Part E - The FPF Constitution and Authoring Guides
Preface node
heading:part-e-the-fpf-constitution-and-authoring-guides:97787
What this page is
This is generated FPF reference text from the specification preface or supporting sections. It helps interpret FPF; it is not FPF Reference product documentation.
Methodology
Use it to understand how the specification wants to be read, then return to a route, pattern, or work packet for active work. Cite generated IDs only when the wording changes the task decision.
Content
Vision & Mission
Status: Stable
Problem frame
Use this charter when proposing an FPF rule or artifact and deciding whether it serves the framework's purpose. Name the reader's task, show which core invariant the rule supports, and state a measurable stakeholder benefit or justify its omission. Use §4's short route and §5.2's worked decision. When the claim also proposes a material realization or physical intervention, apply the separate feasibility branch in §5.1.
FPF aims to give engineers, researchers and learners a reusable scaffold for carrying ideas through comparison, decision and implementation. A shared vocabulary is useful only where it helps a named use; the charter does not by itself demonstrate improvement over a domain's existing practice.
Problem
When a claim crosses domains without its meaning, basis or conditions, the receiver can repeat work or rely on an inapplicable result. A proposed common rule can repair that problem, but can also add learning and maintenance cost. Decide from the affected use rather than assuming that a shared framework is always superior.
Forces
Mission Statement
Enable motivated people and AI agents to transform a raw idea into a reproducible, auditable change in the physical world through incremental, falsifiable cycles.
Vision Statement
Make reliable reasoning accessible across disciplines: combine the relevant shared grounds and reusable methods of FPF Core with suitable domain methods, and keep decisions traceable to their premises, evidence and revisions.
Solution — Shared Grounds and Reusable Methods
FPF helps practitioners construct and revise ways of doing from explicit shared grounds. Its first principles are cross-domain starting commitments that govern the framework's architecture, use, evaluation and development, including the distinctions, inferences and grounds for claims used in reasoning. The constitutional and defining patterns explain these commitments and their scope. A premise introduced in one task model has the scope justified for that model; using it does not establish a framework-wide commitment. All eleven pillars in E.2 retain their role as FPF's constitutional first principles. For example, P-2 gives human comprehension priority over theoretical or tooling purity, and P-7 directs proofs, metrics and models towards real-world objectives.
The Kernel is the subset of Core content that defines admitted universal meta-concepts and their relations. The eleven pillars are constitutional requirements outside that subset; they govern the Kernel and other Core content. P-2 and P-7 therefore remain first principles without becoming Kernel definitions. Broader Core content also explains reusable methods. Describing a method does not establish that someone can perform it or has performed it. Practitioners use domain methods to perform the subject work; domain evidence supports bounded claims about the suitability of a method or its result for the intended use. These grounds and methods remain open to justified revision under the invariants below and E.2.
FPF expresses these shared grounds and reusable contributions through:
-
a Kernel of admitted universal meta-concepts and the rules that define their meanings and relations;
-
patterns that explain reusable methods, state their conditions and grounds, and apply or extend shared distinctions without replacing their governing meanings;
-
a pattern language (Architectural ► why/ how; Definitional ► what) with embedded Conformance Checklist (CC);
-
Design Rationale Records (DRRs) that preserve the grounds for changes to FPF and make them available for audit;
-
three core invariants that every artefact must honour
- Evolvability — change is expected and governed;
- Cross‑Scale Coherence — preserve shared distinctions across scales and qualify each part–whole algebra for its subject and the structure it preserves;
- Didactic Transparency — each element exposes its reasoning path.
For a proposed rule or artifact, take one affected use through these steps:
- State the intended user, decision or action, and the failure the proposal addresses.
- For a normative rule, name at least one core invariant and explain the rule's contribution in that use. A copied invariant label is insufficient; if the contribution is unresolved, retain the proposal and return the missing argument or case before treating it as charter-conformant.
- State the artifact's intended measurable benefit, its comparison basis and how it could be checked, or the reason a benefit statement is inapplicable. Distinguish an aim or supported prediction from an observed result. Withdraw an unsupported outcome claim or obtain the evidence needed by that claim; do not substitute conformance for benefit.
- Apply only the additional feasibility and lexical conditions that the claim activates. State any remaining gap and the next useful action. A changed benefit basis or invariant argument reopens the affected claim, not unrelated conclusions.
Conformance Checklist
Mathematical constructions retain their validity and composition conditions. Material feasibility becomes an additional question when a claim proposes realizing a construction materially or intervening in a physical system. A missing manufacturing method, resource or constraint basis leaves that proposed action unresolved; a physical expression of a mathematical object establishes no realization of its elements.
An in-principle feasibility claim does not assert that production occurred. Work, change, entity inception and production completion require their separate bases under A.12, A.15.1 and A.15.PROD when those claims are made.
Three questions about construction
- Mathematical product. A construction of R² × R³ needs its defined elements, product operations and mathematical laws. With no proposed material realization, it needs no manufacturing actor or resource budget. Printing the construction on paper produces a carrier, not physical copies of all its elements.
- Proposed pump assembly. An engineer proposes assembling a named shaft and housing with a required clearance of 0.05–0.10 mm. The feasibility claim cites the manufacturing/assembly method, available equipment and materials, and the tolerance calculation supporting the stated clearance under those conditions. If the equipment capability or tolerances are unknown, return that missing basis before claiming feasibility for the decision. A valid drawing alone is insufficient.
- Manufactured pump. Claiming that pump P17 was produced additionally needs the assembly Work and change, and the independently applicable inception/completion facts. In an illustrative trace, a measured 0.07 mm clearance can support that one completion criterion; it supplies neither the whole production account nor an untested performance claim. A.12, A.15.1 and A.15.PROD govern those occurrence claims, while any actual composition retains its composition rules.
Decide a proposed reuse rule and its benefit
An engineer proposes instruction R1: “Before reusing a result, compare its subject/edition, question/use, relied sources/dependencies and qualification/currentness with this decision. Reuse only if they still answer the concern; reopen the affected question otherwise.” The artifact is a short instruction with that comparison and a filled case.
R1 supports Didactic Transparency by making the reason for reuse or reopening inspectable, and Evolvability by identifying which changed premise reopens the result. Its intended measurable benefit for the engineer is fewer invalid reuse decisions than the simpler rule “reuse the result with the same label.” The scope of the following evidence is only this stipulated desk comparison:
There is one invalid reuse under the label-only rule and none under R1 in these two cases. This supports the stated failure-prevention argument; it is not an observation of reader behavior or proof that review is faster. The added comparison has a cost. If a receiver claims “R1 makes all reviews 20% faster” without observations, remove that outcome claim and retain the bounded aim, or select a task-matched inquiry before making it. If the proposed rule merely requires a new label and says “supports Evolvability,” return the missing continuity or change-handling contribution; the label alone does not satisfy CC-Vision.2.
R1 can guide the stipulated reuse decision. Admission to FPF requires the charter argument; a team adopting R1 uses its own decision authority. A change of population reopens applicability when the grounds relied on do not already cover the target. A claim about actual reading cost or invalid reuse decisions can be investigated through E.12. A proposal whose contribution to a core invariant is still unclear needs that argument before admission.
Consequences
Intended gains — Make cross-domain claims easier to recover, make their support traceable, and help learners use the concepts. An artifact states and supports its benefit; this charter is not evidence that those gains have already occurred. Trade‑offs — Authors face an initial learning curve and must trace every rule to an invariant; traceability is required to prevent variant sprawl.
SoTA-Echoing — connect conformance to intended use
Practice question. Does a well-formed, traceable rule deserve adoption when its contribution and stakeholder benefit remain unclear? The selected line couples rule conformance with a bounded intended-use argument. The simpler default checks required fields and invariant labels, accepting the artifact when that trace is complete. That default is inexpensive and can suffice when the substantive argument is already established and unchanged; it cannot establish the argument for a new rule by itself.
The NASA Systems Engineering Handbook's product verification and validation distinction, §§5.3–5.4 supplies the substantive line: requirements conformance and satisfaction of stakeholder expectations in intended use answer different questions. Adapt that distinction to CC-Vision.2/4 and §4 steps 2–3. The handbook concerns engineered products; its authority neither admits an FPF invariant nor proves the universal value of a reasoning framework. The conformance-only shortcut is the compared local default, not NASA's complete method.
The GOV.UK Service Standard's success-definition guidance supports the complementary move of identifying useful performance measures together with user research. Adapt it to the benefit and evidence boundary in §5.2. Reject treating a metric target as an achieved effect. Neither source supplies a required improvement percentage or a universal framework-effectiveness test.
At the same proposal and two-case scope, §5.2's substantive replay detects the unsafe label-only reuse that a complete traceability field would leave unanswered. It costs a small comparison; wider reader or performance claims cost separately justified inquiry. Reuse an already adequate argument when its conditions match, and justify an omitted benefit where CC-Vision.4 permits it. Reopen only when the intended task, invariant contribution, benefit evidence or cheaper alternative changes enough to affect adoption.
Relations & Precedence
E.1 supplies the charter commitments for E.2 Eleven Pillars and E.5 Guard-Rails. If a proposed FPF rule contradicts an applicable charter commitment, identify the conflicting provisions. The proposal MUST be revised through E.9 before admission under the current charter. For a team using the method, its own governing terms determine who may adopt a local rule.
E.1:End
FPF's Eleven Pillars and Bitter-Lesson Preference (BLP)
Status: Stable
Problem frame
Use the Eleven Pillars in §4 to assess whether an FPF rule or artifact meets its constitutional obligations. Use the separate BLP policy in §6 when a computational choice claims a scale/generalization advantage or invokes a declared local generality policy: start with C.19.1's cheap scale-claim probe. An ordinary bounded fixed procedure can stop with its task-specific justification; specialization alone creates no scale audit, heuristic-debt entry or waiver. Independently applicable assurance or oversight requirements retain their checks.
E.1 sets the aim of making reliable reasoning accessible across disciplines through shared grounds, reusable Core methods and suitable domain methods. To turn that aim into a durable architecture, FPF needs a small, explicit constitution - principles that remain stable while everything built on top of them can evolve. Without such invariants, domain silos, vocabulary drift, and tool-centric shortcuts quickly erode coherence and reproducibility across disciplines.
The pillars are also the first-principles basis of FPF. They are the minimal commitments from which pattern-level work derives: decisive structure, teachability, maturing formality, open kernel, layering, register discipline, practical payoff, cross-scale consistency, explicit state, open-ended evolution, and SoTA renewal. Later patterns can support this basis by making a concrete argument about pillar support more inspectable; they do not replace pillar authority.
Problem
Without binding first principles, contributors cannot distinguish commitments that proposals must preserve from choices they may revise. Treating every choice as fixed blocks adaptation; treating every commitment as optional leaves incompatible rules without a basis for resolution.
Forces
Solution
FPF rests on eleven binding pillars. Follow the proposed change through the work of the people who would use it. Which decisions, explanations or obligations would change, and under which conditions? Use those consequences to identify the applicable pillar constraints below. Where several uses are claimed, establish which grounds extend to them together and where their conditions differ. E.3 resolves conflicting provisions. A DRR for a normative change records its impact on the pillars. E.2.DA serves the broader commission to evaluate how FPF realizes them.
A pillar-impact argument can use ordinary accepted mathematical theory. Use C.29 when a separate question remains about choosing a mathematical representation, transferring a result, or relying on the represented structure. State the correspondence, preserved and lost structure, and limits needed for that use; E.2 continues to govern the pillar claim.
A proposed FPF rule that conflicts with a pillar can be revised to respect it. To proceed with the conflicting proposal, the body empowered to change the constitution must enact the pillar amendment first. The DRR states the grounds and Pillar Impact Analysis.
Conformance Checklist
Policy — Bitter‑Lesson Preference (BLP)
Intent and activation. In search, learning, planning and related computational work, investigate a claimed advantage from increased compute, data or search capacity under C.19.1. The empirical Bitter Lesson motivates that comparison; it does not decide every choice between a specialized and a general approach. An analogy or generality preference for another bearer needs a separately declared local policy, receiving purpose, scale predicate and evidence basis. Ordinary bounded specialization can stop with its task-specific justification.
BLP-1 — Claim-matched comparison and audit. Begin an activated BLP use with C.19.1's cheap scale-claim probe. Name the candidate, task/use, scale variable, objective, usable budget range, receiving conditions and evidence form. Missing comparison premises return no scale claim yet; select evidence depth by claim and risk. An independently activated profile can require a fuller Scale-Audit. That audit retains:
- (a) Parity and admissibility: comparable tasks, budget basis, editions, freshness and risk/safety envelope. Use G.5/G.9 when their selector/parity contracts apply. Pareto comparison is the default over the declared objective vector; another comparison needs a lawful declared policy under E.3.
- (b) Usable budgets: vary the dimensions relevant to the claim—compute, data or freedom of action where applicable—and pin held conditions. Compare only over the range available to the receiving use; extrapolation needs its own basis.
- (c) Response and uncertainty: report performance over that range with an uncertainty treatment appropriate to the evidence. Slopes may characterize response but cannot replace its level, admissibility or receiving floor. Repeatable stochastic trials retain seeds/replicates and confidence intervals where used; deterministic derivations retain their assumptions and validity bounds.
- (d) Resources and assurance: retain decision-material resource accounts through A.15.1, A.15.2, B.1.6, C.16 and A.10 as applicable. A named assurance claim uses B.3; independently applicable assurance and oversight requirements remain binding.
- (e) Objectives and tolerances: declare quality, risk, cost and any illumination coordinate promoted by an explicit CAL policy. State units and the budget/assurance tolerances α/δ used by the comparison. Publish the audit's edition and policy pins through G.11.
BLP-2 — Usable-performance preference. Among admissible options under the receiving conditions, with comparable assurance within δ and budgets within α when those tolerances apply, prefer the option whose usable performance is shown to Pareto-dominate across the audited range. Evaluate uncertainty in the performance contrast, including dependence between paired observations. Return no scale-based preference when the comparison establishes equivalence, establishes a trade-off, or leaves the ordering open. Say which outcome the evidence supports. A project policy under E.3 can decide the cases covered by its terms. Cite that policy and the cost of applying it for the resulting choice.
BLP-3 — Task-appropriate prescription. Positive procedures and prohibitions are both available under their task or control needs. A mandated safety procedure keeps its force. Minimal prescription and autonomous sequencing need their justified policy, authority and guards; neither follows from a scale comparison. Ordinary use of a positive procedure requires no BLP waiver solely because a general search alternative exists.
BLP-4 — Conditional heuristic debt. Apply C.19.1's debt branch when an admitted heuristic carries a durable scale/generality preference, reusable solution-family policy, selector-facing preference or override within an activated BLP use. Retain scope, review responsibility, expiry/review window, replacement target and de-hardening plan in CalibrationLedger/BCT and SCR as applicable. A bounded task-specific tactic that makes none of those claims is not debt merely because it is specialized.
BLP-5 — Separately authorized adaptation. Enable, require or prohibit adaptation under the product's objective, evidence, permission, change authority and risk/Guard-Rail conditions. A fixed method needs no BLP adaptation waiver solely for remaining fixed. A product policy that requires adaptation or its review retains that obligation and its exception rule.
BLP‑6 — Precedence & Safeguards. BLP combines governance and architectural principles; classify each compared principle under E.3 rather than assigning one class to the whole policy. It is instantiated by Pillars P‑10 (Open‑Ended Evolution), P‑11 (SoTA Alignment), P‑7 (Pragmatic Utility), and P‑1 (Cognitive Elegance). It does not override applicable safety and ethics requirements, E.5 Guard‑Rails, or E.3 precedence rulings; where BLP conflicts with Guard‑Rails, Guard‑Rails prevail. When NQD/E/E‑LOG elevates illumination to dominance for exploration mandates, BLP adopts that lens rather than overriding it.
BLP applies to comparisons such as prompt programming versus fine-tuning and model-based planning versus hand-tuned control when they make the scale or generality claims specified above. Their tasks and receiving conditions determine the comparison; G.5/G.9 apply when their selector/parity contracts hold.
Conformance Checklist — BLP
Usable response and ordinary non-use
These constructed cases distinguish the rule; they are not empirical findings about real methods.
- Bounded arithmetic: summing one known finite list with a fixed procedure makes no scale or generality claim. Use the correct arithmetic and task budget; no scale audit, heuristic-debt entry or adaptation waiver follows from the availability of a general learner.
- Positive safety procedure: a product's applicable isolation-and-verification procedure remains required for the named maintenance action. Its task/control basis warrants its positive steps. BLP neither replaces it with prohibitions nor demands a waiver merely for its form.
- Response versus slope: for budget b in [1,10], A(b)=10+2b ranges from 12 to 30, while B(b)=90+0.1b ranges from 90.1 to 91. With the receiving quality floor 80, A fails everywhere despite its larger slope. B satisfies that floor; any broader preference still respects the other declared objectives, uncertainty and admissibility conditions. The floor is this use's condition, not a BLP constant.
- Non-dominance: at the same declared task, one admissible option has quality 95 and cost 10; another has quality 85 and cost 2. With quality maximized and cost minimized, neither dominates. BLP returns no scale-based preference. A declared generality policy may choose only within the options left lawful by higher rules and must identify that policy basis.
Relations
The relations in this section govern the Bitter-Lesson Preference.
- Instantiates pillars: P‑10, P‑11, P‑7, P‑1.
- Depends on: G.5/G.9 (admission/comparator/selector and parity harness), G.11 (refresh telemetry), A.15.1 (dated Work), A.15.2 (planned work and budgets), B.1.6 (resource aggregation), C.16 (resource and cost measurement), and A.10 (provenance), C.18 (NQD-CAL), C.19 (E/E-LOG), and F.7/F.9 (Bridges, CL/Φ/Ψ).
- Constrained by: E.5 Guard‑Rails (DevOps Lexical Firewall; Notational Independence; Unidirectional Dependency; Cross‑Disciplinary Bias Audit) and E.3 precedence.
Definitions
α (budget tolerance) may be relative or absolute; declare units (e.g., % cost, wall‑time, energy). δ (assurance tolerance) is the permissible delta in assurance under B.3; declare measure and floor(s).
Consequences
Positive
- Gives contributors explicit constitutional obligations.
- Delivers a falsifiable checklist for evaluating proposals.
Trade‑offs
- Constitutional review adds friction to rapid, informal changes.
- Amending the pillar set demands high‑bar governance.
Rationale
The pillars interlock: Cognitive Elegance (P‑1) enables Didactic Primacy (P‑2); Open‑Ended Kernel (P‑4) and FPF Layering (P‑5) make Open‑Ended Evolution (P‑10) and SoTA alignment (P‑11) feasible; Cross‑Scale Consistency (P‑8) provides the algebraic backbone for Scalable Formality (P‑3). Together the pillars balance stability with change, rigor with accessibility, and abstraction with measurable impact.
C.29 supports a pillar-impact argument when its mathematical representation choice, transfer, or reliance remains unresolved. It states the relevant correspondence, loss and use condition; E.2 remains authority over the pillars.
Relations
- Depends on:
E.1– pillars operationalise the mission. - Applied by: Later Core patterns use the pillars relevant to their decisions. When a pattern also claims strict model refinement, compare the two models for the same inquiry. The proposed model must admit a feasible completion, preserve the earlier commitments in every completion and exclude a formerly admissible answer. B.5.MR:4.2 gives the test when the models use different languages.
- Mathematical support path:
C.29supports pillar-impact arguments only for adequacy of mathematical lenses used to express first-principles structure. It does not amend pillar content, priority, or conformance. - Governs: FPF proposals and artifacts where a pillar sets a condition of conformance. A team using an FPF publication can identify which provisions it has adopted for its own work.
E.2:End
Evaluate How FPF Realizes Its Pillars (Pillar-Adequacy CharacteristicSpace)
Status: Stable.
Problem frame
Use E.2.DA when the object under improvement is an FPF-level object and the question is whether it realizes the E.2 Pillars adequately for a declared use. The object may be an FPF edition or release candidate, a pattern family or selected host set, a set of projections, a public-entry change (README, Architectural Reference or ToC), or an access route such as a skill or MCP-backed service.
Use it after a broad cleanup, new pattern family, projection repair, source-use repair, FPF-form change under E.4.FPF, or corpus-level improvement when local pattern quality is not enough. A set of good local patterns can harm FPF entry, naming, layering, source use, projection integrity, or open-ended evolution.
Use the local method in place of E.2.DA for one authored pattern version (E.21), one DRR (E.9.DA), one local wording repair (F.19), or one pattern-use entry problem (E.11). Open E.10 or a precision-restoration neighbor for an unresolved FPF-specific wording question.
First useful move: name the FPF object under improvement by value, declared use, reader family, and qualification window; then evaluate all eleven Pillar coordinates. If a Pillar seems unaffected, give it a value and a short rationale saying what is preserved.
What goes wrong if missed: FPF can become locally polished but globally worse. Readers may find fewer useful entry points, precision repairs may erase the working move, source rows can turn decorative, or several patterns can grow local variants of the same doctrine.
Primary EntityOfConcern: the scoped FPF object under improvement as an E.2 Pillar-realizing language object.
Problem
E.2 gives the Pillars and their constitutional meaning. It does not by itself evaluate whether the scoped FPF object realizes those Pillars well enough for its declared use. Without E.2.DA, corpus-level reviews tend to collapse into local pattern scores, process state, review praise, or broad claims that "FPF improved."
The failures are:
- Local pattern quality is averaged into FPF adequacy.
- Entry projections and companion files become separately maintained sources of definitions, constraints, or tests supplied by pattern bodies.
- Precision repair improves terminology but damages first-use comprehension or changes the FPF kind carried by the repaired text.
- Source and SoTA rows are counted rather than checked for changes in FPF moves.
- Evaluation and selection expressions such as
all 5s,Paretoandshortlistare used as praise instead of claims with their own criteria. - Corpus-level stop claims hide what became worse.
- Pillar values become repair targets, so the corpus gains projection proof, entry apparatus, source rows, or review evidence while the FPF object becomes less usable, less layered, or less decisive.
Forces
Solution
E.2.DA is the Pillar-adequacy specialization of A.19.ECS. An evaluator applies its questions to one FPF object under improvement against all eleven E.2 Pillars for a declared use.
An E.2.DA result gives every Pillar coordinate a value, short rationale, evidence locus, and shared evidence basis for the FPF object being evaluated. Local pattern quality, DRR adequacy, and wording repair use their object-under-improvement questions. A bounded diagnostic may borrow selected Pillar questions, but makes neither an E.2.DA-result nor an FPF-adequacy status claim.
Local names and kind settlement
These names are local to the evaluation unless F.18 promotes a durable name. They name FPF content objects and evaluation fields, not release state, review state, or project evidence.
Evaluation record
E.22 may frame the evaluation purpose when the caller needs floor evaluation, exceptional improvement, trade-off inspection, open-question discovery, absorption, or proposal portfolios. E.23 governs repeated improvement after the evaluation returns findings or candidate proposals.
Ordinal coordinate scale
The values are ordinal content evaluations. They are not an aggregate scalar score, maturity ladder, release gate, or proof that development ends.
Required Pillar coordinates
Evidence and coordinate separation
Follow the practitioner’s work from its ordinary entry to the decision or explanation that uses the result. Establish what each contribution supplies and whether the receiving work can use it directly or with a warranted transfer to its own conditions. One argument can cover the declared range. If a permitted difference could change the conclusion, examine it or bound its effect. A single source can support several coordinates when the grounds for each value are clear. The following distinctions carry most repairs:
If a distinction cannot be recovered from the FPF object, lower the affected coordinate and state the first repair. Do not add a new local doctrine table to explain around the missing content.
E.21 and E.9.DA results are evidence loci for E.2.DA, not inputs to be averaged. A pattern-quality value can support a Pillar only by pointing to the evaluated pattern's beneficial or harmful FPF-level effect.
Result-row discipline and calibration
An E.2.DA result uses this table shape:
For values 1..4, explain why the lower adjacent value would understate the evidence and the higher adjacent value would overstate it. For 0, explain why 1 would overstate the evidence and what would raise the value or reopen it. For 5, explain why 4 would understate the evidence and what would lower the value or reopen it.
A Pillar essay, local-quality average, two-column table, or result whose value depends on unchecked corpus, projection or source evidence is draft evaluation material. Check the needed evidence before assigning the affected value. An observed defect lowers a value according to its coordinate meaning; absence of a check does not by itself establish that defect. A completed E.2.DA result evaluates all eleven coordinates.
Common calibration points:
Status and stop condition
The stop condition states the declared floor, values, smallest reopen locus, and first repair when the declared use is not yet admissible. Add a non-use boundary only when an independently grounded reading by a plausible intended reader changes a named use.
Compact result form
For a small release decision, the coordinate table may be compact if it retains every required row and field. Status is not assigned from prose, a checklist count, a local-pattern average, a two-column table, or a result missing evidence loci needed by its values.
When E.22, E.23, absorption, or exceptional-improvement framing asks for improvement, below-floor Pillar coordinates return findings or repair. Above-floor coordinates receive proposal rows only for substantive non-dominated FPF-level content opportunities inside the declared use: for example, better entry recognition, placement of defining, constraining, or testing content, source-currentness carry-through, projection thinning, corpus-ecology repair, kind-preserving precision restoration, open-ended evolution support, or deletion or relocation of apparatus that weakens the FPF object. Apparatus-only additions do not qualify. Do not treat every value below 5 as a defect. A 4 may be the correct stop value only with loci showing why further Pillar-content movement is dominated, unavailable, or outside scope.
Worked slices
Broad precision cleanup. A wording pass makes many patterns more admissible but several Problem frames now explain less about why the distinction matters, or a cleaned phrase changes the governed kind while the trigger word disappears. Reassess P2, P6, and P7 against the work each measures. A reader may still make a warranted decision even when the words no longer identify which kind of thing the decision concerns. Lower each value for the loss its evidence establishes.
Ontic architecture repair. A campaign adds E.24.CD, E.24.PUB, or a new ontic host. Follow what the change enables in each evaluated use. A new distinction can let readers settle a previously unresolved question even when no competing definition existed. Removing duplicate definitions can ease maintenance while leaving an already well-supported practitioner decision intact. Assess each Pillar on the effect that its coordinate measures. For extra terminology or publication apparatus, follow the resulting change to the affected uses before assigning values to P1, P2, P5, P6, and P7.
Cleanup that removes useful content. A corpus pass removes repeated guards, instructions, references or architecture-placement prose, but leaves several patterns with less positive ontology, method, norm or worked action than before. Trace the missing contribution to the recognition, use or later revision that depended on it. Reassess the affected coordinates on those grounds.
Projection repair. README scenarios, ToC rows, E.11 entry-distribution loci and worked comparisons improve search but can become separately maintained sources of pattern rules. A second source of authority can lower P5. Reassess P9 where it also makes the governing state, edition, change or effective window harder to establish. The repair places each authored definition, constraint, or test in the pattern body that supplies it and preserves the public aid's E.11 function. A locator remains a locator; an ordinary entry or Practical-Use Card retains its source-linked first-use guidance, including the first useful result or blocker and stop or wrong-turn return.
Source absorption. A new source can support a practice already in use, qualify its limits or change the Method. Establish which claim or choice its findings bear on. A citation on its own leaves that contribution open. Reassess P7 when the source changes what a practitioner can understand, judge or do.
Bias annotation
This pattern biases FPF toward whole-language adequacy. The bias is useful because local repairs often hide corpus-level loss.
The bias is bounded by the object-under-improvement declaration. E.2.DA does not replace E.21, E.9.DA, E.10, E.11, or E.23; it evaluates their FPF-level Pillar effect when the scoped FPF object includes their results.
Conformance checklist
Common anti-patterns and repairs
Relations
Rationale
FPF needs a corpus-level quality instrument because the language can degrade while individual pattern edits look successful. The complete eleven-coordinate evaluation examines accumulated effects on Pillar realization even when each constituent change is a bounded local repair.
The result uses short rationales and evidence loci named by value. Obtaining their grounds can require substantial work; reuse adequate evidence and select further inquiry for the question it can resolve. The result requires neither a new review process nor an exhaustive audit or source dossier.
SoTA-Echoing
Consequences
E.2.DA:End
Principle Taxonomy, Precedence and Agent Autonomy Profiles (ABL)
Status: Stable
Problem frame
When several provisions govern a decision, start with the decision and the applicable provisions in §4.2. Establish whether you can meet them together, which rule takes priority if they conflict, and who can settle an outstanding case or change the standing rule. Use §6 to declare and apply an autonomy oversight profile for agentic tool use, binding it to the action, budget and permission window. E.2 supplies the eleven constitutional pillars; their names alone do not settle a conflict.
Problem
When applicable pillars or derived principles prescribe incompatible actions, participants need an agreed basis for identifying which governs or which action must wait for an authorized resolution.
Forces
Solution
Principle Taxonomy
For this precedence classification, assign each principle exactly one class from { Gov, Arch, Epist, Prag, Did }.
Epistemological sub‑concerns (reasoning, falsifiability) reside inside Epist, keeping semantics and trust in one class.
E.3:4.2 - Precedence Stack
Identify who would do what, for whom or to what, and when. Which provisions govern that decision? Can a course of action meet them together? A rule can require an action, forbid it, reserve a choice to a participant, or let another decision maker choose. A team can select a common format if the governing rule delegates that choice to the team. If the rule reserves the choice to each sender, a team instruction that removes the sender’s choice conflicts with it. Apply compatible provisions together. If applicability turns on a missing fact, compare what each possible rule would allow or require. When all support the same action on the same terms, it can proceed. Obtain the fact if it changes the decision or the required procedure, or if the task asks which rule applies.
The FPF precedence rank orders applicable provisions in the FPF constitution. A smaller number gives the provision priority where the two conflict. Rank 1 belongs to the eleven pillars themselves. The derived BLP policy also printed in E.2 has its own terms in §6. A passage that quotes a binding rule takes its force from that rule. When an organization adopts guidance for its own work, the terms of adoption establish where it binds.
Apply the priority that governs the conflict. Within higher-ranking constraints, first use an explicit priority for the provisions in conflict. For a remaining conflict between classified principles, use the default class order:
Gov ≫ Arch ≫ Epist ≫ Prag ≫ Did
Apply the priority to the disputed choice and retain the other obligations. Several provisions can leave an obligation with no permitted way to satisfy it even when every pair is compatible. To establish that conflict, show how the provisions close every permitted alternative. If the alternatives are still unknown, identify what would establish them. After resolving a conflict, check the proposed course against the provisions together. Separately establish whether the people and resources needed for that course are available.
When a conflict remains, identify the provisions responsible, the affected action and the conditions that make them apply. Hold the dependent action. Independent permitted work can proceed. An existing ruling may already settle this case. If a holder is empowered to resolve this conflict, apply their actual decision within its scope and effective period. Until they decide, a known decision maker gives the parties someone to ask.
When the case would need an exception, use only an exception the governing terms authorize for these circumstances. If the mandate does not cover the conflict, keep the affected action on hold. To change a rule or priority for future cases, obtain the decision of the body or bodies empowered to change it. Record the rationale, affected scope, Pillar Impact Analysis and edge effects of an FPF change under E.9. For an exception governed by the FPF policy in §6, its DRR, rationale and expiry are required. An organization’s own policy sets the terms for exceptions within its authority. An earlier ruling governs a later case only when the governing terms give it that reach.
Graph Rule — The precedence graph MUST be acyclic; any new edge that would form a cycle is rejected.
A release-schedule rule classified as Gov takes priority over a conflicting performance-tuning preference classified as Prag when the ranks and applicable explicit priorities leave the choice to the class order.
Conformance Checklist
Illustrative Conflict Resolution
Same-kind wording. First recover the intended kind under A.1. For an identified U.System, compare the plain name System with its technical expression U.System; both name the same kind. P-1 supports a precise, economical expression, while P-2 expressly gives human comprehension priority over theoretical or tooling purity. Use the familiar wording where it preserves the required meaning and make the technical designation recoverable where needed. The default Arch-over-Did order does not reverse that explicit P-2 priority. U.System is a proper subtype of U.Holon: replacing a non-System Holon with System would change the subject’s extension before any legitimate wording comparison.
Two Gov rules with no priority. G1 requires publishing report R to audience A at noon. G2 forbids the same publication. Both are Gov and have the same FPF precedence rank. Hold the report. Suppose the charter gives a duty officer power to decide individual report disputes. Naming the officer leaves today's decision open. If the officer rules that R must be withheld from A today, apply that ruling. Tomorrow's report still falls under the standing rules. A proposed change to their priority needs a decision from the authority empowered to change it.
Changed case. Suppose an authorized G2-over-G1 priority is in force for R, A and the noon window. G2 governs, so publication is forbidden; this pair needs no further priority resolution. A proposed G1-over-G2 edge would create a cycle and is rejected. The existing priority and G2's prohibition remain in force.
Three provisions, two notice channels. A service must send a notice today through either channel A or channel B. One provision closes A and another closes B. Both channels are working, yet there is no permitted way to send the required notice. Any two of these provisions could be met together. Report all three and the outstanding duty. If an authorized, effective exception opens A for this notice, send it through A. The restriction on B and the terms for tomorrow’s notices still apply.
BLP and a guard. If an applicable E.5 guard excludes a proposed implementation dependency, BLP-6 gives that guard priority over BLP. Compare those rules in their current scope, not the E.2 and E.5 files as whole nodes. B.3 enters the policy ordering for an assurance requirement consumed by the use.
Local product policy and a pillar. A product preference cannot override an applicable pillar merely because a local policy says it wins. A disclosed local tie-break can settle only a preference left open by the applicable higher rules. Classify the compared principles; a compound policy such as BLP need not receive one class for all its provisions.
BLP-policy precedence (high → low, for applicable rules). Law & Regulation → E.5 Guard-Rails → B.3 assurance requirements consumed by the use → E.3 governance decisions → E/E-LOG policies (editioned) → BLP derived policy (E.2 §6) → Product Policies → Implementation Tactics. This scope-specific ordering implements the stated BLP safeguards; it does not place a derived policy above a pillar or create unconditional file-to-file edges.
Notes.
- BLP is a constitutional policy (E.2) governing an activated scale comparison. E.5 Guard‑Rails and applicable B.3 assurance floors retain their force. A project can apply a separate generality preference only under the conditions its policy specifies.
- Wherever NQD/E/E‑LOG promotes illumination telemetry to dominance (via an explicit CAL policy; policy‑id recorded in SCR), BLP adopts that lens rather than overriding it (see E.2 BLP‑6).
- Any exception to policy MUST include a DRR with rationale and expiry.
- BLP Override (Waiver). A waiver is needed only when an applicable declared generality policy would otherwise decide the use and its exception rule is invoked. Name that policy, the rationale, responsible review System and direct responsibility basis, expiry/review, and the evidence required by the exception. Apply C.19.1's waiver and debt conditions. Ordinary bounded specialization is not an override. Add an AutonomyProfileId and GateDecision authority only when an independently applicable agentic profile governs the action.
Set-returning parity. Activated scale comparisons preserve parity and admissible operations under A.18; use G.5/G.9 where their selector/parity contracts apply. Independently applicable stronger profiles retain those requirements. Mixed-scale scalarization without an admissible operation remains prohibited.
BLP — Bitter-Lesson Hooks into Precedence
- Comparison and tie-breaking. Under E.2 BLP-2, compare usable performance over the declared range and receiving conditions, with uncertainty in the contrast and applicable α/δ tolerances. When the comparison yields no scale-based preference, identify whether it establishes equivalence, establishes a trade-off or leaves the order open. A declared generality policy can settle the case if its terms cover that outcome and higher rules permit the choice.
- Procedure and search. Select a positive procedure or general search under the task/control requirements and applicable policy. A mandated safety procedure keeps its force. BLP adds no universal prohibition-first rule or waiver for a fixed procedure.
- Publication. A BLP precedence result retains the policy/edition and comparison basis. A performed full audit publishes its material resource and assurance accounts under E.2 BLP-1d and G.11. A bounded use with no scale claim creates no audit package.
ABL — Autonomy‑Budget & Oversight Profiles (GateProfile) This section defines an extensible family of autonomy oversight profiles for agentic tool use: each profile specifies (i) a budget envelope, (ii) a Freedom‑of‑Action (FoA) descriptor, and (iii) the required publication of a gate decision authorizing execution under that envelope. The labels L0…L4 are treated here as profile identifiers (not a fixed managerial ladder): projects MAY introduce additional profiles or sub‑profiles by minting new profile ids, provided they publish the same fields (budgets, FoA, decision roles, telemetry requirements) and keep profile changes explicit and auditable. For a further profile, declare which standard obligations it inherits and which additional safeguards govern it. Here “L1+” and “L2+” apply to the named standard profiles at or above those identifiers and to any additional profile that inherits the specified obligation. In documents that also use model abstraction levels, write autonomy profile L2 for the oversight rule and model abstraction level L2 for the modelling use in B.5.MR. A model at any abstraction level can be discussed under any applicable oversight profile.
Normative requirements by profile.
- Budgets. Each profile MUST declare ceilings for time / compute / cost / risk and a FoA descriptor; each ceiling's meaning and scale must be explicit, with units where applicable under C.16. Planned ceilings remain A.15.2 WorkPlan content, and run‑time consumption is tied to dated Work, aggregation, and provenance under A.15.1, B.1.6, and A.10. Budgets are hard gates at run‑time (A.21; C.Agent‑Tools‑CAL ATC‑3 for stopping or replanning when a ceiling is breached).
- Profile binding & change visibility. Every CallPlan MUST declare the active profile id. A profile change MUST retain its from/to profiles, policy editions, effective window and A.21 permission result in the DecisionLog. Recompute the decision when its subject, action, rule, checks, scope or window changes. Cite an E.18 GateCrossing only when an independently selected TFS, exact source/receiving positions and changed CtxState bindings establish that structural transition. Add an F.9 Bridge only for its independently obtaining relation between the two F.17 local senses.
- Assurance floors. Apply B.3 to the named assurance claim and receiving use. A formality requirement names the expression properties needed under C.2.3, or a justified threshold on a declared local scale; any R threshold, composition, or CL loss needs its justified quantity, scale, and model under B.3/C.2.2. No universal F/R weakest-link minimum is imposed by an autonomy-profile label, and F does not substitute for R. Any justified profile-specific tightening for broader FoA MUST be declared on the profile and pinned by policy-id; the protection and decision-authority conditions remain binding. Pre-deployment assurance deltas MUST be recorded for L2+.
- Exploration discipline. When a live C.19 pool constrains the
CallPlan, cite itsPoolPolicyResult. Include theEmitterPolicyused by the plan andexplore_shareonly when the plan uses that profile, including0when it explicitly plans no exploration. Apply the declared profile's exception rule to a deviation; do not invent live-pool refs after the action is fixed and pool treatment is irrelevant. - Provenance. L1+ MUST emit a CallGraph with Service/Method editions, any used EmitterPolicyRef, budget deltas, and observation hooks (C.Agent‑Tools‑CAL ATC‑5).
- BLP conformance. Apply E.2 BLP when selection makes a scale claim or invokes a declared generality policy, retaining its α/δ tolerances and conditional debt requirements. L2+ alone does not establish a scale claim. A profile that independently requires a fuller comparison or audit must state that trigger and basis; its other budget, oversight and assurance duties remain binding.
- Learning/Adaptation. L3–L4 MAY enable adaptation only under product permission, change authority, E.5 and privacy controls. L0–L2 retain their profile default off unless their authorized policy change documents mitigation. Remaining fixed needs no BLP waiver; changing an applicable product requirement follows that requirement's exception rule.
- Human‑in‑the‑Loop (HITL). HITL obligations are expressed as gate decisions and pause/resume hooks, not an implicit “approval ladder”:
- L0–L1: execution MAY start only after a GateDecision authorizing the CallPlan is present in the declared window.
- L2: sentinels MUST be able to pause execution; resumption requires a new GateDecision recorded in the DecisionLog.
- L3: the profile MUST declare periodic review windows; continued execution across a review boundary requires a GateDecision.
- L4: continuous telemetry review; the default execution context is sandboxed. Leaving the sandbox requires an A.21 GateDecision for that bounded action, with the applicable signers, budget, scope/window and stop conditions. Publish the decision and profile/policy change. When the case also satisfies E.18’s structural crossing conditions, cite its CrossingRef and required per-binding accounts. Unchanged local meanings need no invented F.9 Bridge.
Gate‑decision authority map (default signers; who may author GateDecisions).
- L0: EoR or appointed maintainer.
- L1: EoR and peer reviewer (two‑person rule).
- L2: Team Lead and Safety representative.
- L3: Product Owner and Safety and Legal/Privacy.
- L4: Gov‑CAL Board (multi‑disciplinary) with documented scope, time‑boxed trial budget, and rollback criteria.
Profile promotion / demotion triggers.
- Promotion. Use the active profile’s required performance and assurance evidence for the actions and conditions it covers, including stability within the declared tolerances and budget for ≥ N_policy runs. Establish what the proposed profile would newly permit, which commitments and controls must hold together and what evidence covers those uses. Promotion requires its A.21 GateDecision. A supporting scale claim uses the usable-response comparison under E.2 BLP-1c. All independently required profile checks apply.
- Restriction and reassessment. Demote the profile when its declared risk or budget ceiling is breached, assurance drops below a required floor, or a significant heuristic‑debt item expires without replacement. A policy change calls for reassessing the affected permissions. Retain, restrict or extend them as the new terms, authority and grounds warrant. Publish a demotion as a profile-change and permission result with the applicable policy, budgets and effective window. E.18 governs a structural crossing when its conditions obtain.
Profile-change cases. A local budget revision from twenty to thirty minutes with unchanged meanings requires the authorized profile/policy change and its permission result; absent a selected TFS transition, it has no GateCrossing. If the profile label remains L2, the original twenty-minute permission cannot authorize execution at minute twenty-five. Apply the authorized new ceiling, required checks and effective window before continuing; otherwise stop at the old ceiling. A case that changes a CtxState binding between two positions in an independently selected TFS also satisfies E.18’s crossing obligations. Leaving a sandbox with unchanged F.17 senses needs the declared authorization and oversight, but supplies no semantic Bridge. Stop/resume, signer and budget duties remain applicable in each case under its profile.
Conformance Checklist — E.3 ↔ BLP Interop
Consequences
Positive — Makes priorities and unresolved conflicts inspectable before the dependent action.
Cost — A conflict can delay the dependent work until a competent holder decides the case. Changing the priority for future uses can take longer and calls for the authority to amend the governing terms.
Rationale
The taxonomy groups principles according to what they govern: change (Gov), framework structure (Arch), meaning, evidence and trust (Epist), practical value (Prag), or comprehension (Did). Explicit override edges and the default hierarchy determine a winner where the declared ordering suffices. Where it does not, the hold leaves the decision with the holder empowered to settle this conflict. That mandate determines whether the ruling covers this case or changes future decisions. The taxonomy supports that decision; it does not make every pair comparable.
Relations
- Depends on:
E.1,E.2 - Governs: Classification and precedence when an FPF pattern or DRR introduces, changes or compares principles. Guard‑Rail patterns reference CC‑PT. Section 6 also governs the oversight profile for agentic tool use.
SoTA-Echoing — resolve rule conflicts and bind oversight to the action
Practice question and choice. How can an agentic action use applicable governance rules and a budget profile without acquiring an undeclared priority or extra authority? The selected answer combines §4.2's partial order and unresolved-action hold with §6's editioned profile, concrete ceilings, checks and permission window. The simpler default uses one fixed conflict-combining rule and a stable table of profiles. That default is cheaper where every relevant rule fits its declared semantics and the table fully specifies the action; the added governance work is justified when obligations conflict or a profile's conditions change.
Under Cedar's authorization algorithm, an applicable forbid overrides permits, otherwise an applicable permit allows the request, and the default is denial. Adopt explicit request scope and a declared combination rule where that permission policy is intended. Reject treating it as a universal ordering of obligations: a requirement to publish and a prohibition of publication can remain normatively inconsistent even when access is denied. Cedar does not claim to repair that inconsistency. Its policy-error handling also does not replace A.21's treatment of required check results. The source defines an authorization language, not FPF rule-change authority.
Engineering configuration management gives a practical test for the reach of an exception. After an authorized waiver for one item, a team inspecting the next item still works from the effective baseline. Apply that distinction to case rulings and standing amendments. The mandate governing the actual decision determines who can grant either and what the ruling covers. NASA Systems Engineering Handbook, §6.5.1.2.3
For oversight, §6 draws on NIST AI RMF 1.0 §6, pp.33–34. Its profiles depend on the use, requirements, risk tolerance and resources; they can distinguish current and target conditions. Adapt that contextual treatment to §6's budgets, signers, checks and effective window. The source does not prescribe ABL's L0–L4 labels, numeric defaults or FPF permission semantics. A project's defaults need a basis appropriate to its use; an ordinal-looking label alone supplies none. This comparison uses the published 1.0 method.
At the same action and input scope, the two cases expose the chosen trade-off:
The taxonomy groups principles for the declared fallback order; its five classes are not a claim that all governing duties reduce to permission predicates. Use the simpler default when its scope and maintained inputs suffice. Reopen only the affected priority or profile comparison when a new incompatible rule, changed permission condition, or equally adequate cheaper policy changes the action. BLP comparisons remain conditional on a scale claim or declared generality policy.
E.3:End
FPF Ecosystem Architecture: Framework Families, Products and DPF Suites
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative.
Problem frame
This pattern helps FPF users, framework authors and stewards answer any of these questions as they create, extend or use an FPF-grounded pattern ecosystem. What belongs to FPF itself, to the FPF Core or to a domain or local framework? Which records carry relation and edition claims? Which patterns define publication, access, naming, source, currentness and quality work?
Primary EntityOfConcern: the FPF-grounded pattern ecosystem for one named ecosystem question. For a first answer, name the question and classify the likely case. Point to the pattern that addresses it, or state which distinction is missing. Open a complete ecosystem-architecture record only when the answer must settle durable architecture or support later reliance.
The distinction determines the next contribution: E.4.FPF handles the form of FPF itself; E.11 and E.17 handle first entry and publication; E.4.DPF handles domain or local framework authorship. Use §4.2 for a proposed Suite and §4.1 for support units and adjacent products.
Problem
FPF has grown from a single core pattern set into an ecosystem of core rules, tools, companions, domain frameworks, local practice frameworks, source packs, decisions, quality records, publication and access-facing presentation carriers, and access routes. If those objects are described only by file names, abbreviations, or reader-facing tables of contents, several different kinds collapse:
- a pattern set is treated as a publication or access carrier;
- a local practice framework is treated as an FPF Core amendment;
- a relation record is treated as a method order;
- a dependency on a framework edition is treated as a specialization relation;
- a source or generated carrier is treated as architecture evidence without source-return and preservation claims.
The result is a framework that may look organized but cannot answer ordinary architecture questions: what structure is selected, what depends on what, what can change independently, what is preserved by a projection, and which stronger claim requires another pattern before it is used.
Forces
Solution
Describe an FPF-grounded pattern ecosystem as a family of framework editions and publication and access-facing presentation carriers, plus access routes, over selected structures. For each durable ecosystem-architecture claim, or technical claim on which later work will rely, state the exact subject and relation and cite the defining or constraining ClaimGraph in its subject pattern. The smallest route below needs no ClaimGraph citation when ordinary guidance or a stop already answers the question.
A principle framework edition renders a selected architecture in pattern-language form for a declared reader and use. That architecture covers recurring problem situations, forces, known failure modes, reusable SoTA solution moves, consequences, cases, relation records, evaluation methods, and refresh conditions. Known failure modes include beginner mistakes and experienced-practitioner failures caused by stale, local-only, or non-SoTA practice.
Start with the smallest route that answers the current question:
-
Name the ecosystem question and who needs the answer.
-
Classify the likely case: a framework-family boundary, an adjacent result or service, a publication carrier, access-facing presentation carrier, or access route, a DPF-suite question, or another relation that an existing pattern addresses.
-
Point to the pattern that addresses the case and state the next useful move, or stop with the missing distinction.
-
Open the complete ecosystem-architecture record only when the answer must persist as ecosystem architecture or later work must rely on the selected structures and relations.
This route is ordinary guidance, not a new record or package. A supplying pattern or stop is a complete first result when no durable ecosystem-architecture record is needed.
When the available methods are insufficient for the needed result under the relevant conditions, use C.39 to find or develop a missing operation or connection. When a method and its connections need a reusable explanation in a framework, use E.4.CM to develop that explanation, choose its placement and provide a reader entry. An adequate available method can be used directly.
Create an ecosystem-architecture record only if the answer must settle durable architecture or support later reliance. Use the fields the claim needs:
This record answers the declared ecosystem question for its intended use. Include blockedOverreadRefs only when F.19's grounded-contribution test admits them; otherwise omit the field. The record carries a contextual ecosystem answer; the subject claims and patterns it cites remain its content sources and semantic loci.
Classify the family members as follows:
Conceptual Core names FPF's general conceptual content within the existing Core, Tooling Reference and Pedagogical Companion family partition. An FPF edition is an identified body of framework content. Its publication units present that content, and readers obtain them through available access routes. The FPF Core pattern set consists of the selected Core subject patterns in that FPF edition. A framework-edition dependency names the relied-on FPF edition and the Core claims required for its use.
The Kernel is the subset of Core content that defines admitted universal meta-concepts and the rules for their meanings and relations. Membership follows that defining content, wherever it occurs in the patterns. Other Core content explains reusable methods, states their conditions and grounds, or governs the framework. All eleven E.2 pillars remain constitutional first principles outside the Kernel subset; they govern both Kernel and other Core content.
Readers obtain the relevant content from an FPF publication and use the supplying patterns for the question at hand. A pattern can contain both a definition and an explanation of its application. A Part heading, file boundary or selected rendering therefore does not define the Kernel's membership or constitute a separate Kernel edition. The Architectural Reference explains the framework as a whole; the pattern bodies retain the definitions, conditions and methods on which a use relies.
A domain or local framework identifies the Core content and edition on which it relies. That content may include a Kernel definition, a general method or a constitutional authoring requirement. The dependent framework supplies its domain or local problems, methods, sources and use conditions. Authors can revise those contributions while preserving the relied-on meanings and obligations; a change to a supplying claim reopens the uses that depend on it. A shared publication carrier supplies access to these frameworks without merging their content or edition boundaries.
Place support units and adjacent products
In this pattern, product is Plain management wording for a deliberately identified result or service boundary. It helps a team state intended use, identity or current state, access, later change and retirement rules, and any maintenance that obtains. It is not one FPF technical kind and it creates no U.Product.
Before making a product-boundary claim, name the direct subject—the thing the claim is about—and the relation that carries its identity, edition, current state, provision, publication, availability, or maintenance. The subject may be, for example, a framework-edition episteme, an evidence-package episteme, an admitted System, an admitted service arrangement, a Method, a programme-description episteme, or another result already admitted by its subject pattern.
Constitution or publication establishes only the claims made by those acts; maintenance and future Work need their own evidence. If the direct kind or relation is not settled, keep the management boundary as a proposal and return that question instead of inventing a common object kind.
A framework edition is an episteme. Treat its Readme, architectural reference, table of contents, pattern-body collection, framework-scale structure or coverage account, relation or edition note, and refresh route as named publication units in the same managed boundary when they share the edition's declared readers and use, edition boundary, access, and change rule. Being outside the pattern set or in another file does not by itself create another product or a maintenance claim.
Make a separate adjacent product only when people need to change, cite, or use its direct subject independently. Look for an independently useful identity, edition or current state, named users and use, an intensional rule for what belongs, access, a later-review or retirement rule, or cross-framework reuse or reliance. A separately established maintenance relation may also matter, but product identity does not require it. Examples include a registry, MethodDescription collection, evidence package, guide, tool reference, access service, or inquiry programme; other direct subjects may also justify a separate boundary.
The label does not settle the kind: a guide or evidence package may be an editioned episteme; a tool reference may identify an episteme, a tool System, or both; and an access service needs its own service and provider-System claims. File location does not decide the boundary.
When the direct subject is independently used or changed, keep it separate and point from the framework to the edition or current state being used. An annex may carry a declared snapshot or projection, but it returns to the authoritative subject and does not fork it. When no independent boundary is useful and ordinary framework use needs the material, keep it as a named support publication unit of the framework edition.
An architectural reference is the public account of how a named method language is assembled and why that arrangement helps its intended users. It explains the whole's purpose, constituents, relations, shared conditions, alternatives and limits, with the rationale and source synthesis needed to understand and adapt it. Give the unit the full public name <name of the explained whole> Architectural Reference, for example FPF Library Architectural Reference, Engineering Suite Architectural Reference or Method Engineering Principle Framework Architectural Reference. A shorter mention is sufficient once the explained whole is unambiguous. The qualified name identifies the account's subject; it does not select a filename, section key or independent product boundary.
Provide this account at each level of language assembly: patterns in a framework, constituent frameworks in a Suite, and Core, Suites and standalone frameworks in FPF Library. Each account explains the connections added at its level and returns to the full constituent accounts and patterns. That descriptive assembly does not make every Method part of one composite Method or every described discipline part of one discipline. A Suite's disciplinary composition still needs C.20; a library-wide connection can instead be use, specialization, shared terminology or another stated relation.
The account belongs to the explained language. In a one-file framework it can be embedded; in a multi-file assembly it has a discoverable unit at that assembly level, which can be carried separately. Keep its relation to the whole explicit in either form. Source hosts, directories, generated pages and site slices do not create further language levels merely by dividing files. Do not duplicate an account for each carrier, and do not require it before a known pattern can be used directly.
One presentation carrier may expose several managed products without merging their subjects. Each constituent keeps its identity, edition or state, form, access, later-change and retirement rules, and any separately established maintenance relation; the outer navigation names the constituents and stays neutral. A result reused by several DPFs may therefore be managed as an ecosystem companion or service product. Shared use does not make it a parent DPF. Open another DPF only when its field-boundary assessment finds recurring practitioner problems, constructive Methods, an independently useful first cut, evidence practice, and its own edition and change boundary.
When programme is used, start with what continues. An inquiry programme may be managed as a continuing programme or service product, but neither label says what persists. If a subject pattern admits the programme as a System or another arrangement, name it. Otherwise name the current programme-description episteme and any provider System, maintenance relation, accepted commitment, or service state that independently obtains.
Bounded inquiry projects require independent A.15.1 admission as separate Work occurrences, and their results remain separate epistemes. A maintained inquiry evidence package is its own editioned episteme. The management boundary may coordinate these subjects and relations, but it does not turn them into one indefinitely continuing U.Work or one generic Product. If the persisting arrangement is still unclear, return that architecture question.
DRRs, build manifests, quality runs, digests, logs, and campaign state remain development or process evidence by default. Publish the architectural reasons that affect use in the language's architectural reference or the relevant pattern, so readers need not reconstruct them from those records. The dated development, review and release history stays in its records. A selected public use can give a record its own product identity and publication route; ordinary understanding of the language does not require that separate publication.
Use these tests in order:
- Name the intended managed boundary and ordinary use.
- Identify every subject the decision makes a claim about, its kind and the identity or current-state relation the decision uses for that subject.
- Group only publication units that share the framework edition, readers, access and change rule.
- Test a proposed adjacent subject for independent use and change.
- Select the smallest useful boundary.
- Record exact pointers to those subjects, the return from each snapshot and the navigation from the neutral carrier.
Establish maintenance only when a maintained claim is made. If a needed kind or relation remains unresolved, record that question and stop short of the technical product claim.
Describe a composite discipline with a DPF Suite
Use this branch when a proposed field contains several subdisciplines and the reader needs both their own Methods and the Methods that connect their contributions. First establish the discipline and its composition through C.20. Then describe that whole in a DPF Suite. A convenient selection of publications, several suppliers to one project, or a common audience can justify a reading route without establishing a composite discipline.
A DPF Suite is a domain principle framework for a composite discipline, assembled from the frameworks of its constituent subdisciplines and any justified whole-level patterns. It is a larger DPF, not another ontological kind of field and not a collection constituted merely by a publisher's inclusion decision. The described discipline is a U.Discipline; a Suite edition is a U.Episteme about that discipline. The constituent DPF editions describe its subdisciplines. A directory, website or combined file presents those descriptions. None of these is identical to the discipline.
Recover the field before arranging its descriptions
State the field's recurring questions, objects of inquiry or intervention, characteristic results, knowledge and reusable practices. Use F.18 for a name that makes that field recognizable. A university subject, profession, conference or community is useful evidence of an established field; it does not by itself establish either the field's exact boundary or a constructive part relation.
For each proposed subdiscipline, recover the C.20 construction:
- the knowledge-bearing and reusable-practice contributions it makes to the larger field;
- the relations by which those contributions constrain, enable, test or revise other required contributions;
- the whole-forming arrangement, including admitted alternatives and how a part can be replaced without losing the field;
- the field's identity through changes to theories, Methods, instruments and institutions;
- a whole characteristic that the composition produces or sustains, with its grounds; and
- how this whole can participate in a larger disciplinary assembly.
Explain the coupling in practice. For example, an account of ground behaviour constrains a foundation design; the design identifies what site evidence is needed; observations and failure investigations can revise the account and design Methods. A bibliography listing geotechnics and structural engineering supplies neither this coupling nor a framework that teaches it. Nor does every construction project need to use every branch of those fields.
Use this field construction to reconsider the current Suite architecture. An intake may justify a new Suite, a split or merger of existing Suites, or a change to their boundaries; inclusion in an existing Suite is one possible outcome.
A Suite may describe a bounded, explicitly incomplete portion of a recognized composite discipline. State which constituent subdisciplines and whole-level problems it currently covers, which are outside that coverage, and what the reader can obtain. Partial coverage must not be presented as a complete account of the field. A planned constituent with no usable description is a gap, not an available DPF.
A first Suite edition may contain only one available constituent DPF when the composite field, its other constituents and their whole-forming relations are already established. The common account must offer a present use: for example, explaining where the available contribution applies, which connections it relies on and which needed results the edition cannot yet supply. Compare this benefit with keeping the DPF standalone until a common assembly is useful. One DPF plus an unspecified intention to add more frameworks does not establish a Suite.
Distinguish composition, specialization and use
Do not infer one relation from another:
An engineering decision may need finance, research, mathematical modelling and ethical judgement. Their use does not make their entire disciplines parts of Engineering. Conversely, a subject-specific engineering branch does not become an external supplier merely because its Methods are used in only some projects. Test its participation in the continuing field assembly, not the participant list of one Work occurrence.
Do not force the discipline structure into a tree for filing convenience. If multiple parthood is proposed, establish each direct relation and explain the shared contribution, whole-forming couplings and consequences of change in each assembly.
First test whether the apparent overlap is a general Core Method, an independent supplier, specialization or two different subject-specific contributions. If it is, publish that distinction instead of duplicating whole DPFs. If shared constituency remains, identify one current constituent description and explicitly assemble it into the affected Suite editions; a casual hyperlink or an unqualified second copy cannot carry that claim.
Build the larger pattern language
Develop the Suite as a framework under E.4.DPF and E.8. Its entrance explains the composite field, the contribution of each constituent and the important relations between them. A long mantra can help a practitioner retain a connected Method, but neither that phrase nor a Reference substitutes for the Methods it recalls.
Give a whole-level Method its own pattern when it removes a recurring difficulty that the constituent Methods do not independently solve. State its problem frame, problem, forces, construction, limits and cases. The construction may consume results from several DPFs and return questions to them. If a contribution is transdisciplinary, develop it in Core instead of hiding it in a Suite introduction. If the missing work is specific to one subdiscipline, develop it there. If it reveals another field, reopen the field architecture rather than stretching a pattern or an appendix around it.
Expose enough of the language for a reader to select constituent Methods, understand their contributions, combine them where justified and retain their separate assurance obligations. Preserve alternative configurations; a Suite is not a mandatory sequence through every DPF. Use E.11.PFP for the Suite's framework entrance and constituent frameworks. Use E.11.DSG to develop the Suite's architectural reference, including useful problem-led entries. It covers the whole's architecture and can return both constituent and external contributions, labelled by their relation.
Develop a missing operation or composition where the practitioner needs it. Complete the existing pattern when the operation is part of its promised Method. An independently reusable Method with its own working problem and result can warrant a separate E.8 pattern. The architectural reference explains how contributions fit together and directs readers to the construction they will use.
A teaching course or Guide with a separate learning purpose is a pedagogical companion, while explanations and cases needed to use or adapt the language remain with that language.
Make the publication reflect the architecture
Give each current full Method description and constituent DPF an unambiguous authoritative home. Assemble the Suite's publication from its selected constituent editions and its own framework-level explanation; make that composition and the editions used recoverable. A Suite directory can be a convenient physical assembly of the complete constituent publications. A standalone DPF needs no artificial Suite. The FPF Library Architectural Reference explains their library-wide connections and boundaries; the library index locates the publications. Neither the index nor that account establishes another discipline above them.
Keep four kinds of change distinct:
- A change in the discipline's knowledge, practice or composition is judged through C.20's identity and part relations.
- A change to the framework description produces or revises a framework edition under C.2.1 and C.2.P.
- A change to files, layout or access can preserve that edition while changing its presentation or availability.
- A change in publication or maintenance responsibility changes those relations. That change alone does not change the described field.
An editorial decision can select what a Suite edition describes. It cannot make a non-constituent field become a subdiscipline, and it cannot preserve an unsupported part claim until someone formally removes it. Correct the description when its claim is false; qualify incompleteness or uncertainty when the evidence is insufficient. Conversely, an unavailable file or retired framework publication does not erase a discipline.
For a release or reproducible use, identify the Suite edition and the constituent editions it assembles. An optional configuration description can pin those selections; it remains an episteme about the assembly, not the assembly's disciplinary subject. A separately maintained Reference can evolve independently while preserving its source returns. Apply E.4.PFIP when reconstructing publications: preserve useful Methods and their conditions, correct erroneous architecture, and update maintained links to the current full explanations.
Keep edition series separate from disciplinary structure
When a use needs continuity across releases, a framework product series is a continuing collection of edition epistemes. A constitution decision starts it by identifying at least one existing edition, its intended readers and use, content-selection and edition-admission rules, a reidentification rule, and later-review and retirement conditions. When that decision takes effect, the series begins and the first edition is admitted. The decision Work and its record remain separate from the series and the edition's belonging to it. Publication, maintenance and future availability require their own claims.
A later edition joins only when its source-edition relation and the series rule hold and a further admission decision takes effect. A fork or translation is not admitted by provenance alone.
The series need not have one total version order. Supersession, unavailability, non-currentness or edition retirement does not end the edition's belonging while the same series continues, and a return to availability does not create a new occurrence.
If the series ends or is reidentified as another series, current belonging to the old series ends; the past fact remains. Another series needs its own admission decision. If a claimed earlier admission never satisfied the rule, correct the false claim instead of preserving it as historical truth.
The same publishing arrangement can be used for a constituent DPF, a Suite framework or an independently managed Reference. It describes their publication continuity, not DisciplinePartOfRelation. A series may end while the described discipline continues, and a revised Suite edition may correct its account of that discipline. Establish publication, availability and maintenance separately through E.17, E.24.PUB, C.2.P and G.11 when those claims matter. Do not make a routine reader reconstruct admission decisions merely to use a current Method.
State the ecosystem architecture for the declared use
For a durable ecosystem-architecture answer, use the following operations as their conditions apply:
- Declare the ecosystem scope and intended architecture use. Cite the exact source, pattern host, selected architecture structure, publication relation, or bounded model-use structure only when the record relies on it.
- Name the family member being created, used, or changed.
- Name only the fields from
FPFEcosystemArchitectureRecord@Contextthat this architecture claim uses. For PF work, the pattern-language publication carrier exposes a reader-facing expression of the selected problem-and-solution architecture. - If the family member is FPF itself as a framework edition, open
E.4.FPFfor form, presentation carriers, access routes, and whole-FPF adequacy routing. - Identify the content another edition supplies and the work that needs it. For Conceptual Core, Tooling and Pedagogy, check the permitted import directions in
E.5.3. For framework editions, applyE.4.PFR:3.4. An LPF may rely on a DPF, and either may rely directly on Core. Core must supply its transdisciplinary rule and grounds independently of those domain and local frameworks. - State publication and first-entry claims using
E.11andE.17; state framework-carrier structure-account assertions usingE.4.FPFfor FPF itself orE.4.DPF/E.4.DPF.DAfor domain and local frameworks. - State pattern-use recommendation claims using
E.11.PUR. - When a framework-architecture question is open, record the selected answer in one
E.9DRR and useE.4.PFADto profile its framework-specific content. UseC.32.PADonly for a project architecture decision andC.32.ADRonly to project such a decision into an ADR-like publication. - State relation, dependency, compatibility, deprecation, and edition claims using
E.4.PFRonly when its named maintenance use requires that representation; otherwise use the direct subject assertion. - Settle names using
F.18. - Ground source-use claims in the relevant source content. Apply
E.8:11to a SoTA comparison. Reuse a sufficient direct source answer; useG.2when a broadCG-Framesource pack and its downstream handoffs are needed. - State currentness, refresh, and edition-change claims using
G.11, the exact edition values, and their source/currentness assertions. - Before using a carrier or transformed or generated view as evidence, state the exact source-return or preservation assertion under the predicate defined in
C.33,C.34, orC.35. - Evaluate whole-FPF adequacy through
E.2.DA, DPF or local-framework package adequacy throughE.4.DPF.DA, individual pattern quality throughE.21, improve throughE.23, and useE.19for a selected admission, refresh, or return-for-repair question.
When placing a foundational contribution, identify what would settle a disagreement about it. Give a Core rule the premises its users need. If several Core definitions depend on one commitment and their separate passages cannot reliably keep it consistent, state it once in Core and refer to it from those definitions. Put a method of deriving those premises or testing their physical interpretation with the discipline that can correct it. A separate edition earns its place when independent consumers must adopt the same commitments and check compatibility as they change. A reading order can be supplied in the Reference. Use B.5.MR when the claim concerns refinement or abstraction between models.
Core, DPFs and LPFs can supply models at several levels. When each pair passes the B.5.MR comparison for the same inquiry, read the chain from a formal construction through a transdisciplinary principle, a disciplinary model and a local model towards an individual model as levels of model refinement. In reverse, read levels of model abstraction. A DPF can carry a foundational construction as well as a disciplinary model. A method for testing a physical interpretation belongs with the discipline that can correct it, even when its result supplies a premise for Core.
Use this routing table when a proposed change is ambiguous. Its rows are common routes, not a closed taxonomy:
This pattern should leave the reader able to state the architecture directly. Name the family member and selected pattern-language architecture, then state its dependency editions, publication or access carriers, preserved structures, and each neighboring claim under its exact predicate with the subject pattern as locator.
Archetypal Grounding
A team developing a hydroponic-cucumber domain framework uses FPF's distinction between a method, its description and performed work, together with E.8's framework-authoring requirements. Horticultural sources supply the crop-specific methods and conditions. The team identifies the relied-on Core content and edition, develops its domain patterns and publishes the framework for growers or agronomists. A changed crop-specific threshold reopens the affected domain explanation and its uses; it does not by itself change the meanings of Method or Work.
The edition labels in this example are illustrative. FPF@C1 names one stipulated FPF edition containing the Core claims cited here. HydroponicCucumberPF@2026Q3 uses A.3.1:4.3 from that edition to distinguish the nutrient-monitoring method, its description and performed monitoring work. Its revision guidance also applies E.8:4.1.2, item 6, from the same FPF edition: authors repair examples and direct consumers made stale by a changed pattern interface in the same authoring increment. Removing or materially changing either relied-on claim reopens the corresponding domain explanation or revision guidance.
Mini-example:
Local practice. A Codex-process local practice framework may depend on FPF Core and selected architecture-domain patterns. Its handoff patterns, prelanding patterns, and process runbooks are local framework material. A Core-amendment decision under E.9 remains the route for changing FPF Core.
Generated relation graph. A graph over pattern names can help inspect missing relation assertions. After C.35 admits the generated result for its intended architecture use, state each supported relation directly. Open a reusable E.4.PFR row only when a named maintenance consumer requires it.
Shared carrier. In the cucumber DPF, the Readme, table of contents, pattern collection, and coverage account share one framework edition, reader use, access route, and change rule, so they remain publication units of one product. A greenhouse-calibration source registry has its own edition rule and is reused by another crop DPF, so its current registry edition is a separate episteme. One web carrier may expose both while preserving their identities and direct relations.
Bias-Annotation
Scope: limited. This pattern helps make architecture claims about FPF-grounded framework ecosystems and their publication, access, companion, service, and separately established maintenance boundaries. Product taxonomy, service design, programme ontology, and content management remain with the patterns that define them.
The recurrent drift is publication-first architecture: the visible file, all-in-one carrier, card deck, table of contents, or graph is treated as the architecture because it is what a reader sees first. The repair is to name the selected structures and dependency direction first, then use publication patterns to expose them.
A useful local rule may suggest a Core amendment. Identify what makes the rule hold across the domains where readers would apply it. If its conclusions still depend on one practice's convention or authority, publish those conclusions with that practice. Credit the local source for the discovery when a transdisciplinary argument can stand on its own.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
This pattern makes FPF ecosystem work slower at the beginning because a framework author must name family membership, dependency direction, selected structures, and the patterns needed for neighbouring claims. The gain is that later work can evolve without hidden Core changes, hidden publication substitutions, or hidden source loss.
It also makes some attractive names and short labels provisional until F.18 settles them. The delay is deliberate because short names are useful only after the value being named, its source-local meaning and its intended use have been stated.
Rationale
An ecosystem-architecture record identifies the selected structures across FPF patterns, frameworks, source packs, presentation carriers, access routes, quality records, and decisions. Direct assertions state relation meaning and decision rationale. Source-return and currentness patterns qualify carriers and access routes. Architecture work therefore names the selected structures and applies the pattern that defines or constrains each neighboring claim.
The Core, Tooling Reference, and Pedagogical Companion distinction remains valuable, but it is only one family partition. Domain and local principle frameworks need their own framework editions so they can depend on Core without redefining it.
SoTA-Echoing
Use official catalogues, vocabulary standards, current release pages, tool documentation, lineage sources, and source-maintenance checks only for their stated source or default contribution. Product kind, service kind, publication identity, relation truth, and SoTA rank each require their direct evidence and pattern.
Relations
-
Builds on:
E.2/P-5 FPF LayeringandE.5.3for modular extension, directed dependency, and family-order discipline. -
Coordinates with:
E.4.FPFwhen the work concerns FPF itself as a first-principles framework edition, its presentation carriers, access routes, and whole-FPF adequacy route. -
Coordinates with:
E.2.DAwhen the scoped FPF object needs whole-FPF Pillar adequacy evaluation. -
Coordinates with:
E.4.PFADwhen the ecosystem-architecture record opens a framework-architecture question;E.4.PFADprofiles the framework-specific content,E.9supplies the decision-record method and content requirements, and the resulting DRR records the selected answer. -
Uses:
C.20to establish a discipline and its composition before choosing a Suite architecture. -
Coordinates with:
E.4.DPFwhen the work is to author a domain principle framework, including a DPF Suite, or a local practice framework. -
Coordinates with:
E.4.PFRwhen a named maintenance consumer needs a reusable relation, edition, dependency, compatibility, deprecation, or preservation record; otherwise state the direct relation assertion. -
Coordinates with:
E.4.DPF.DAwhen a domain or local framework package must be evaluated as a package rather than as an average of its pattern bodies. -
Coordinates with:
E.11for discoverability,E.11.PFPfor the common publication form of FPF, DPF, or LPF constituents,E.11.DSGfor architectural references and their problem-led entries at framework, Suite and Library levels,E.11.PURfor pattern-use recommendation,E.17for a source-backed publication face and return to source, andE.24.PUBfor the publication occurrence, form, carrier, audience, bounded use, and availability. -
Coordinates with:
G.2,G.11,C.33,C.34, andC.35for source, currentness, preservation, and admission of generated or discovered results for architecture use.
E.4:End
Develop Connected Methods as Framework Contributions
Type: Method pattern Status: Stable Normativity: Normative
Problem frame
Use this when you author or maintain FPF or a domain principle framework and its readers can find relevant patterns but have to invent how to use their methods, rules or results together in a recurring situation. They may need one composite Method or an explanation of relations among independently usable Methods. A source synthesis, worked solution or practical-use card may contain part of the construction; independently developed methods may leave its connections to be built.
The broader work is developing Methods for useful action. C.39:4 supplies the general way to construct, connect, explain and change them without assuming framework authorship or a pattern format. This pattern applies that way to a public framework explanation. It adds comparison with the available framework supply, choice of a maintained public owner, the content functions and form of E.8, and preservation of direct uses and shared explanations under E.11.PFP:4.7.
The gain is that later practitioners can perform, combine and adapt the Methods from the public framework without reconstructing essential connections from private development material. A connected account need not assert one composite Method. The number of patterns or levels does not establish that identity.
First useful move. Start with the practice the recipient needs to understand, carry out or design. What result must its explanation let that person obtain, and which choices does the work leave to them? Identify the practice at issue. The Method for designing a practice can have different constituent Methods from the one its practitioners perform. Where composition matters, establish what is part of what for each relevant kind of whole. Does the explanation need a rule that holds at each scale, an inference from one scale to another, or the result that the constituents can produce together? At each scale choose what the model must distinguish for this reader’s work. Use an available account or attempt to establish a needed relation. If it settles one step, follow that contribution back to the whole entrusted work. Develop the missing connection through C.39:4.2–4.6 and place it where the practitioner needs it.
An adequate public Method can be used directly. Use C.39 alone when the result is an ordinary practical way or description with no framework-publication need. A private one-off solution need not become a maintained pattern. Repair discovery when the complete explanation already exists and only its entry is missing. E.4.CM requires neither a new primitive operation nor a new framework for each useful composition. C.32.MWA supplies a related architecture account when the practical decision concerns Methods together with performance, support and development.
Problem
A catalogue can cover each constituent operation and still leave a recurring whole problem unexplained. A practitioner must then select compatible operations, resolve different meanings, recognize what belongs to the environment, and determine which lower action changes when a higher requirement changes. A catalogue can also leave relationships among independent Methods unexplained: where one receives another's result, which conditions they share, and when a failure calls for developing a method rather than repeating it. Only readers who can invent these connections benefit from the catalogue.
The opposite failure is to publish a broad umbrella that repeats its suppliers without constructing the connections. Its apparent completeness hides an unavailable operation or treats alternatives as mandatory stages. A third failure uses the author's successful synthesis as evidence that the synthesis need not be made available to anyone else.
Forces
Solution
Apply C.39's general construction and explanation to the framework reader's recurring difficulty, then give the resulting content its appropriate public owner and form. For a composite Method, explain the organization that produces its whole result. For independent Methods, explain their material relations while preserving direct use. The steps below add the framework-authoring decisions; they do not make framework publication a prerequisite for developing Methods.
Recover the working question before assigning parts
Start with the practitioner, working situation, sought result and conditions that make the result useful. Recover how the practice is supposed to produce that result. A general argument, an actual attempt, a failure or the proposed conditions can supply the grounds. Select the material for what it can establish. Keep the original need when exploring a smaller attainable result. If the difficulty is not yet named, follow an ordinary result to its receiving use with B.5.PI: a document can have been delivered while its recipient still cannot continue the work.
Who requested the result, what did the performer agree to supply, and who will use or accept it? Distinguish a proposed change of product or date from the revised undertaking the parties actually agreed. Where the conclusion depends on a handover, establish the act the agreement calls for and what shows that it occurred. Depositing an identified report can finish an order to deliver it. If the order requires acceptance, identify who can accept it and which act has that meaning under the agreement. An approved result from a supplier can finish an order to obtain it. When the reader must devise the way of working, identify the decisions they must make with the information and help available to them. When their assignment is to find out what result is needed, begin with the actions of its prospective users. If two interpretations would change the promised action, find out who can settle the difference. If both support the same action and its required justification, give that answer while the other difference remains open.
Find candidate methods, rules and results by the change needed. In each candidate description recover the input, action, result, conditions and limits. A shared topic helps find a text; the receiving action determines what can be used from it. Include the necessary performer, preparation, means and access: an obtainable description does not supply them.
For a source-derived Method, recover the source's constructive synthesis as well as its component results: the proposed objects, relations, operations, dependencies, choices and claimed result. Verify those contributions at the scope being used. Known ingredients are not grounds for discarding the source's organization; the source's novelty claim is not proof that the organization works.
Method-engineering sources may call their reusable units fragments, chunks, slices or practices. Recover what each unit contains and what its source composition rules mean. Transfer the useful selection, adaptation or joining move, then establish the receiving FPF claim through A.3.1 and B.1.5.
A source unit can combine an action account, notation, role, deliverable and guidance; its name alone does not make it a constituent Method. Use E.10:0.2c.21a for explanatory naming without erasing the source's distinctions.
Trace what participants do, what each action needs and what it makes available. Include preparation or support that changes the result. Distinguish constituent actions from documents describing them, learning how to perform them, and environmental conditions. A table of pattern names describes neither their enactment nor their composition.
Ask where the same connection will matter again: another input, performer, receiving use, configuration or scale. That variation gives a candidate reusable problem. It need not establish a universal law or require observed failures in every discipline. State a proposed use as proposed.
Use B.1.5 to determine whether the actions form one Method: the whole needs its own reusable action, applicability, result, constituent relations and preserved conditions. Keep a repertoire of alternatives as a repertoire when no common whole is being asserted. A composition can contain other compositions, and a constituent can serve several wholes. Neither Method depth nor the number of patterns establishes a framework boundary.
When several Methods remain independently usable, select the relations their readers need explained. Preparing a recording, analysing movement in it, showing it to an audience and developing a better preparation method can be connected without constituting one Method. Recover the result uses, shared conditions and returns that make a difference.
This connected account is ordinary content about Methods and their uses. Call it a MethodDescription of one Method only when that Method and the description's membership have been established under A.3.1 and A.3.2. A list, diagram or larger time window does not establish a new whole or a U.Structure.
Develop the missing way before choosing its framework form
Use C.39:4.3 to work backward from the reader's needed result and forward through a proposed connection. Its difficulty/remedy account in C.39:4.3.2 covers incompatible intermediate results, whole conditions, missing useful applications, hidden assistance, replacements, unavailable coordination, unclear returns, mistaken distinctions and unwarranted sequence. Work on the failure you found, using the contribution that supplies its remedy when you need it. The table need not be traversed for every framework contribution.
For framework authoring, distinguish two gaps. An operation or relation may be missing from the proposed Method; construct it with the relevant professional Methods. Alternatively, the way may already be understood by its developers while the public explanation fails to teach it. Develop that explanation through C.39:4.5 and E.8:4.2.1.1. A reader's successful private invention exposes useful candidate content; it does not establish that the framework already supplies it.
Follow the construction to its receiving use and include necessary preparation, performer capability, means and access. Unknown facts, wrong object distinctions, missing operations and incompatible demands call for different remedies. Two of them may arise at the same join. General authoring guidance can explain the needed contribution without possessing a professional means to supply it; name what is missing instead of promising that the action can be carried out.
Then identify what the public framework must carry: the reusable action or relation, its reason, applicability and limits, and an obtainable return to each supplied explanation. Preserve source constructions as well as their ingredients. A familiar set of operations can still have a useful new organization through selection, joining, coordination, adaptation or return.
Apply the relations at the framework's chosen scale
C.39:4.3.1 supplies the general distinctions among result use, continuation, Method composition, enactment, variation, refinement, replacement and cross-domain use. C.30.ASV:4.5a helps choose the relations needed for the reader's question. Develop their account for that question; an FPF or DPF publication does not require one universal view set.
Choose the composition scales and levels of model abstraction for the reader's decision with C.39.SA:4. Which operations form the Method? Which performances made up this job? Which premises support its conclusion? A system performs the Work, the Work can enact the Method, and a report can describe it. Trace the parts and the enclosing whole for each kind that matters. For each model ask which possible arrangements it must distinguish. A limit on total hours bounds the hours of each constituent job when all hours are nonnegative. To infer a result for the whole from its parts, establish how the parts combine. A rule can apply to every part while their interaction changes the result of the whole. To use a schedule and an authorisation together, establish that both cover the agreed work and time, or give the grounds for carrying a conclusion across the difference. C.37.MM:4.1.1 supplies that joint use, and B.5.MR:4.2 supplies the comparison when strict concretisation of one model is claimed.
For example, an author connecting recording preparation to analysis must explain how a corrected recording obtains its interval qualification and how the analysis uses that result. Three marker offsets alone do not justify the interval, and a correction number alone is not an edited recording. The public account must expose the independent interval basis and editing operation or state which of them is missing. C.39:5.2–5.3 develops both the missing-way case and the qualified conditional connection.
If several Methods remain independently usable, explain shared results and conditions, independent branches and selective returns. Where constituent operations form one Method, qualify that composition under B.1.5. Where their performance constitutes encompassing Work, use B.1.5.EW. Where continuations vary by condition, explain them through the ordinary question in A.22.CGUS and use its formal branch only for the stronger claim. A chapter order, heading tree or longer time window establishes none of those relations.
A source's procedure or reusable unit can guide construction without becoming an FPF Method part by name. Recover what the source makes available and establish the receiving claim. When two Methods are related by variation or refinement, explain which meaning is preserved and what changes under A.3.1:4.7–4.8. When one replaces another, follow affected uses with B.1.5.RS.
Keep framework-scope relations equally precise. FPF→DPF→LPF concerns framework scope and basis, not a ready-made composition or execution order. A DPF's use of a mathematical result does not place its domain below mathematics in a universal hierarchy. A profile can inherit shared explanation and state a narrower use without becoming a part of one composite Method. Several such relations may hold together; label each where it changes the reader's decision.
If a diagram uses “vertical” for Method composition and “horizontal” for result use, state those conventions under A.22:4.3a. Its directions do not provide the explanation. Keep the content with the source that supplies it when the reader can find and use it; add the missing receiving relation at the scope where it applies.
Compare the proposed contribution with what readers already receive
Use E.8:4.1.3 on the same working situation and comparable reader preparation. Follow the available public explanation to a result, then follow the candidate. Name the first changed action, result, condition, stop or amount of necessary reconstruction.
A reference to a supplier is sufficient only for the operation that supplier explains. If the author had to invent a consequential joining rule, that rule is a candidate contribution, even when the supplier list is unchanged. Conversely, if an existing public treatment gives the joining rule and its use, prefer that owner or repair discovery.
Test the proposed savings against its maintenance cost and burden. A second full copy of every constituent is usually worse than a developed whole explanation with precise returns to reusable suppliers. A wrapper that adds formal records but no useful move should be reduced or omitted. An incorrect synthesis does not become valuable merely because it differs.
Record the gap when neither treatment answers the question. Keep it open if an expert solves the problem with a contribution neither treatment gives its reader.
When selecting the composite would settle a consequential audience, support or shared-change boundary, use E.4.PFAD:4.1.1. Compare the complete useful arrangement for the intended human, human-with-AI or AI use, including acquisition and recurring costs; do not count constituent availability as delivered whole capability.
E.8:4.2.1.1 allows a competent-reader method account, fuller pedagogical companions and short reminders to serve different uses without multiplying authoritative Methods. A straightforward completion within an existing pattern can retain the local comparison above.
Choose the public owner by problem and scale
Keep the contribution in the existing pattern when it completes that pattern's promised result without introducing a separately selectable recurring problem. Give it a normal pattern when practitioners need to recognize and apply the whole's distinct problem–move–result relationship. Its constituents can remain independently usable patterns.
Make that owner decision from the action and result being supplied, not from whether the wording or its ingredients are new. A full explanation hidden in a Reference can describe a separately useful Method.
For example, a practitioner may need a repeatable way to assemble compatible answers, choose the next unresolved question and revise dependent answers after change. That Method of reasoning is distinct from the Methods whose answers it connects. Recover its whole problem, action and result and compare them with the complete candidate owner. Existing constituent Methods, a nonlinear teaching order and the absence of a new theory do not settle that comparison.
An Architectural Reference can explain why a language is organized as it is, how its Methods relate and how an already described Method applies to that assembly. Name the ordinary pattern that supplies such a Method. When a separately selectable recurring Method has no adequate pattern owner, give its full account one; do not leave it discoverable only as a Reference section. Preserve the complete explanation during the move, while keeping architectural rationale and useful applications in the Reference. This does not turn every explanatory relation or worked application into another Method.
Place transdisciplinary guidance in FPF and domain-specific fillings in the relevant DPF. Use E.4.PFAD when a material placement, framework boundary or cross-framework relation remains open. Shared constituents do not establish a parent DPF, and a cross-DPF method does not by itself require a Suite.
A developed example or reference treatment can be the right owner when it explains an application of an already available Method rather than a new recurring move. For several independently usable Methods, develop the material relations in the pattern language's Architectural Reference as E.11.PFP:4.7 provides. Explain entry, result use, composition where present, independent branches, changed conditions and returns. Such an account need not become a new Method pattern.
Preserve the reusable explanation at a stable public location needed by that use. Campaign decisions and private source notes can retain provenance, but must not carry essential practitioner instructions that the public owner omits.
There can be several useful scales of pattern. Select them by independently recognizable difficulties and reusable remedies, not by equal text length, one file per step, or a rule against “large” methods.
Apply the pattern functions to the developed explanation
First obtain the practical account through C.39:4.5–4.6: intended use and preparation, inputs and actions, reasons for the connections, supported result and limits, and the consequences of a changed condition. Its general questions about construction, fit and practical worth apply to the Method and its connected uses before any framework form is chosen. Select their evidence at the strength of the claim; a worked derivation is neither empirical effectiveness nor unassisted practitioner transfer.
For a pattern about that Method, apply E.8:4.1.4's twelve content functions.
- The problem frame helps readers recognize the use. The problem and forces expose the difficulty and trade-offs.
- The Solution explains the way.
- Grounding, biases, conformance and anti-patterns show readers how to examine its use and limits.
- Consequences, rationale, SoTA comparison and relations support informed choice and adaptation.
Each function asks for substantive content, not a filled heading. The functions organize and examine the developed Method explanation; they do not derive the Method from twelve labels.
For an account of several Methods, apply the corresponding questions to their material relations: why and where they connect, what each supplies, how their conditions affect use and what changes when one result or receiver changes. E.11.PFP:4.7 governs its placement in the framework's Architectural Reference and declared support units. Exact inherited answers can satisfy a question when the public reader can obtain and use them. Neither this account nor its profiles automatically become a new Method or pattern.
Choose the expression that makes those answers usable: prose, notation, reading order, media and publication location. Each declared pattern keeps E.8's body form; the whole account keeps E.11.PFP's publication form. These are additional obligations of this framework use, not requirements on every ordinary Method description. A translation or short entry can preserve the Method; omitting an applicability condition can change it even when every Solution step remains.
Develop a complete receiving case. Show which consequential conditions can change and why the conclusion follows over the range you claim. The general grounds can occur inside the example. For a choice assigned to the reader, start at the point before the example supplied its answer. Can the reader use the published explanation, their preparation and available support to make that choice? A heading, a form or a supplier may already have made it for them. Preserve the resulting work and use C.39.GA:4 to restore the missing step when the commission includes design. A different case can reveal a hidden restriction. An argument covering the allowed variation can settle the general question directly. Distinguish a change to the Method, another execution and a revision of its description. Reuse the branches and evidence whose conditions still hold.
Keep the grounds connected as C.39:4.5 requires: need, nature of the proposal, development choices, the content supplied and claims. The framework adds a publication concern: the arrangement described and examined must be the one its readers receive, with necessary source reasoning and support still accessible. A local repair belongs at the source of the connected omission. Short cards, headings and source lists cannot close an absent public explanation.
In §5.4, the timing requirements motivate a stipulated use, the minimax reasoning selects the constant correction, and independent professional premises support the conditional interval result. The framework account must expose those grounds. Nothing in the pattern form supplies evidence of improved movement analysis, lower learning cost or widespread studio need.
For a new pattern-quality or admission conclusion, apply E.21 and E.19 at their declared scope. Preserve valid source assurance and unaffected constituent results. A correct constituent, a compatible repertoire and a useful composite are different units of judgement; publication form does not make one answer stand for the others.
Expose a short entry and keep the explanation usable
A practical-use card or long mantra can remind a reader how to use one composite Method or several connected Methods. It should lead to their developed public explanation. E.11 selects when that entry is useful. A card that only helps choose among methods can remain navigation.
Discover useful combinations from recurring situations as well as from named patterns. Follow a result that readers repeatedly struggle to use, or a connection they repeatedly invent. Compare the methods, rules and worked constructions supplied by independently developed patterns and earlier treatments. Reuse a complete available construction; where none exists, develop the missing joins through §§4.1–4.6. A combination need not have been assembled in one source before the author can construct it. Conversely, co-occurrence in a source or a long list of relevant patterns does not establish a useful combination.
Choose a short form when it helps a reader recognize or perform the Method. A reminder can point to a full explanation available at the moment of use. A separately issued assignment must itself preserve the work entrusted to its reader, the choices it leaves open and the information available when those choices are made.
When a supplier or receiving condition changes, return to the affected connection and encompassing uses. Separate permitted Method variation from a different Method with A.3.1 when that question matters. B.1.5.RS governs replacing a constituent; E.19 governs a corresponding pattern-edition change. Retain the explanation's useful conditions even when a shorter public form is desired.
Produce source-grounded Method knowledge in pattern-language form
This specialization connects source recovery with a reusable pattern language for the source-derived construction problem in :4.1–4.7. For Method knowledge that needs no public framework, use the general C.39 construction and the applicable source, description and language contributions; publication is not a condition of that work.
Problem frame
Use this when. You have unlike sources and a recurring practice that needs actionable Method knowledge, but no adequate current description or language supplies the answer. You need to recover, reconcile, describe and connect the useful contributions while keeping their grounds and limits usable after change.
The intended reader is a practitioner developing Method knowledge or a framework author, able to examine the relevant practice and sources. The first useful result may be one justified allocation, a usable description, or a statement of which basis is missing. A complete language is warranted only when its connected answers repay their construction and maintenance.
Use A.3.2.DE directly when one description is enough. Use E.8.ML when the content is already recovered and allocated and only its language architecture is missing. Use C.39.MC when a finished language or existing result supplies the present continuation. Those narrower results remain valid exits.
Problem
A practitioner developing Method knowledge has a handbook, current research, an existing pattern library and a returned account of local practice. They partly agree, but neither chapter boundaries nor a ready-made process tells the practitioner which reusable actions, conditions and unresolved claims should survive. Later, a new fact can change the useful next question or the basis of an earlier answer.
Copying source chapters can preserve their words while losing the receiving action; reorganizing them first can lose a source condition or turn an observed practice account into an admitted Method. A finished language can hide the same loss when its entries cannot reconstruct the promised result or return to a changed source.
Forces
The author faces several tensions:
- Source fidelity and a usable synthesis.
- Reusable knowledge and situation-specific judgment.
- Enough preserved context and affordable reading.
- Local repair and hidden dependencies across patterns.
The construction keeps serious alternatives visible. A direct description or small synthesis note may be cheaper. A stable source-processing sequence may be appropriate. Neither is excluded by the ability to unfold use under changing conditions.
Solution
Produce the source-grounded language
Construct Method descriptions and their connections from the source contributions and current framework content. Keep each source's role and limits and the reasons for allocating its contributions. Test whether the language reconstructs the promised source contributions, and revisit affected explanations when their grounds change. C.39 governs the general development of that knowledge; E.4.CM applies when the receiving product is a public framework contribution. The result can contain several one-Method descriptions and neighboring contributions. Call that whole result one MethodDescription only when its subject and substantive claims satisfy A.3.2.
This account develops the connections among the existing recovery, allocation, description, language and reconstruction Methods. The following actions and result dependencies explain their use together without establishing a decomposition into part Methods. When the receiving use depends on a constituent Method, use B.1.5.CM/B.1.5 to identify it and establish its relation to the whole; B.1.5.EW recovers how its work is performed within the encompassing work. A further claim about one composite Method requires its own whole-forming grounds; the developed explanation and constructed case do not establish success in recurring production practice.
Inputs and applicability. Start with a named receiving practice and use, inspectable source contributions, their roles and limits, and the relevant current FPF/DPF supply. Recover the source's connected explanation before selecting its Method contributions.
When several actions rely on a principle or argument, use B.5.RA to recover its reasons and dependent conclusions; F.0.2 compares those explanations across source ontologies. Reuse an adequate recovered account. Definitions, explanatory principles and empirical claims retain their contribution and status rather than becoming steps merely to enter this Method-development branch.
Documentary Method contributions enter through A.3.1.DR. Observations, artifacts, logs and interviews about dated Work first require an A.3.1.MR recovery result. Investigate a remaining consequential evidence question through the relevant research practice. A returned candidate account retains rival explanations and gaps. If the receiving use or required source cannot be established, name the missing use or source as the first useful result.
Reusable action. Select source depth by the question and recover source-local meaning. Compare current supply and serious alternatives. Reconcile the surviving contributions without inventing Method identities. Allocate their useful claims and describe the Methods needed. Organize independently useful pattern questions and material relations, then attempt the reconstruction the language promises. Preserve a direct-source remainder where it is sufficient and recoverable. Each action consumes only the results it needs. A missing condition can return the work to that question before the rest is complete.
Source-profile variation. Use the profile that matches each source and receiving claim:
F.0.1/F.1/F.0.2 govern source meaning, selection and conceptual synthesis. F.0.2.MA consumes their results for the Method-specific allocation: reuse current supply, add or amend the professional contribution, retain a direct-source or unresolved remainder, or return a general question to FPF. The useful output states what changed and why, rather than merely listing source destinations.
Results and stops. The complete result for the selected use is an actionable language with identifiable description subjects, conditions, material relations, recoverable source decisions and an affected-refresh route. A smaller completed allocation, comparison or reconstruction result can be used independently without claiming that the whole language is complete. Stop a dependent use when its required basis is missing; preserve independently supported results. Do not promise corpus-complete coverage from a sample or gap-only pass.
Variation and identity. Source order, source profiles, tools, reader entries and representations may vary while preserving the explained actions and their source/evidence discipline. A change to the required allocation, source-return discipline, language result, reconstruction obligation or dependency direction reopens the affected construction. When the use makes a claim about one Method and its changed whole action, A.3.1.QM/A.3.1.MV govern that identity question separately. Changing a description or publishing tool does not by itself create a Method variant.
Enactment boundary. A prospective action description is not a dated Work occurrence or a WorkPlan. An account of a production attempt identifies its performers, conditions, support and observed results. During ongoing Work, A.15.7 supplies a separate next-action decision that names the chooser and performer. A.3.1.WT can obtain trial evidence; A.3.1.FT and C.11.MV separately judge transfer and worth. The constructed application below makes the description inspectable but does not supply those empirical results.
Make the joins between contributions usable
Recover a source-local answer before deciding where it belongs. A.3.1.DR supplies the documentary claims, actions, conditions and unresolved meanings. Keep the Work account recovered under A.3.1.MR separate from any additional investigation of its remaining consequential evidence question. Select the research practice for that question. These results can inform one production effort without receiving the same evidence status.
Give F.0.2.MA the recovered contributions and the receiving use. Its allocation must explain what current supply answers, what needs an addition, and what remains direct-source or unresolved. A matching topic or a destination filename supplies no such answer. If two sources disagree about a relied-on condition, preserve that conflict and return its interpretation or evidence question before using it as a settled premise.
For each allocated Method claim, obtain the identification or qualification it needs through A.3.1 and A.3.1.QM; retain a candidate account when identity remains open. A.3.2.DE develops the description for its named use with that status preserved. A useful source allocation can finish independently, but the language cannot promise an action whose necessary content remains unwritten.
Pass the developed content and its independent questions to E.8.ML. Its result establishes which body supplies which answer and how another use depends on it, chooses an alternative or returns after failure. It neither invents Method parts from body boundaries nor requires every user to enact production. When parthood is a needed claim, obtain B.1.5.CM's whole-forming account separately.
Challenge the promised receiving answer through F.0.2.RC, using the source contribution and the full pattern bodies. Include the argument when the reader needs it to choose, connect or adapt the Methods: can the public explanation recover the premise, justify the consequential inference and show when it fails?
Successful routing to a constituent pattern or completion of one local action does not establish that connected understanding.
A missing condition returns to its recovery, allocation or description owner; a wrong connection returns to E.8.ML. Retain unaffected answers while repairing the failed join. C.37.CF compares content and form only when that distinction can change the revision decision. No coverage claim extends beyond the recovered source and tested use boundary.
The connection is reusable even when the work starts at a later missing result. A sufficient recovery, allocation or description can supply its input; a source change can reopen an earlier reliance after later work has finished. Continue until the selected receiving use has its complete answer or a statement of the condition that blocks it. Use C.39.MC for the separate situated continuation question.
Archetypal Grounding
Worked connected application — H, L and W
This is a constructed application, not a report of completed empirical validation. A practitioner wants a small language for reviewing and adapting a Method description.
Handbook H, edition 3, §4, supports reuse of an earlier review after a display-only change. Paper L, edition 1, §2, limits that reliance to matching reviewed claims, question and qualification window. A returned MR account W, version 2, reports that a team shortened a warning; it preserves two rival explanations, harmless redundancy and a missed condition. The new language must not turn that report into an admitted shortcut Method.
First use. A.3.1.DR preserves H and L source-locally; W remains a separately returned candidate account. F.0.2.MA reuses A.3.1.CC's existing coherence contribution and carries L's narrower condition into the description. Tool-specific layout instructions remain at H. The allocation produces a conditional reuse answer and an unresolved W remainder, not a new Method count.
If that one description settles the receiving use, A.3.2.DE is sufficient and the practitioner stops. For recurring allocation, review and source-return questions, E.8.ML connects independently useful entries: the review relies on the allocated condition, and a missing coverage relation returns to the affected source decision. This arrangement does not establish three parts of a review Method.
C.37.CF compares earlier prose A with revised prose B that adds the matching-condition and source-return stop, then compares B with a table C preserving those same claims. Merely comparing A with C could not separate the added content from the new presentation. Textual comparison can reveal that A lacks a needed stop. A claim that C helps readers more than B needs an appropriate comparison of their actual use; the table's appearance is not that evidence.
F.0.2.RC reconstructs the promised reliance answer. A topic entry and source citation are insufficient if the condition cannot be recovered. Its first result names the missing condition or confirms recoverability only for this bounded case.
Changed condition. H becomes available only as excerpts. A new receiving use requires rechecking an argument outside them, and the retained allocation contains only its locator, not the argument. The recheck now has an unavailable basis. Return that dependent use to source recovery, use a separately justified sufficient basis, or narrow or stop it. Keep the earlier result with its historical basis, L's independently supported condition and W's unresolved status. Another pattern that copied the missing premise is affected even if its file did not change.
C.39.MC continues from this newly missing result. If only source availability changed, reconsider the current judgment. If the reader instead asks whether publication is authorized, identify that new use and its required result. A prior review is not publication authorization. In Work, identify the chooser and performer.
A useful answer to this case preserves the first action, condition and stop, distinguishes the three source roles, makes no invented part Method, and localizes reconsideration without losing another dependent use. A rehearsal of these answers tests the explanation. It does not measure production cost, human performance, transfer or causal superiority.
One sufficient direct description
A practitioner needs to check one retained review under a known matching-condition rule. The current A.3.2.DE description gives the action, source basis and unavailable-argument stop. The practitioner can use it without constructing a pattern language or reopening its source corpus. If a later recurring question asks where that condition came from or which dependent uses a changed source affects, that question can justify the missing allocation or reconstruction contribution. The earlier direct answer remains bounded by its conditions.
Bias-Annotation
Source familiarity, preference for a neat architecture and enthusiasm for pattern languages can bias all these judgments. Preserve the rival account and the cheaper direct route. AI-assisted recovery or comparison retains its model, input and interaction limits when they matter to the claimed result; fluency does not replace source access or independent use evidence.
Source prestige can hide a missing action; familiarity with the intended language can make its author privately supply an omitted join. Inspect what the receiving reader can recover from the allocated contribution and full suppliers.
Conformance Checklist
Recognition and assurance answer different questions. First identify the missing result and a plausible contribution. Then check only the claims on which the receiving use relies, with the evidence and independence that their conclusion needs. Do not treat finding the right heading, obtaining a graph path or reading all patterns as assurance.
For a production result, ask:
- Can the reader recover the receiving action and stop?
- Do the sources and candidate evidence retain their roles?
- Does every merge, split and omission have a reason?
- Does the language keep Method identity separate from body allocation?
- Can the reader identify where to return when a premise changes?
When the promised use spans several patterns, follow the contributions the reader needs into the same receiving work. Establish their compatibility wherever that work depends on it. One argument may establish several connections. Where only one contribution has been checked, return its result and identify the connections still open. Broaden the inquiry when the dependency boundary is unknown.
Common Anti-Patterns and How to Avoid Them
The recurring failures have distinct repairs. A chapter copied as a Method returns to recovery and qualification. A tidy graph that loses a condition returns to description content. A comparison that changes both content and form returns a bundled result, not a form advantage. A coverage table that counts topics returns to reconstruction. A source-access failure stops a required new recheck; it does not retrospectively falsify everything derived from the source.
Consequences
Recoverable source decisions make it possible to localize reconsideration when a premise changes. The burden is additional initial reading and maintenance of the dependencies needed for return. Use a simpler account when maintaining the language does not repay its burden.
Architectural Rationale
The production problem begins before language architecture because authors may still need to establish the source meaning their readers can use and decide where that contribution belongs. E.8.ML consumes those results to construct independent entries and material relations; C.39.MC uses an available answer under current conditions. Keep the full production explanation directly accessible here. Readers who need only language architecture or situated continuation can use E.8.ML or C.39.MC respectively without repeating adequate source recovery. They can also return to this specialization at a later missing result.
The existing C.39/E.4.CM construction, documentary and Work-account recovery, F.0.2.MA allocation, A.3.2.DE description, E.8.ML language construction and F.0.2.RC reconstruction supply the actions and material joins used here. Their connected use is the selected answer; a further independently selectable Method would need an additional action or whole condition that changes this same receiving use.
Situational Method Engineering (SME) contributes situation dependence and deliberate construction of reusable ways of working. The production construction keeps both premises and selects an arrangement in which patterns expose problems, forces, actionable contributions and useful relations, while Method identity and composition remain separately governed. The reuse unit is therefore not indiscriminately a source slice: it can be an independently qualified smaller Method, a description contribution or a neighboring non-Method answer.
This explains the distinct production results. Recovery must remain faithful even when allocation later rejects a proposal. Description comparison must not be replaced by either a cross-use representation profile or a judgment of Method worth. Language architecture must make material relations useful without creating parthood. Reconstruction must be able to find a missing contribution before a contradictory maintained claim and its owner are known. A.3.1.DR, F.0.2.MA, C.37.CF, E.8.ML and F.0.2.RC keep these differences actionable; none requires an obligatory pipeline.
SoTA-Echoing
The following sources supply questions and alternatives, not a proof that this construction is universally best:
Alexander et al.'s A Pattern Language (1977) supplies the historical problem-first, connected-guidance idea. Minsky's frame account (1974) connects situation knowledge, expectations and adaptation. The architectural analogy drawn here concerns those knowledge-use roles; historical descent of pattern languages from frame systems is not established by it. Declarative and case-management approaches, including CMMN 1.1, remain operational comparison targets, not the ontology adopted by this language.
The production construction uses Core's general qualification, architecture, representation and assurance Methods. Their shared use does not establish a separate discipline or require another framework. A domain framework would need a different justified field promise. A prose-only synthesis remains a lower-maintenance option; a fixed route remains legitimate under stable conditions. Reopen the production choice when current supply removes a distinct result, reconstruction fails for a promised use, source conditions change, or the maintenance burden defeats the practical gain.
Relations
F.0.1/F.1/F.0.2 supply source interpretation, selection and conceptual synthesis. A.3.1.DR supplies documentary recovery; A.3.1.MR recovers an account from Work evidence; an additional consequential evidence question calls for the relevant research practice.
F.0.2.MA allocates the useful content; A.3.1.QM and A.3.2.DE preserve Method qualification and description boundaries. E.8.ML constructs the language, and F.0.2.RC challenges reconstruction and refreshes affected explanations.
C.37.CF separates content and form comparisons. C.39.MC supplies independent situated use. A.3.1 and B.1.5.CM govern any claimed production Method identity and composition when those claims are needed; the supplier list alone establishes no decomposition. A.3.1.WT, A.3.1.FT and C.11.MV retain trial, transfer and worth questions.
C.39 supplies the general construction for a use that does not require framework publication. This E.4.CM specialization connects the source-grounded production contributions for a public framework use; E.11.PFP governs preservation of their complete public explanations and direct entries.
Archetypal Grounding
Assemble a comparison that another reader can use
A team has two public operations: grouping records and comparing paired observations. Its receiving question requires each error/non-error pair to share both setup and temperature. The existing grouping uses only setup.
The available data are T1=(error, S1, warm), T2=(acceptable, S1, cool), T3=(error, S2, warm), T4=(acceptable, S2, warm). Grouping by setup offers T1–T2 and T3–T4. Only the latter satisfies the comparison's precondition.
The author locates a composition difficulty: the producer's grouping key is weaker than the receiver's eligibility rule. The repair groups by setup and temperature, returns T3–T4 as eligible and states that no S1 comparison is available. Comparison then uses the eligible pair; it cannot claim a result for S1. A request for another record is warranted only if that unresolved use matters.
C.40.CD:5.1 develops this comparison. E.4.CM therefore selects reuse of that explanation rather than another pattern for these same operations. If a proposed card merely lists grouping and comparison, it should point to the available construction instead of asking the reader to invent the missing key.
Now change the receiving rule to require a shared instrument revision too. Temperature matching alone no longer answers eligibility. Matching revision identifiers can establish the new condition. A qualified record that the same instrument kept its revision throughout both observations can also establish it, even if the identifier is unknown. Evidence that the revisions differ rules out that pair. When both a match and a mismatch remain possible, leave its eligibility open and preserve any other pair already supported. This change illustrates what the public explanation must teach: derive the joining rule from the receiving condition and preserve partial results. It does not establish causality from one matched pair.
For an order to supply the eligible pair, the team can accept T3–T4 from a qualified supplier under the agreed rule. For an order to design comparisons for future jobs, the reader needs to obtain the receiving question and derive the matching conditions. Giving them a form headed “setup and temperature” supplies that particular choice. They can still apply the rule, check the entries or identify a missing revision. If the supplier's report documents who originally derived the rule and on what grounds, that evidence can also answer a question about the earlier design.
A larger contribution survives subtraction
Suppose an author has explanations for probing an unfamiliar device, comparing architecture alternatives and retaining a shared way of working. A practitioner still needs to turn an unexpected response into a repeatable delivery use.
The necessary connection is not “do all three.” A response supports only a bounded claim; the intended use determines whether that claim is sufficient. If it is insufficient, the next probe must distinguish the relevant alternatives. A chosen use then determines which preparation and environmental support to preserve, and a receiving participant's attempt can reopen the arrangement rather than the original causal question.
C.40.CU develops this whole. Its suppliers remain useful, but subtracting their individually supplied operations leaves the selection and return relations above. A normal FPF pattern is appropriate because those relations recur across constructs and uses.
The explanation connects several relations. A supported fixed-task response can permit use while an unfamiliar-task claim remains unknown and calls for another probe; these are conditioned continuations, not a compulsory inquiry sequence. Guide inspection and loading are constituent actions of delivery, and the delivery deadline constrains their combined performance. The tested configuration's success remains distinct from the controller's independently held capability. Each distinction changes the construction.
A short entry can say “take the response, bound the claim, construct a useful arrangement, and reproduce it under the receiving conditions.” That sentence is a reminder, not the complete explanation.
Integrate overlapping checks without losing either use
An intake Method accepts a reading when its unit is °C and its value lies from 0 through 100. A calibration-use Method repeats that same check and also requires a current calibration. The two Methods consume the same immutable reading version and use the same meanings of unit, endpoints and rejection. Under those conditions the common check can be performed once and its result supplied to both uses.
For the readings R1=(42, °C, current), R2=(42, °F, current), R3=(120, °C, current), R4=(60, °C, expired) and R5=(60, °C, current), the common check accepts R1, R4 and R5. Intake retains all three. Calibration use additionally tests calibration currentness, retaining R1 and R5. Sharing the check does not permit removing that additional condition or imposing it on intake.
Now the calibration use changes its upper limit to 50 °C. Use B.1.5.RS to follow the replacement through both receiving uses. Retain the shared 0–100 check for intake and add the narrower condition on the calibration branch; that branch now retains only R1. Changing the common threshold to 50 would incorrectly remove R4 and R5 from intake. Alternatively, keep separate checks when different observation versions, meanings or required failure responses prevent sharing their result.
The constructive choice is to identify the identical predicate and input, preserve each distinct obligation, and reconnect their results to the two receivers. It integrates the overlap without flattening the Methods into one stronger admission rule. No additional pattern is needed merely to give these checks a second name.
Apply the general studio construction to a framework publication
C.39:5.3 constructs an ordinary studio arrangement without requiring framework authorship: preparation supplies a corrected recording, analysis and viewing use it independently, and an unmet requirement returns the affected connection to method development. Now suppose an author is developing a movement-analysis DPF whose supplied methods already cover recording, editing, analysis and viewing, but whose public text leaves those joins to the reader. This is a constructed framework-authoring case, not a claim that such a DPF is already available.
Reuse the general construction. The author first follows C.39's account. The marker offsets 40, 42 and 45 ms give a constant correction of 42.5 ms and marker residuals −2.5, −0.5 and 2.5 ms. The independent engineer's bound of 40–45 ms throughout the interval, together with the qualified distortion-free editing operation, permits the interval conclusion. Both premises remain explicit. A number alone is not the corrected recording or its professional qualification.
Under that basis, one prepared recording meets analysis at 3 ms and viewing at 8 ms. At 2 ms, analysis fails for every constant correction of these offsets, while viewing retains its previous qualification. Preparation, analysis, viewing and method development remain distinguishable. Calculation both precedes editing and participates in encompassing preparation when it is performed; the account gives these different relations their own meanings.
Decide what the framework adds. The author compares a smaller alternative: links to the existing four method bodies. Those links allow an expert to reconstruct the joins but leave the interval basis, distinct receiving tolerances and selective return unexplained. A shared public account would supply that missing explanation. Its useful difference lies in those relations, not in a new primitive operation or a claim that the four Methods constitute one whole.
Prepare that account for publication in the DPF's Architectural Reference, with direct returns from the affected patterns and a short Readme entry. Keep the arithmetic and general change explanation with C.39:5.3/C.39.RO:5.1 and the professional operations with their qualified suppliers. Explain the domain interpretation and receiving conditions in the DPF.
For example, its practical instruction can say: “Obtain the recording's interval qualification; compare its bound separately with each receiving use; pass it to a use only when that use's tolerance is met; develop another preparation for an unmet requirement and retain qualified uses.” The earlier preparation explanation supplies how the qualification is obtained; this paragraph supplies how the receivers use it.
Apply the content functions rather than infer a Method from a template. The construction supplies much of the explanation; applying E.8 and E.11.PFP also exposes a domain-source question that it cannot answer:
To complete the SoTA function, obtain current domain sources for acquisition and synchronization, timing-error validation, and the movement-analysis requirements behind the receiving tolerances. Compare a serious alternative for the same receiving task, equipment and access constraints, and error measures; distinguish supported bounds from stipulated values.
Under E.8:11, state the best-known line, the alternative's action-changing defect or the deliberate trade-off, and each source's role and limits. Show how the comparison retains, narrows or changes the preparation, qualification or receiving rule, including the 2 ms return to development. Reopen if a changed source, requirement or observed failure defeats that choice. E.4.CM:11's method-engineering sources do not supply these domain claims.
The answers can use connected prose with references to the passages that supply each answer. They do not require twelve new headings in the Architectural Reference. If a separately recognizable recurring difficulty needs a new Method pattern, that pattern uses E.8's canonical body form and receives its own developed Solution and limits. Here the shared account and existing suppliers answer the missing-joins question; pattern creation is not forced by the number of Methods.
A short reminder can say: “Recover the receiving uses; construct and qualify preparation; apply each use's conditions; return the failed connection to development and preserve qualified branches.” In the resulting publication it returns to the full explanation, including the domain comparison and its limits. Reading the reminder in order requires neither development during every use nor a sequence between the independent receivers.
The ordinary studio practitioner obtained a workable conditional arrangement through C.39. The framework author has drafted its public explanation, chosen its maintained scope and placement, made direct uses reachable, and identified the missing domain SoTA comparison. The numerical argument and selective return remain useful under their stated premises. The DPF account is complete for publication only after that comparison is answered and its consequences incorporated.
Bias-Annotation
An expert author can forget how much synthesis they supplied between references. Compare the public inputs readers receive with the reasoning used in the worked case. Conversely, enthusiasm for a source can create a named umbrella with no added contribution. The same-situation comparison tests both errors.
Conformance Checklist
- The public account names a recognizable recurring difficulty, useful result and applicability conditions, and distinguishes one composite Method from several connected Methods.
- For every selected composition difficulty, give a constructive remedy or state what is missing.
- Connections explain meanings, dependencies and result use, including material branches and returns.
- Composition, result use and continuation, Method variation or refinement, and the use of methods and knowledge from different domains remain distinct; their consequential interactions are explained through the relevant structures. An orientation cue names its represented structure and relation.
- The selected objects and relations support the construction; source novelty or familiar component names do not settle it.
- Whole conditions change constituent actions where required; Method scale is not reduced to sequence order.
- Each reused description supplies the contribution claimed for it.
- The same-situation comparison preserves a useful remainder without copying complete suppliers.
- The public owner contains the explanation needed without private development material.
- The receiving case exposes the choices entrusted to the practitioner, the grounds that extend across the claimed conditions and any further use that needs evidence.
- Construction, fit for the receiving result and practical worth have proportionate grounds. Constituent, repertoire and whole-Method claims keep their own limits; necessary training, tools and support remain visible.
- The need and its basis, nature of the proposal, consequential development decisions, relied-on content and each asserted consequence are connected to their grounds. Evidence concerns the described arrangement and intended use; a missing ground is supplied, the construction is revised or the dependent claim is narrowed.
- A short entry preserves the work entrusted to its reader. It gives or leads to the grounds needed when the reader must act, develop the explanation or design the Method.
Common Anti-Patterns and How to Avoid Them
Consequences
Practitioners can use and adapt methods at several scales without reconstructing every joining rule. Constituent explanations remain reusable. Authors incur the cost of maintaining the whole's material relations and checking their affected uses when a supplier changes.
The method cannot make an unavailable operation, skill or resource obtainable by description alone. It can make that gap explicit and direct the required development or provision.
Rationale
Pattern authoring and scientific Method development often perform methodological synthesis: they construct ways of acting and the conditions under which their operations and results can be combined. Semantic synthesis can supply or revise the claims and distinctions used in that construction. The two results can develop together; neither the presence of action words nor publication in one document makes them the same result.
Composition is productive work because successful individual operations do not determine their joint conditions, suitable selection or response to change. The same need for construction arises among independently usable Methods. A new primitive operation is therefore neither necessary nor sufficient for a useful framework contribution.
Normal patterns can describe these solutions at the scale where practitioners need them. Publication length is a separate choice. Keeping a short first entry and a developed body avoids both a compulsory long first read and an unexplained whole.
SoTA-Echoing
Working question and selected answer. How should a framework author turn available methods into a public explanation that readers can combine and adapt without inventing the essential connections? The selected answer is the FPF construction in §§4.1–4.7: recover the receiving use, establish the participants and relations, develop missing joins, preserve independently useful branches, and test changed conditions while keeping each claim connected to its grounds. The resulting Method or explanation can be original even when its ingredients are known.
The method-engineering alternative constructs or configures a situational method from reusable units, with requirements and evaluation guiding adaptation. Wilke et al., Method Engineering — a Systematic Literature Review on Scopus Base, §§2–5, compares definitions and approaches including assembly and configuration; its Scopus search of August 2022 is bounded evidence of that repertoire, not proof of one universal ontology or the best answer today.
OMG Essence 2.0 beta 2, §§7.3–7.4, 9.3.2.12/.15 and 9.4, is a comparator: a common language and kernel support coherent Methods composed from Practices, recursive Practice composition, adaptation, and composition with conflict resolution. Its activities can be sequential, parallel or conditional.
Ralyté, Koutsopoulos and Stirna, Verification, validation, and evaluation of modeling methods: experiences and recommendations, §§2–4.1, 4.8 and 6.3, separates assessment purposes: construction correctness, fit to requirements and practical worth. It also covers problem and requirements analysis, evidence suited to development maturity, parts, method base, whole method and supporting provision. Its modeling-method cases and recommendations do not validate FPF or settle FPF participant kinds.
Cash, Daalhuizen and Hekkert, Evaluating the efficacy and effectiveness of design methods: A systematic review and assessment framework, §§2.1–2.5 and Table 1, supplies a different test: follow the evidence connecting a method's motivation, nature, development, content and claims. Its focus is design methods involving human activity or interaction; its study assesses reported evidence in selected design journals from 2020 (§3). It does not establish a universal method-development procedure or assess the whole method market (§§5.1–5.2).
The two treatments overlap in needs, content and outcomes. VVE distinguishes what an assessment seeks to establish; the evidence chain makes the grounds and alignment of the proposal inspectable. A positive trial can answer a bounded use question while leaving a need claim, development choice or wider effect unsupported. E.4.CM therefore combines the three assessment questions with explicit claim grounding in §4.6 and the checklist. §§4.1 and 4.3–4.4 develop the need and connections; §4.6 links their grounds to the conclusions drawn. This combination helps locate which premise or instruction must change when an assessment fails.
Why this construction is selected for the question. A direct route through existing descriptions is cheaper when it supplies the missing connection, as §5.1 shows. A common method-engineering model can be useful when the receiving project needs its compositional conventions and tooling. Neither a reading route nor transferring the source's unit model by itself establishes the receiving FPF composition, result-use and Work claims. This is a transfer limit, not a claim that Essence cannot describe the case or accept additional rules.
The studio case in §5.4 gives the bounded difference. The mathematical correction remains valid at the marks; the interval-wide warrant is separate. Tightening analysis from 3 ms to 2 ms defeats one receiving use, while 8 ms viewing remains available. The account identifies the connection to redevelop without making analysis and viewing parts of preparation, repeating unaffected work, or asserting empirical success. C.30.ASV:4.5a makes the relevant structure choices reusable. The selection concerns this conceptual construction. Claims about comparative effectiveness or learning cost need evidence from those uses.
The accepted cost is explaining and maintaining those consequential relations, with reusable detail left at its supplying owner. Reopen when a directly usable treatment supplies the same connections with less reconstruction, an applicable current rival changes the construction or evaluation question, a constituent or receiving condition changes, or an actual use exposes a missing rule, support need or moved burden. A better supported answer at lower total burden should replace this one.
Relations
- E.4 and E.4.PFAD place the contribution in the ecosystem when that architecture question is open.
- E.4.FPF and E.4.DPF govern the receiving framework kind; E.8 develops its pattern prose.
- F.0.2 develops semantic synthesis across source ontologies; A.7.1 repairs consequence-changing object distinctions.
- A.3.1 distinguishes Method identity, variation and directed refinement; A.22 and C.30.ASV:4.5a select the relevant structures and their representations; A.22.CGUS explains constrained continuations when that question is current.
- B.1.5 qualifies the whole's composition; B.1.5.EW and B.1.5.RS explain enactment and replacement.
- C.39:4 supplies the general construction, connection, explanation and selective change applied here; ordinary developers can use it without framework authorship. C.39.RO makes a working construction reusable. C.40.CD develops problems and ways together.
- C.32.MWA synthesizes an architecture account of Methods and their use when several structures must be related for a practical decision.
- C.40.CU is a worked transdisciplinary composition from uncertain response to reproducible use.
- E.11 selects useful short entries; E.11.PFP supplies the substantive questions, publication form and placement for accounts at their chosen scales.
- E.21 and E.19 supply quality and admission or refresh conclusions when those conclusions are sought.
E.4.CM:End
FPF Edition Assembly: Publication Forms, Carriers and Access Routes
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative.
Problem frame
Use this pattern when an FPF steward, framework author, reviewer, or AI agent must assemble, expose, or evaluate one FPF edition and distinguish it from its publication units, forms, presentation carriers, access routes, and domain or local dependents.
The primary EntityOfConcern is the scoped FirstPrinciplesFrameworkEdition. For assembly, prepare the record needed to rebuild the edition. For exposure, establish which edition the reader can actually obtain and use. For an evaluation, identify the edition under judgement and the grounds the commissioned conclusion requires.
The rebuildability record names the first-principles scope, selected Core patterns, recurring cross-domain problems and reusable solution moves, selected publication units, particular expressions and reusable forms, independently identified publication- or access-facing carriers with actual expression-bearing relations, access routes, edition relations, and the applicable whole-FPF quality result from E.2.DA.
Add a reference identifying the competing reading only when it passes the F.19:4 test: independent local ground, plausibility for the intended reader, a change in truth, understanding, or use, and the smallest clear correction.
Use this pattern when the work changes FPF publication units or forms—such as the public opening, Readme, Architectural Reference, ToC, first-entry view, cards, or split pattern collection—or assembles the exact physical or digital carrier whose actual state or realization yields their particular expression, such as an all-in-one file, site snapshot, volume, or bundle.
Use it also when a skill-pack carrier or an MCP, retrieval, search, or assistant access route exposes the edition. When a public presentation carrier is in scope, use E.11.PFP for its common reader-facing form and this pattern for FPF-specific source selection and assembly.
Use E.4.DPF when the framework is a domain or local framework grounded in FPF. Use E.4 when the live question is only family placement and routing among framework members.
If an edition needs new method content, use E.8:4.2.1.1 and :4.2.3 to write for the practitioner who must act. Use E.4.CM:4 when that practitioner needs a way to connect several contributions. An edition can faithfully reproduce a calculation whose author chose the model, while its intended reader still has to choose one for a new assignment. Follow that difference into the selected pattern before judging what its explanation supplies.
A reader can identify the FPF edition, the reader-facing units and forms that expose it, and the exact presentation carriers that bear them. The account also tells the reader how to reach the edition and where to evaluate whole-FPF adequacy. A DPF or local framework may depend on FPF Core while remaining its own framework edition.
Problem
DPF and local-framework packages can have detailed rules for publication units, forms, carriers, relations, naming, quality, and refresh while authors leave FPF's form unstated because they consider it "the main spec".
That creates several failures:
- an all-in-one file or site, one of its Readme, Architectural Reference, ToC, pattern-body, card, or first-entry units, a skill-pack bundle, or an MCP route is mistaken for the FPF edition itself;
- FPF is described as one more DPF, losing the first-principles and transdisciplinary burden that makes DPFs possible;
- whole-FPF quality is checked by local pattern scores, landing status, or DPF package scales instead of
E.2.DA; - adoption units, forms, or access surfaces grow user-facing explanations that quietly become shadow authority beside subject patterns;
- skill or MCP access makes FPF look like a callable service, tool permission layer, or runtime dependency rather than a framework edition reached through an access route or borne by an exact access-facing presentation carrier.
Use an FPF-specific form rule. FPF must keep first-principles distinctions usable across domains while allowing domain and local frameworks to grow from it; E.4.DPF governs those dependent frameworks.
Forces
Solution
Treat FPF as a FirstPrinciplesFrameworkEdition: one transdisciplinary edition with a selected Core pattern set and a stated first-principles scope. Record its recurring cross-domain problems, reusable solution moves, edition relations, publication units and forms, exact presentation carriers, and access routes under their direct subjects and relations. Coordinate them within that edition.
Units and forms expose selected content; presentation carriers bear the selected forms; access routes help an audience or system reach them.
Use these local names:
The progressive-minimum F.18 NameCard NC-FPF-EDITION-REBUILDABILITY-RECORD names the family of claim-bearing records defined by the FPFEditionRebuildabilityRecord row and declaration in E.4.FPF:4. That section is the record family's subject-pattern locator. This is an ordinary record family. Keep each under its direct kind: an individual record, its Markdown carrier, the assembly Method, the Work that enacts it, and an E.10 Map (a mapping U.Method).
The NameCard uses FPFCoreReferenceScheme by value. In that scheme, FPFEditionRebuildabilityRecord designates only the record family whose instances concern one FPF edition and name the exact sources, publication units and forms, presentation carriers, access routes, relations, projections, quality and currentness results, and refresh routes needed to reconstruct its public form. No Bridge is claimed.
Use the Tech designation in edition and rebuildability records, maintainer diagnostics, and direct consumers; use the Plain designation “record for rebuilding one FPF edition” in ordinary practitioner explanation.
The name comparison covers FPFEditionRebuildabilityRecord, FPFEditionAssemblyRecord, FPFEditionSourceAndCarrierIndex, and the predecessor FPFFormMap: rebuildability-record, assembly-record, index, and mapping-Method readings.
AssemblyRecord is too narrow because the record also names relations, projections, quality, currentness, and refresh inputs. Assembly follows. SourceAndCarrierIndex is too narrow because the record must also keep publication units, forms, and access routes distinct. FormMap is retired rather than kept as an alias because E.10 reserves Map for a mapping U.Method.
Reopen this choice if the named family becomes such a Method, ceases to concern one edition's reconstruction inputs and routes, FPFCoreReferenceScheme or the local-sense claim changes, a direct consumer needs another distinction, or a narrower admitted record kind covers every current field and use.
Current FPF practical-entry declaration. Apply E.11's content test to every proposed example. The current English FPF Readme declaration is:
Use this declaration as the sole list of selectable keys and their ordinary-entry or card forms when authoring, assembling, or validating the Readme. The Readme must return a reader whose question fits no example to the Table of Contents, another finding aid, or the direct patterns. Any additional coverage correction follows E.11.PFP:4.3.
Four related questions can be recovered without a separate selectable entry for each. PRACTICE-ARCHITECTURE and IMPROVEMENT serve CAPABILITY-DEVELOPMENT. OPTION-COMPARISON serves COSTLY-ACTION, and WORKING-DOCUMENTS serves DESCRIPTION-USE. For DPF-AUTHORING, use E.4.DPF to develop the explanations the domain practitioner needs. The SOTA-PORTFOLIO entry contributes a comparison when that work depends on choosing among source traditions.
UNFAMILIAR-THEORY connects recovering a construction and argument with transfer to the project's representation. PHYSICAL-RESULT connects physical modeling, mathematics, computation and realization, including returns when one contribution changes. Their mantras retain those dependencies while the direct patterns supply the operations.
WORK-OPPORTUNITY starts at the first missing contribution and can end at a local result, an exact gap or continued work without advice. Follow each linked pattern when its condition arises. The route does not require all of them. The Readme presents WORK-OPPORTUNITY as an ordinary entry.
TIME, CAUSAL-USE, MEASUREMENT, and MATHEMATICAL-MODELING are ordinary examples because each starts with one direct pattern and can stop at its first useful result without a cross-pattern mantra. No MODELING-FOR-ACTION card joins them.
LIVE-WORK-STEERING and METHOD-RECOVERY are ordinary examples for the same reason: each begins at one direct pattern and may stop at its first useful result or blocker.
PROFESSIONAL-RESULT is also ordinary: it starts with A.15.9, tests an already-available result before any new request, and can stop at bounded reuse, the smallest missing-result request, or a blocker without a cross-pattern mantra.
COMMUNICATION-FOR-USE starts with the communication and its intended use. Its ordinary explanation distinguishes interpretation, response, later action and effect, then opens a wording, representation, evidence, causal or repair question only when that question is needed. The reader can stop at a supported use or an explicit blocker.
RESULT-TO-NEXT-MOVE starts with the obtained result and its direct pattern. Its ordinary explanation preserves the conditions for a later interpretation, reliance, characterization, comparison or live-choice question and stops before taking up a question that is not current for the result or its receiving use.
CONSEQUENCE-BEARERS starts with the bounded consequence account under A.1.CSD. Its ordinary explanation retains the boundary challenge, conditional systemhood check, separate obtaining and modal paths, each bearer's changes and uncertainty, and the return when the receiving use changes.
ACTUAL-TEMPORAL-STRUCTURE starts with what the temporal claim concerns. Its ordinary explanation distinguishes actual subjects and obtaining relations, a selected structure and its account, and future specifications or representations. A coordination trial follows only when it can change the decision and warrants its burden; otherwise the reader stops at the supported answer or missing basis.
PUBLICATION-FORM and DPF-SUITE-REFERENCE point readers to E.11.PFP and E.11.DSG without adding selected examples. Those patterns supply the detailed content. The Readme carries only the recognition, cross-pattern dependency, and return needed for discoverability.
For every card row, keep the card and its practical-use guidance findable by the same key. The current English FPF counts whitespace-separated tokens and requires at most 80 for a card mantra and 220 for a complete compact card. Neither limit sets a minimum.
These token ceilings are calibrated publication limits for this English FPF, not psychometric thresholds or universal DPF, LPF, or translated-publication limits.
Reopen the smallest affected declaration row and consumer when a cold-reader replay loses a choice-changing distinction, a useful card cannot fit without copying direct-pattern apparatus, or either ceiling can be lowered without losing use value. The optional @FPFReadme support records may carry FPF links but do not define shared conformance.
For a commission to assemble or republish an FPF edition, create the rebuildability record below. When exposing or evaluating an edition that already exists, use its record for the identity, lineage or expression actually needed for the commissioned result.
firstPrinciplesFrameworkEditionRef resolves to the edition record for the selected FPF edition; relationAndEditionRefs resolves that edition's status and dependency assertions. Keep DPF and LPF package records outside this record. The rebuildability record supplies reconstruction inputs. Base downstream assembly and use claims on their direct results and relations.
Use E.24.PUB to distinguish each expression from the reusable form whose rules it follows. Recovering the expression from the independently identified carrier supplies the basis for the expression-bearing claim. A continuing file, an immutable snapshot value and an individual copy have different identity rules. A route or filename alone settles none of them.
A form-mediated expression or bearing claim retains its existential expression witness, and combined availability requires the same witness.
For assembly or republication, use the steps below to produce the edition and its rebuildability record. For access to an existing edition or an evaluation, take the identity, availability and other facts the commissioned use requires from the edition and its records. Carry out a rebuild when the commission or a missing result requires it.
-
Identify the FPF edition or candidate by its stable designation and the record for that edition.
-
State the first-principles scope: FPF supplies transdisciplinary distinctions that can be reused across domains. Domain doctrines remain with their DPFs; the declared scope sets the useful coverage boundary.
-
Identify the selected Core pattern set and any companion or projection loci that expose it.
-
When a public presentation carrier is being assembled or checked, use
E.11.PFPfor the common publication form. Keep a product-declared compact opening and separate exact title and Readme H1 values.Represent Readme and Architectural Reference in the product's established ToC grammar before one logical pattern index. Keep one selected practical-entry set with E.11.PFP:4.3’s guidance for keeping the reader’s question through the choice of entry, five-field ordinary examples, six-field selected cards, and one integrated source-hazard plus rendered-structure check.
Apply
E.11's use test before assigning card form, and use the current FPF declaration above for every selected key and form plus the English compactness measure and two limits. Add another public cue only when a named reader decision or action needs it.For the established all-in-one FPF carrier, add Readme through the same non-pattern table grammar already used for Architectural Reference, preserve the compact pre-ToC shape, and keep the exact line-position and native-ToC assertions in the builder regression. Keep FPF-specific source selection, body order, and assembly here; the carrier and form remain separate from the FPF edition.
-
Readme, Architectural Reference, ToC, the public opening, cards, and the pattern collection are publication units. Reusable rules for arranging them supply the publication form. Identify the particular expression of the selected edition that follows those rules.
Name the physical or digital carrier—for example, a managed Markdown file, site snapshot, PDF volume, split-file bundle, skill-pack bundle, index file, or response document—under its own domain identity rule and establish that its actual state or realization yields that expression.
Record an MCP service, retrieval route, search function, or assistant integration as an access route. Name any returned carrier separately. Record these units, expressions, forms, carriers, and routes as distinct subjects and relations for the
FirstPrinciplesFrameworkEdition. -
State relation, dependency, edition, deprecation, supersession, publication, and access claims directly. Open an
E.4.PFRrow only for a named maintenance consumer. -
DPFs and local practice frameworks may depend on FPF Core. Incorporate an accepted transdisciplinary contribution through a deliberate Core amendment before Core uses it. The amended Core supplies that contribution without depending on the DPF or local framework.
-
Complete the
FPFEditionRebuildabilityRecordwith exact selected source, publication-unit, publication-expression, publication-form, presentation-carrier, access-route, relation, practical-entry declaration, currentness, and refresh references. Use that one declaration of keys and card assignments for Readme assembly and its checks. Do not add a competing manifest or duplicate the rebuildability account. -
Assemble the all-in-one edition candidate from the exact predecessor, the selected edition record, the matching
FPFEditionRebuildabilityRecord, and every selected complete pattern source. Give each replacement or insertion an explicit boundary. Derive the logical index and pattern bodies from the same selection, verify one index row per selected PatternID, report which source supplied each assembled unit, and verify that every unselected predecessor span is unchanged.A missing or duplicate record, unresolved ref, index/body mismatch, ambiguous boundary, source mismatch, or changed unselected span stops construction.
The construction result reports the assembled candidate and source correspondence. Acceptance and publication require their separate decisions and relations. Keep repository paths, commands, helper options, template names, and insertion syntax in maintainer documentation or the selected tool's help.
-
When the assembled publication claims accepted-source integration or continuity with its predecessor, use
E.4.PFIPfor that comparison.For whole-FPF adequacy, use
E.2.DAover the scoped FPF object and declared use. UseE.21for individual pattern bodies,E.9.DAfor a DRR, andE.4.DPF.DAonly for DPF or local-framework packages. -
For first entry and reader-facing exposure, use
E.11andE.17. Keep the Readme and ToC concise enough for entry. Develop the Architectural Reference enough to explain the whole and the connections needed to select, combine and adapt its contributions. Definitions and separately selectable Methods retain their subject-pattern owners. -
In the FPF Readme, Architectural Reference, and ToC, state the reader and use they serve, which first-principles structures they foreground, what they deliberately coarsen, abstract, omit, or defer, and where the reader returns for subject-pattern detail.
Use
E.11.PFPfor the common publication-form structure. Preserve the product-declared compact opening, put the direct Readme/Architectural Reference route before the logical pattern index, and keep source paths, digests, machine identity blocks, candidate records, and build instructions outside reader front matter. -
For source-front, currentness, and refresh claims, use the direct source and currentness assertions under
G.2andG.11. Publication units, forms, presentation carriers, and access routes contribute only their stated publication or access facts. -
For skill packs or MCP-backed access, expose edition identity, dependency boundary, and currentness or refusal conditions. Distinguish the exact skill-pack, index, or response carrier from the service or route that returns it.
Use
C.35for generated candidate text intended for architecture use. Keep tool capability and Work claims separate, using the applicable tool pattern for the former andA.15plus the pattern for the exact Work for the latter. Use the applicable patterns for assurance, evidence, and decision-authority claims.
Choose the pattern for your question:
Archetypal Grounding
Edition revision. A later FPF edition updates its public front, Readme, Architectural Reference, ToC, and several pattern bodies. The FPF edition rebuildability record names one scoped edition, lists its Core pattern set, publication units and forms, exact presentation carriers, and access routes, records which entry text is only a projection, and points to E.2.DA for whole-FPF adequacy. It records Readme and Architectural Reference as publication units. Naming a presentation carrier requires the PublicationExpressionBearingRelation to its expression. When that expression conforms to the selected form, it also witnesses PublicationFormBearingRelation.
Domain framework. A domain principle framework for one practice depends on FPF Core and may cite architecture, representation, precision, and improvement patterns from FPF. Its publication units, forms, exact presentation carriers, and access routes belong to that DPF edition. Use E.4.DPF.DA to assess that package. FPF Core retains its transdisciplinary scope and its own amendment route.
Skill-pack access. An FPF skill pack exposes pattern lookup, first-entry guidance, and short-use prompts. A versioned skill-pack bundle can supply an access-facing carrier when its independently identified actual state or realization yields the selected expression. The service, endpoint, or assistant integration that returns it is an access route. Their descriptions name the FPF edition they expose and its refresh condition. Authority claims return to the named subject pattern and edition.
Rebuild an edition and catch a boundary error
A steward is preparing Cedar 2, a fictional FPF edition for engineers who start from a working question. Cedar 1 is its frozen predecessor. The example follows three changed bodies inside the full edition; all other bodies must carry forward unchanged.
The accepted complete sources are DUA-2 for C.11.DUA, PUR-2 for E.11.PUR, and Ways-1 for C.39. DUA-2 replaces DUA-1 and includes continuing without optional advice. PUR-2 replaces PUR-1 and distinguishes applicability from recommendation. Ways-1 adds the method-development contribution between the existing C.38 and C.40 bodies.
Readme R2 adds their practical entry. Architectural Reference P1 remains unchanged.
The first useful result is rebuildability record RC2, which gives the assembler the following inputs and boundaries:
Each carrier's actual state yields its named expression. Each expression follows the corresponding reusable arrangement rules and expresses the selected content. Neither file is the edition itself. The standalone HTML expresses R2 without becoming another independently authored Readme. No access service has been selected.
Assembly replaces the two bounded bodies, inserts Ways-1, and derives I2 from that same membership. I2 must contain one row each for C.11.DUA, E.11.PUR and C.39, with no duplicate C.39 row. C.39 is the only addition to the body inventory. All other membership and order stay as in Cedar 1.
A first candidate has exact copies of DUA-2, PUR-2 and Ways-1 and the expected index, yet fails the predecessor comparison. The Readme replacement ended two sentences inside P1 instead of at its boundary. The lost predecessor text was: “Start from your working question. Stop when the answer is sufficient; the entry list is not a required sequence.” The candidate has lost useful stopping guidance although every selected body matches its source.
The complete predecessor comparison exposes that unselected deletion. Stop the assembly at this failed preservation condition. Correct the Readme boundary and restore both sentences from frozen P1. Do not rewrite the accepted pattern bodies. Rebuild and repeat source correspondence, the full inventory and the complete publication comparison.
In the repaired candidate, DUA-2, PUR-2 and Ways-1 match their selected sources, I2 and K2 agree, R2 appears in both selected carriers, and P1 and every other unselected predecessor span are unchanged. The remaining differences are the two declared body replacements, one insertion and index row, R2 and its opening references. That is a completed construction with a stated preservation result under E.4.PFIP. Acceptance, publication and whole-FPF adequacy require their own conclusions.
Bias-Annotation
Scope: This pattern covers the publication units and forms, assembly, exact presentation carriers, access routes, and whole-framework evaluation route of an FPF edition. Repository layout, assembly tool, carrier split, and publication service remain implementation choices. A one-sentence repair that leaves these FPF-level values unchanged uses its direct wording and source route.
Conformance checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Authors can reproduce an FPF publication by building its Readme, Architectural Reference, ToC and pattern-body units, arranging them in all-in-one or split forms, and publishing them through identified files, sites, volumes or bundles. Skill, MCP, search or retrieval access can expose that publication while keeping edition, publication and access claims distinct. The same declaration keeps FPF's few direct and cross-pattern examples, their forms, and the two length limits aligned across authoring, assembly, and validation. The selected examples show how to use the pattern language, while the unmatched-question return keeps further discovery available.
Stewards face an extra distinction: whole-FPF form differs from DPF authoring, package adequacy, individual pattern quality, and first-entry publication. The extra work is justified because those problems require different evidence and can fail in different ways.
Rationale
FPF carries transdisciplinary first-principles distinctions that can seed and discipline many domain and local frameworks. Those frameworks supply their domain or local subjects and practices.
These cross-domain uses make FPF form an architecture concern. Separate publication units, forms, exact presentation carriers, access routes, and the framework edition so adoption work returns authority to the subject patterns. E.4.DPF.DA answers the DPF package question. Use E.21 for individual patterns and E.2.DA for whole-FPF adequacy.
SoTA-Echoing
The current practical-entry declaration and the internal FPF quality, dependency, carrier, and access-route rules are governed by the Solution, checks, and Relations. These internal rules do not supply external SoTA evidence about this pattern.
Relations
- Specializes:
E.4for work on the form of FPF. - Builds on:
E.2Pillars throughE.2.DAfor whole-FPF adequacy. - Coordinates with:
E.4.PFRwhen a named maintenance consumer needs a reusable relation, edition, dependency, publication, access, deprecation, or supersession record. Otherwise state the relation. - Coordinates with:
E.4.DPFandE.4.DPF.DAas sibling patterns for domain and local frameworks, not as FPF-level substitutes. - Coordinates with:
E.11.PFPfor the common framework publication form and withE.11,E.17for first entry, projection, and publication or access use. - Coordinates with:
G.2,G.11,C.33,C.34, andC.35for source, currentness, structural preservation, and admission of generated or discovered results for architecture use. - Coordinates with:
E.21,E.22,E.23, andE.9.DAwhen individual pattern quality, evaluation framing, improvement loops, or DRR adequacy provide evidence for FPF-level changes.
E.4.FPF:End
Principle-Framework Architecture Decision
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative.
Problem frame
Use this pattern when an author is choosing among a new or revised principle framework, a contribution to an existing framework, another product that is not a framework, a thinner publication or access route, and no new maintained product now. The choice must settle an identity, edition, relation, intended-use, or publication decision that later work must use.
The decision may concern the public field and first use, framework edition, dependencies, initial pattern placement or relations, the kind and identity or change rule of a non-framework product, or the publication or access consequence. Use it when another author or reviewer will need the answer and its rationale to act.
E.4.DPF may return this question before a public product name, PatternIDs, or pattern bodies exist. Open it when the preliminary contribution comparison leaves ownership unresolved, exposes a connected remainder, or identifies a material field, relation, source or refresh, publication, or access consequence that later work must use.
Treat an incoming carrier or organizing cue—for example, a role account, MethodDescription, file, Card, mantra, shared topic, or missing authoring artefact—as evidence for that comparison. Decide from what each candidate contributes and the consequence on which later work depends.
Here product has the Plain management meaning declared in E.4:4.1. It names no technical kind. When a non-framework product is selected, name the direct subject of each claim and its kind, and the identity, current-state, provision, publication, availability, or other relations used by the decision. Name a maintenance relation only when that stronger claim separately obtains and changes the answer. If a kind or relation that can change the answer is unresolved, state the unresolved decision question and do not invent U.Product.
A proposed new or substantially revised DPF also needs an answer about its field boundary. That answer says who can first use the framework and for what, which connected problem families and useful results it covers, what the current FPF and admitted DPFs already provide, and what remains uncovered. It compares serious alternatives, tests one representative case that crosses problem families, states where the evidence runs out, and names the change that will require a refresh.
The field account and comparison must support one independently usable pattern language. One pattern or a narrow authoring slice is not a DPF merely because it has a broad title or a coherent carrier.
If a cheap search, curated reading route, useful contribution to an existing framework, suitable non-framework product, or stop answers the immediate need without settling such a boundary, use that result and stop. Reserve the framework-architecture DRR for a later-used boundary.
When the architecture question is live, use E.4.PFAD to state the framework-specific content of one ordinary E.9 DRR. The decision-maker selects the answer during decision Work. The DRR records it. Practitioners use this profile to find the framework questions and answer content in the DRR, while acceptance remains a separate decision.
Problem
Framework authors repeatedly need to decide whether a recurring practitioner problem calls for a new framework, an existing framework contribution, another product such as a programme, service, or evidence package, a thinner access result, or no new maintained product now.
When a framework is selected, later work needs its public field promise, first-edition boundary, problem-family coverage, first patterns and their material relations, representative use, important omissions, and publication or access consequence. Framework authors also need the selected FPF edition, the Core claims required for each receiving use, and any other framework-edition dependencies. Generic decision prose can hide those choices.
Authors commonly mistake a small, coherent set under development for a field-scale pattern language because it already has a few patterns and an orderly structure.
Source diagrams can also mislead authors. They may copy one list or hierarchy into the DPF although Methods, Work, subjects, descriptions, capabilities, providers, and cultural change may have different structures.
A large framework-specific form creates the opposite problem by making proposal, acceptance, DRR, edition, authoring, quality review, and publication look like one extra decision object.
The useful result is one readable answer that explains its consequences and limits for the framework. A second decision stage is unnecessary, and exploratory work can stop without a decision record.
Forces
Solution
Decide whether the architecture question is open
Ask whether choosing a framework, a non-framework product, a thinner route, an existing-framework contribution, or stop will settle at least one of the following matters for later authoring or review:
-
the public field promise, a first use that does not depend on unpublished authoring context, or the problem-family coverage of a proposed DPF;
-
an intended or existing framework edition;
-
a dependency on a selected FPF edition for named Core claims and a receiving use, or on another identified framework edition;
-
initial pattern placement or a material relation among those patterns that changes the architecture;
-
the direct subjects of the decision's claims and the identity or change rules for a continuing programme, an admitted service, or a separate editioned result, plus any maintenance relation when later work separately claims or uses it;
-
a publication or access consequence; or
-
for a proposed DPF Suite, the composite discipline, its constituent subdisciplines and whole-forming relations, the distribution of Methods between Core, constituent DPFs and external suppliers, and the framework assembly that describes those distinctions.
When E.4.DPF supplies a preliminary contribution account, compare the contributions it identifies. If same-situation comparison shows that exact current FPF or admitted-DPF owners carry every action, first result, return, and source or refresh duty—or exact external results supply them—and no decision above remains, take the smallest useful result and close without PFAD.
For connected Methods, establish the supplied whole use through E.4.CM:4.4 before treating the constituent suppliers as a complete answer.
Keep the architecture question open when exact ownership remains unresolved, a coherent action-bearing remainder survives, or closure would erase a material relation or field or refresh responsibility that changes a later use. Use carrier form, role label, shared topic, names, PatternID status, and counts to locate candidate contributions. The four dispositions and later-used consequence determine whether to close or select an outcome.
If none of these decisions and no receiving use are present, take the exploratory result as the answer. If a decision for later use is present, the decision-maker selects a framework, non-framework product, thinner route, existing-framework contribution, or stop during decision Work. One E.9 DRR records that answer. The cheap exit and the architecture decision are alternative entry outcomes.
Start a comparison of products from the work the recipient needs. Establish the direct subject of each alternative, such as a service arrangement, a Method, a framework or the carrier a person must receive. Hold the receiving question and conditions fixed. For each alternative, retain the distinctions needed to establish its result and the choices the recipient would still have to make. Where the alternatives concern different kinds of thing or composition scales, show how each bears on the same decision.
Use a quality-management, service-management, publication, or content-management scheme to test a possible distinction. Use the FPF pattern for the direct subject to settle its kind. If an unresolved kind can change the selected answer, keep the product proposed and make that kind the next decision question.
Choose a contribution through attainable use and its whole cost
Use this construction when the architecture choice turns on who can benefit from the proposed content, how that benefit becomes available, or which uses must change together. Use the comparison as grounds for the answer in :4.2. The comparison adds no record and requires no study before a small repair. If one adequate route already answers the question without a consequential boundary choice, keep the :4.1 cheap exit.
Start with a useful occasion, not a catalogue of possible mistakes. Describe a recurring or credibly anticipated situation, the result worth obtaining, the presently available way of obtaining it, and the consequential difficulty. Distinguish frequency of the occasion, frequency of failure conditional on that occasion, severity of its consequence, and the proposed remedy's effect.
A frequently encountered task can justify assistance even when mistakes are rare. A rare severe failure can justify protection without being called frequent. Source occurrence is evidence that a method was described, not that the intended population needs another description.
State which claims are observed, source-supported, estimated or still hypotheses. Use C.11.DUA to decide which uncertainty is worth investigating.
Describe the actual reader and performer arrangement. Separate general and domain education, skill in the relevant work, familiarity with FPF/DPF distinctions, and the means available on this occasion. Preparation can differ for recognizing, performing, checking, adapting and teaching the same Method. Use only differences that could change the arrangement. Do not classify people or models by a single rank of intelligence.
These are common arrangements, not an exhaustive role taxonomy or three mandatory editions. A learner can later need only a reminder. An experienced practitioner can need the full reasoning when conditions change.
If the intended result is the person's independent judgement, keep that result in the comparison instead of substituting a good answer delivered by the tool. If the intended result is obtaining the answer from a provider, do not require every recipient to learn the provider's Method.
Separately name the subject whose state, relations or capability the Method is intended to change, and any distinct recipient of its result. Changing a teacher from a human to an AI system can leave the human learner and intended result unchanged. Replacing that learner with a robot changes the targeted participant and may require a different domain Method.
Use A.3.1:4.7–4.8 to decide what remains the same, and F.0.2.MA to retain a justified common construction and each action-changing domain contribution. A common interface or result label does not settle Method identity or discipline membership. Return the membership question to :4.2 and C.20.
Follow the route by which the benefit could occur. Begin with the work question the user recognizes, without supplying a PatternID the user would have to discover. Follow discovery, recovery of the task-relevant distinctions and connections, selection of a suitable action, performance of the Method or obtaining the required result from a capable provider, and the receiving use of that result. E.11 and A.15.11 supply discovery and occasion; C.2.8 makes recovery relative to preparation, accessible material and effort.
Find the first necessary contribution that the proposed arrangement cannot supply. It can be a missing method connection, inaccessible explanation, unacquired skill, unavailable assistance, or a result that does not serve the intended work. These require different repairs.
If the contribution cannot be found or its necessary structure cannot be recovered in this arrangement, the promised benefit through that route is unavailable. Search and reading can still consume resources. This is a causal constraint on value, not a conversion of structural information into money. More extractable structure can also be irrelevant to the task. Distinguish an observed failed attempt, a reasoned walkthrough and an untested access assumption. Unknown performance is not a measured zero.
Construct serious alternatives for the same result. Compare the available public method and no addition with the consequential alternatives: repairing discovery; a reminder or checklist for an already possessed capability; a developed pattern or composite; a pedagogical companion; external competent provision; or distinct variants with a small shared supplier.
A checklist can prompt a judgement without teaching how to make it. A companion can supply explanation, practice and feedback without creating a different subject Method. E.8:4.2.1.1 assigns their respective content and tests whether teaching can be constructed without inventing the Method. A short route and a full explanation can belong to one architecture.
For one composite Method or connected independently usable Methods, including alternatives, compare the whole receiving use and its necessary connections under E.4.CM:4.4, not only the available constituents.
For an unprepared audience, include learning or external help in the comparison, with its costs and availability. An assumed expert reader hides those requirements. Conversely, do not force experts through elementary instruction that changes no relevant result. A result beyond the present capability and budget can warrant a different allocation, staged development or a use that remains unmet.
Compare the whole cost of obtaining and sustaining useful application. Use a stated horizon and plausible number of users and occasions. Include only terms that can change the decision, but include them where they fall: developing and maintaining the contribution; finding and interpreting it; learning, practice and feedback; obtaining help; adapting and integrating it into work; recurring application, checking and recovery; and responding to consequential changes.
For AI-supported work, computation and retrieval are costs alongside human prompting, checking, waiting and correction. Include displaced work and the cost of retaining a capability or dependence when those matter. Using a free document consumes resources, including the time to find, learn and apply its contents.
C.11.CRC compares complete configurations rather than summing isolated pattern merits. C.11.DUA compares added work with attainable gain. C.19.2 limits how much decision apparatus is worth applying.
A useful ordinary calculation is one-time acquisition and setup, plus recurring cost over the horizon, plus consequential change and retirement costs. Count shared development once at the chosen population boundary. State any allocation to individuals. Do not charge an author's investment in full to every reader, count the same correction both as expected loss and paid rework, or amortize over an invented infinite audience.
Human time, compute, money, delay and serious non-compensable losses need not collapse into one score. Use ranges or break-even conditions when a point estimate would pretend to knowledge.
Compare acceptable result quality and constraints together with these costs. Lower cost for a different or inadequate result is not a saving. Past learning is part of the available starting capability. Further learning can create benefits beyond this one application. State those future uses as supported expectations or hypotheses.
Identify who pays and who benefits: an aggregate saving can leave the intended user unable to afford adoption. A rare but unacceptable hazard is handled under its governing obligation. A small average cost cannot justify trading it away.
Challenge sharing by an independent change. Vary one premise that matters to one intended use or audience: for example, a domain criterion, teaching prerequisite, available tool, or required outcome. Determine which explanations, interfaces, dependent uses and release obligations would need to change.
Common ideas, reused Method descriptions and a common running implementation create different dependencies. Separate files alone do not create independent change. A shared carrier alone does not prove shared semantics.
Compare a shared supplier with a bounded interface, independent variants, and limited deliberate duplication. Retain sharing where it saves enough construction and maintenance while preserving useful independence. Split where accumulating exceptions, migration, compatibility or coordinated releases defeats that gain. Use the C.31 modularity and reuse characteristics that answer this question. Neither maximum reuse nor every coordinate is required. E.4.PFR handles edition dependence when that relation obtains.
Spend the next effort where it could change the choice. Try the smallest consequential uncertainty: can the intended reader find the contribution; can the intended reader recover a necessary connection from the available public explanation; can a user complete a supported case; does a changed premise force unrelated variants to change?
If the reader must invent a consequential connection, retain that construction as a proposed contribution and compare its useful effect and cost. The private invention does not establish that the public explanation supplies the connection. A prospective benefit is a hypothesis that can justify a bounded experiment, not a realized return.
Separate a content-recovery probe from an effectiveness or learning trial. In a teaching claim, assisted performance and later independent performance can require different observations. Stop expanding the analysis when the available distinction already selects the next action, or when further inquiry costs more than the decision it could improve.
Return the selected contribution and arrangement, the useful result, preparation and support on which it relies, material costs and uncertainty, the chosen sharing boundary, and the observation that could reverse the decision. Use these grounds in the :4.2 answer. No separate form is required.
State the compact framework answer
When the architecture question is open, the framework-specific part of the DRR states:
-
the intended practitioner, public field name and promise, recurring problem, and bounded architecture question;
-
the selected outcome: a new or revised framework edition, a contribution to an existing framework, a non-framework product, a thinner publication or access route, or no new maintained product now. For a non-framework product, also name the kind of each direct subject of the answer and the identity, current-state, provision, publication, availability, or maintenance relations used by the decision;
-
its field boundary: who can first use it without unpublished authoring context and for what; the connected problem families and useful results; what the current FPF and admitted DPFs already provide and what remains uncovered; alternatives that can change this boundary; the limits of evidence; and what change will require a refresh;
-
the selected problem-family pattern sets, first patterns and their material relations, representative cross-problem application, and important omissions;
-
which structures of Methods, their use, support and development change the answer, and how the identified Methods, descriptions, patterns, direct subjects and managed result boundaries fit together. When those structures do not line up one-for-one, use a completed
C.32.MWAsynthesis; useE.23.CDIonly when capability development for a named Work family changes the answer; -
the existing or intended framework-edition boundary; the selected FPF edition, the Core claims this framework relies on and the receiving use that requires each claim; and any other framework-edition dependencies required by this answer, under
E.4.PFR; -
the sources to revisit for each important claim, whether the evidence supports, suggests, or only motivates it, the limits of that evidence, and the publication or access consequence; and
-
material alternatives, accepted costs or losses, practical consequences, the first authoring action or stop, and the condition for reopening the decision. For each framework-edition dependency, distinguish a change of the relied-on edition from a material change to a relied-on claim, and identify the dependent use that each reopens.
The field-boundary comparison considers applicable alternatives: retaining, splitting or merging existing and proposed frameworks; moving or extracting existing Methods into a new or revised DPF; and developing a Suite when C.20 establishes the fields as constituent subdisciplines of one composite discipline.
Also consider using existing sources directly, contributing to one framework, composing exact contributions across several DPFs and FPF, selecting a programme, service or separate evidence-package episteme, and selecting no new maintained product now.
For any opened question, record only candidate contributions that can change the outcome. For each, state whether an exact current FPF or admitted-DPF owner carries its action, first result, return, and source or refresh duty; an exact external result supplies it; an action-bearing remainder survives; or ownership remains unresolved. A shared topic or broad framework name can help find candidates. Assign a disposition only after comparing what each contributes to the same situation.
Then settle any field or organization question that the comparison exposed. Finding a Method in an existing DPF establishes its availability, not that its present placement is the best boundary for the receiving practice.
Reconstruct the candidate field from its characteristic problems, Methods and knowledge, material relations, intended practitioners and source/change responsibilities. C.20 supplies discipline-composition conditions when a discipline claim matters. A job title or conference is evidence about a field, not a substitute for its constructive account.
Compare the present distribution with plausible regrouping, including the already described Methods. Explain what each choice changes in obtaining, learning, adapting and maintaining the whole. Retention needs the same substantive grounds as a split or new framework. Count migration costs, but avoiding a new name or file does not establish adequacy.
A residual comparison can therefore return one missing operation, a new composition, a whole field or several fields. Conversely, an available field framework may already supply the required language, including its relations, so that a bounded extension or better entry is enough. Keep these outcomes open until the comparison settles them.
A pattern whose internal sections address several separately usable problem/result families is a candidate for repartition, even if its old identifier is stable. Length alone establishes neither that defect nor the right split.
When professional Method coverage can change point 5, the same compact framework answer projects five connected claim groups from points 1, 3, 4, 5, 7, and 8. Keep these claims in that answer and its one ordinary E.9 DRR. Fill each group to the grain that changes first use, using a representation suited to the case, before DPF authoring:
- Practice claims and first use: identify every bounded practice claim or promised contribution by its subject and scope. Distinguish a claim about existing practice from a proposed practice, and state what is supported, refuted or unresolved. Name the practitioner, recurring or anticipated difficulty, sought result, first use, stop or wrong-turn return, qualification window, and receiving decision. Include only non-use boundaries admitted by
F.19's grounded-contribution test. - Project and Method positions: name the direct project subjects, use and environment, materially different solution forms, and Methods under their actual operational, system-change, solution, Method-of-interest, or Method-development relations. Keep incumbent Work, development or trial Work, candidate-practice Work, and intended Work distinct.
- Selected structures and correspondences: include only the Method, Work, subject, transformation-flow, capability/provider, description, contribution, Method-development, and cultural structures whose correspondence, conflict, or non-isomorphism changes the answer.
- Pressures and evidence: keep constraints, conflicts, failures, environment or interest changes, and observed, source-supported, estimated, contradicted, and missing links distinct from causal history and temporal unfolding.
- Contribution, subtraction, gaps, and reopen: apply the four dispositions above, then name each receiving pattern and domain filling still needed, omissions and gaps, and the observation that reopens the architecture.
Within an existing-framework contribution or another answer that selects no new framework, distinguish one existing DPF receiving a contribution from a use that composes exact contributions supplied by several DPFs and FPF. Record the latter as cross-DPF composition under the selected one of the five outcomes, optionally exposed through a role-centred view or access route. Decide Suite membership separately.
One answer may contain several bounded practice claims with different subjects and support. Every selected question names the claims it consumes. A supported incumbent-practice claim can coexist with an unresolved claim about another existing practice and with a candidate-practice proposal.
Independently obtaining A.13 agency claims and actual development or trial Work retain their own status. They establish candidate-practice coverage only when the required relation and evidence support it. Public coverage remains a separate claim, limited to the supported contribution and package evaluation.
For an obtaining practice claim, name actual recurring difficulties and representative actual Work. A precise Agent-performer branch first supplies A.13's core: the exact admitted System, local agential system-role kind and criterion, classification, obtaining assignment, and needed scope, working situation, and window. Add the agency-characteristic profile only when a Grade, autonomy or profile claim, a criterion-dependent characteristic, or a named assurance use consumes it.
A.15.1 then independently admits the Work from its performance history, Method, extent, and containment. Only after admission does F.6 supply any precise assignment-bound attribution through that same obtaining assignment. State evidence limits. A missing F.6 relation leaves admitted Work intact and only the attribution unresolved.
For a proposed practice, recover intended use, any incumbent Work or Method and available problem evidence, candidate Methods and architecture, and realization conditions. State what the construction, sources, derivation or earlier validation support for that use and which material questions remain open.
Plan a representative trial when a binding requirement demands it, or when needed evidence is missing and the attainable result justifies the trial's full burden. When a trial is warranted, name its acceptance, failure and reopen conditions. Otherwise retain the sufficient existing basis or state the exact unresolved claim.
Incumbent-practice and actual trial-Work claims remain independently grounded; candidate-practice Work and Agents are asserted only when their own admission conditions obtain.
For every selected question, name its receiving pattern and the exact bounded practice claim or claims whose values change first use. If a required group or claim-to-question binding is absent at that grain, return a bounded PFAD gap to the architecture decision for completion before DPF authoring. A completed C.32.MWA synthesis is used only when several selected structures do not line up one-for-one, and E.23.CDI only when capability development changes the answer.
The answer is one identified claim-bearing episteme under C.2.1 and is recorded, with its rationale, in one ordinary E.9 DRR. The decision-maker selects that answer during decision Work. An authorized decision-maker accepts, redirects, rejects, or reopens it through a separately identified acceptance decision. Record the acceptance decision separately and give E.4.DPF the answer it accepted.
Common practice questions include:
If another question changes the answer, name it and the pattern that handles it. A required contribution, transformation, performed Work, capability, provider contribution, or cultural change belongs to its own pattern. Use the complete relation predicate in B.1.5 for Method parthood.
For a DPF Suite answer, establish the composite discipline through C.20 before selecting its framework arrangement. State its subject, knowledge and reusable practices, constituent subdisciplines, direct part relations, whole-forming couplings, identity and substitution conditions, composition-grounded characteristic and possible larger assembly.
Show what each constituent DPF explains and what the Suite must additionally teach. Distinguish external suppliers, specializations, publication support and gaps from constituent fields.
Compare that arrangement with independent DPFs connected by use, a smaller DPF, a general Core contribution and another partition of the existing Methods. A Suite is warranted by the described disciplinary whole and a useful composite pattern language, not by the number of publications, a shared audience or one cross-DPF application.
A broad field may be only partly represented. State the coverage boundary. Do not force genuine multiple parthood into a tree, but do not infer it from use either.
For a disputed constituent relation, hold the required result and access to complete Methods constant while comparing direct membership, independent disciplinary use and any applicable specialization. In the use alternative, the same practitioner may learn, perform, combine, adapt and check the supplying Methods. An external discipline does not imply an additional performer or only a finished result.
Identify whether the receiving whole needs one Method, a connected family or the discipline with its own subject, knowledge and responsibilities for change. Couplings that make the supplying field coherent establish grounds for that field's identity. Establish its relation to the receiving whole separately through C.20.
Find the subject, integration or substitution condition of the receiving whole that a use-only or specialization account would fail to explain, with the same operative Methods and return paths available.
Then apply the proposed ground to a nearby field whose contribution is also necessary but which the arrangement excludes. If the same ground includes that field, reconsider the whole's boundary or develop the missing distinction. Do not preserve the familiar roster by an unexplained exception.
When several part relations are claimed, make this comparison for each receiving whole. Failure to establish one relation leaves that claim unsupported without erasing the supplying discipline or prohibiting a different, supported composite whole.
Keep new Suite formation among the alternatives when no current Suite explains the supported field assembly. The answer may instead change the current partition through a split, merger or boundary revision.
If only one constituent DPF is ready, compare a first incomplete Suite edition with an independent DPF. Choose the Suite when its common account supplies a needed explanation of the DPF's place, connections or limits within the established whole. State the available results and the gaps that prevent a broader combined use. Otherwise retain the independent DPF arrangement until the common assembly is useful.
Selecting either architecture neither admits its edition nor establishes its publication.
The E.9 DRR records the selected description and publication architecture, alternatives, practical consequences and reopen conditions. The decision selects what to build or revise. Disciplinary parthood requires its own grounds.
Identify the Suite edition and constituent editions, an unambiguous home for each full Method, whole-level patterns and the route to external suppliers. Select the Suite's architectural-reference unit and its embedded or separate carrier. Identify whether it shares the framework's edition boundary or has an independently useful change rule under E.4:4.1.
The Suite's Architectural Reference must explain the selected architecture and use-changing reasons without requiring the DRR. Publication, availability, maintenance and product-series continuity remain separate claims under E.4:4.2.
For an existing-framework contribution, non-framework product, thinner route, or stop, state only the parts needed to explain that outcome and the later-used decision.
For a selected product, name the direct subjects of the decision's claims and the relations it uses. For a proposed product, state any unresolved kind. The selected outcome determines which field-assessment or package content applies.
When the architecture keeps, merges, removes, reuses, or omits a load-bearing contribution, record the E.8:4.1.3 same-situation disposition and the action or result that changed. Preserve a difference when it changes action at comparable effort without adding an unsupported or needless burden. A narrower label or example alone is not that difference.
When the answer treats a promised problem family as covered by a result supplied from outside the framework, name the exact external result, the specific content used, the kind of that result, its supplying product and exact edition or current state, receiving use, practical discovery route, and every currentness or availability condition that can change that use. State that the result remains external, and state maintenance only when it changes the receiving use.
When the receiving use requires an availability or compatibility result, unambiguously identify both the result and its supporting basis. If those facts are absent, or the result does not answer the promised use, record a gap or omission rather than relabelling the result as framework content, a MethodDescription, or source evidence.
When the selected keep, merge, removal, profile, external reliance, or omission materially changes the stable set for a promised problem family, obtain a current E.4.DPF.DA D12DomainProblemFamilyCoverageAdequacy result for the resulting exact DPF or LPF edition. Reuse a matching current result while that exact resulting edition and basis are unchanged. Do not require proof that a revisit occurred.
Keep the ordinary E.9 grounds, sources, affected loci, rationale, and consequences in the same DRR. Add naming, quality, admission, currentness, or package details only when they change the answer or a named later use requires them. Each added claim remains under the pattern that defines, constrains, or tests it and appears in PFAD for that named use.
State initial pattern relations directly
When an initial pattern relation changes the selected architecture, state the relation and its participants as an ordinary assertion. For example: Pattern A frames the recurring problem; Patterns B and C specialize its reusable move for two stated situations. Use the pattern that defines or constrains each relation function.
In the architecture answer, state what relation holds between the named participants. Use the predicate that defines that relation. Add an optional E.4.PFR row when a named maintenance use needs that representation.
Keep the answer, DRR, authoring, and publication distinct
The decision-maker selects the answer during decision Work. The E.9 DRR records that answer and its rationale. An authorized decision-maker accepts, redirects, rejects, or reopens it through a separately identified acceptance decision. Later authoring follows an accepted answer.
A framework edition is the pattern-language episteme assembled from accepted sources. Any maintenance relation obtains separately. Publish or project claims about these objects in the selected form—for example, an ADR-like document, site, or PDF—and identify its presentation carrier separately when needed.
When the answer uses C.32.MWA or E.23.CDI, keep each proposed Method distinct from the pattern that describes it, the Work in which it is performed, the result of that Work, the framework answer, the DRR, and the resulting edition. Use a proposal or evidence locator to find the supporting material.
Use C.32.PAD only when the question is a project architecture decision about a named composite project Work, and use C.32.ADR only to project that project decision.
For an ordinary framework answer, a selected public use can justify publishing the decision episteme or a reader-specific projection through E.17 and E.24.PUB.
The framework's Architectural Reference carries the architectural reasons needed for use and adaptation whether or not the DRR itself is published. Publication of the decision record is not a mandatory stage of principle-framework authoring.
Archetypal Grounding
Selecting a DPF for service launch and related work
A systems-management group considers a public DPF for recurring problems in service launch, cross-team coordination, incident response, and feedback-based improvement. A broad FPF route covers several shared distinctions, and an admitted neighboring DPF covers one specialist branch, but neither gives this practitioner group a coherent first use across the four problem families.
The field-boundary assessment compares a new DPF with direct FPF-and-source use, a guide, a contribution to the neighboring DPF, two existing DPF edition series, and no new maintained product now. In the proposed field, the way a service accepts an obligation constrains how it can later transfer work and restore service. Practitioners repeatedly have to design these relations together. The assessed alternatives leave that design to each practitioner. The proposed DPF explains the relations for the declared field and has an independent edition, change and refresh boundary. The service-launch case illustrates the need. If the claim of recurring use is still open, the architecture decision can remain provisional.
The source accounts organize Methods, dated Work, service and equipment subjects, descriptions, provider capabilities, and cultural change differently. A completed C.32.MWA result makes those correspondences and conflicts readable. The selected problem families and relations supply the DPF structure.
The E.9 DRR records the public promise, selected problem-family sets and material relations, representative case, omitted procurement and certification questions, the sources to revisit, which claims the evidence supports, suggests, or only motivates, the publication consequence, first authoring action, and reopen condition. It also identifies the selected FPF edition and any other relied-on framework editions, their required claims and receiving uses.
Capability development does not change this case, so E.23.CDI is absent. PFR rows and proposal locators serve their conditional representation and discovery uses.
Exploratory access result
Existing FPF and source material answer the immediate need through a curated route. The inquiry closes with that route because no later author or reviewer needs a settled framework boundary.
Decision-level access result
A team needs a settled choice among a DPF, an access route, and stop because later work depends on the rationale. With that architecture question open, the team selects the access route and no new framework edition. One E.9 DRR records the selected access consequence, stop, and condition for reconsidering the answer.
Non-framework programme product
A cross-domain inquiry need recurs, but practitioners do not need another pattern language. The decision compares a DPF, an inquiry-programme product, a separate inquiry evidence-package episteme, a curated route, and no new maintained product now. It selects the programme because named users need continuing access to inquiry Methods, bounded-project intake, and result return.
The answer records the admitted direct subject of the programme claim and its relations under the patterns that define them. Any maintenance relation is a separate claim.
The programme's first usable version is a current programme-description episteme that names the users and questions, inquiry Methods, project intake, result return, access, change, and retirement rules. Name any provider System, maintenance relation, accepted commitment, or admitted service state when it independently obtains and changes the answer.
Each bounded inquiry project requires independent A.15.1 admission as a separate Work occurrence, and each returned result is a separate episteme. A subject pattern may instead admit the programme itself as a System or another exact arrangement, in which case the answer names it.
A bounded project may end while the managed programme continues and evolves. The inquiry evidence package remains its own editioned episteme.
A composite engineering field and its Reference
A framework author proposes to assemble geotechnical, structural and construction-engineering DPFs. Their use on the same building project is insufficient. The field account instead explains how ground models constrain foundation design, design consequences determine investigation needs, constructability constrains structural choices, and observations in construction and service revise the models and Methods.
The field account identifies the field's continuing concern with realizable built works and the conditions under which alternative subdisciplinary contributions preserve that whole. These facts support the particular composite-field claim under C.20.
The Suite describes that field assembly and develops any missing whole-level Method for reconciling those contributions. Its edition includes the constituent DPF descriptions with their usable cases and limits. Finance is an external supplier when a project needs investment analysis. Its entire discipline is not made part of the Suite by that use.
The Suite's Architectural Reference explains that composition, the alternatives and the conditions that make the arrangement useful. A problem-led entry can direct a practitioner to the relevant engineering Methods and the external finance result. The account returns to their full explanations. Develop the integration operation in the Method whose result requires it. Give it a separate E.8 pattern when practitioners can recognize and use it as a Method in its own right.
If the author has only a list of three publications and no supported whole construction, the useful first release is a reading route or a bounded DPF proposal. It is not yet a Suite. If the field is supported but one branch is not yet described, the Suite states its partial coverage rather than presenting the missing branch as available.
Existing framework
A local practice framework already has an accepted architecture answer and a source record. Reopen that answer when its selected edition boundary, dependencies, initial pattern architecture, or publication or access consequence changes.
A change of a relied-on edition or a material change to a relied-on claim first reopens the affected use. Reconsider the architecture answer when that affects one of those architectural choices. An example or publication-carrier edit that leaves those choices intact follows the existing answer.
Candidate recognition before product form
Apply the same contribution comparison when an inquiry starts with a carrier, a role, or a count.
Same useful Method, different costs of access and learning
A framework author must make a method for recurring case reviews available to a group. Participants must identify the case question, obtain reasons for proposed changes, and let the case author accept or reject those changes with reasons. The method's operative distinctions and conditions are already settled. The architecture question is how participants with different preparation can learn and keep using it.
Alternative A puts a complete tutorial and the use instructions into one required reading route. Alternative B retains the complete method account for competent readers, supplies a beginner companion with practice and feedback, and exposes a short reminder for participants who already possess the capability.
Both alternatives must achieve the same case-review result. B's shorter reminder alone would be an inadequate third alternative for beginners. Direct qualified facilitation and continuing with the existing method remain alternatives if their availability or cost changes the choice.
For an illustrative cohort calculation, stipulate adequate results for A and B, 20 participants and 20 occasions per participant. All numbers below are assumed human minutes, not measured effects or a universal valuation of time.
B spends more on authoring and initial learning, but saves 2,000 minutes over this stipulated horizon. Its additional shared cost is 1,200 minutes and its per-participant saving is 160: it breaks even at 7.5 participants, so eight or more favor B on this cost alone.
For five participants, A costs 2,475 and B 2,875. B then costs 400 more. Different performance, expected use or support availability can reverse either answer. The author first needs evidence about the recurring burden or likely audience, not another abstract declaration that reuse is desirable.
Now a teaching example proves unsuitable for beginners. With B, revise that companion's example and the teaching result it supports. Retain the unchanged operative method and expert reminder.
If the rule for deciding which reasons warrant a revision changes, inspect all dependent forms. A common publication may still need a release, but unchanged subject content need not acquire a pedagogical exception. This change probe explains the selected boundary. Separate file names alone would not explain it.
An AI does not teach a participant to evaluate reasons merely by supplying a polished case answer. If independent human judgement is the promised result, add the needed practice and feedback and test it. If obtaining a competent case answer is the promised result, direct AI or human provision can instead be compared on that result and its costs.
Cheap exit. A prepared participant repeatedly overlooks one already understood question, and an available checklist prompts it at the right occasion. Keep that route unless a consequential defect appears. No new pattern, companion, audience survey or cost table is needed.
A new Suite with one available constituent DPF
Consider an illustrative Catchment Hydrology DPF Suite, initially covering groundwater and its interaction with surface water. Its subject is the investigation of water storage and movement within catchments. Groundwater hydrology supplies knowledge and Methods for aquifer storage and flow. Surface-water hydrology supplies those for runoff, channels and surface storage. An estimate of recharge supplied by a surface-water account constrains an aquifer model. An estimate of groundwater discharge must agree with the receiving stream's water budget.
Measurements test these exchanges and can require revision of either account. USGS investigation of groundwater/surface-water interaction provides a concrete basis for that coupling. The explained composition supports a mutually consistent and revisable account of catchment water movement. Models and measurement Methods can change while that coupling is preserved. The resulting account can contribute to wider Earth-system inquiry.
Suppose only the groundwater DPF is ready. Surface-water hydrology is an existing constituent field. Its absent DPF is a publication gap. The first edition contains the groundwater patterns and an architectural account that makes their relation to the wider field explicit. It states that surface-water investigation and the complete construction of a coupled catchment model are outside its available Method coverage. A forecast requiring those missing Methods is therefore not offered as obtainable from this edition.
The common account is useful now when a practitioner asks how aquifer pumping affects streamflow: it exposes why the available groundwater contribution is insufficient and which contribution must be obtained elsewhere or developed. In the published edition, an outside contributor can also use the stated composition and coverage to propose the missing framework or challenge the assembly.
If readers can obtain the needed result from the standalone groundwater DPF and do not need the common account, retain that arrangement for now. Naming that file a 'Hydrology Suite' and promising later water and land frameworks supplies neither the constituent-whole grounds nor a useful present assembly.
Bias-Annotation
Scope. This profile decides a later-used architecture boundary for an FPF-grounded framework, adjacent result or service, thinner route, existing-framework contribution, or stop. It does not provide a universal product ontology, a service-design Method, a publication taxonomy, or a mandatory decision form for exploration.
A team can mistake form for a decision when it starts from a schema, row, ADR heading, or status field and assumes that filling it has settled the architecture. Start from the reader's problem, alternatives, later-used boundary, and practical consequence.
Proposal, dependency, quality, naming, and publication material can appear before the reader knows whether a framework decision is needed. Include each item when a named use calls for it.
A small set of patterns under development can receive a broad DPF name before its field promise, several problem families, representative first use, omissions, and refresh boundary have been tested. Treat the slice as a seed or existing-framework contribution until the field-boundary assessment supports a DPF.
Authors can mistake source rows, levels, chapters, or diagrams for the product structure. Recover the Methods, Work, subjects, descriptions, capabilities, providers, cultural change, and their actual relations first. Use C.32.MWA when several structures must be reconciled.
Authors can mistake a table row or reference list for the relation it records. State which relation holds. Add a representation only when a named maintenance or checking use needs it.
Conformance Checklist
When audience access, whole cost or shared change can alter the architecture answer, apply :4.1.1: keep the promised result fixed, identify relevant preparation and support, compare an adequate available alternative, and expose the decisive cost or change dependency. This is a conditional substantive comparison, not a required numerical score.
Common Anti-Patterns and How to Avoid Them
Consequences
Authors get a recognizable framework question, one cheap stop rule, one readable decision account, and one next action. Later authors can recover the public field promise, problem-family coverage, representative use, edition boundary, dependencies, initial pattern architecture, omissions, sources to revisit, publication or access consequence, rationale, and reopen condition without reconciling two decision objects.
A new or substantially revised DPF carries more architecture work than a suitable non-framework product, thin route, or existing-framework contribution, and the PFAD profile adds one locator to maintain.
That cost makes field-scale evidence explicit before a public pattern language is selected. Conditional naming, package, quality, and machine-readable detail stays out until a named use needs it.
Rationale
The PFAD locator makes recurring framework-specific questions discoverable. One ordinary E.9 DRR records the bounded answer, alternatives, rationale, consequences, action, and reopen condition.
The field-boundary assessment reserves field-scale identity for a contribution with the practitioner coverage and independent use that justify it. The several-structure branch grounds the practice architecture in its actual Methods, Work, subjects, descriptions, capabilities, providers, and cultural relations. Direct assertions state each selected initial pattern relation under its own predicate. Optional representations serve their named uses.
SoTA-Echoing
The external comparisons above supply bounded grounds for the selected construction. The whole-cost method is their methodological synthesis with C.11.DUA/C.11.CRC. The synthesis makes no claim that any source established an optimal library size or an audience-independent publication form.
Relations
-
Uses:
E.9to record the one bounded answer selected by the decision-maker during decision Work. -
Uses:
C.11.DUA,C.11.CRCandC.19.2for worthwhile comparison and its burden,C.2.8for attainable structure, andC.31for the modularity and reuse questions that matter. -
Coordinates with:
E.8:4.2.1.1,E.11andE.11.PFPfor method explanation, pedagogical companions and usable publication/access routes. -
Uses:
E.4,E.4.DPF, andE.4.DPF.DAfor framework scale, authoring, field coverage, and package assurance; usesE.4:4.2when one decision selects a DPF Suite andE.11.DSGwhen that Suite has a problem-led DPF Suite Reference. -
Uses:
C.32.MWAwhen several practice structures need one readable synthesis. UseE.23.CDIonly when capability development for a named Work family changes the answer. -
Uses:
A.6.RCD,A.6.REL, and the patterns that define those relations for material relation assertions among initial patterns. -
Coordinates with:
A.22,C.30.STRAT,B.1.5,A.15.1,C.30.AD, andC.36for selected structures and architecture distinctions. -
Coordinates with:
E.4.PFRfor framework-edition dependency predicates and optional relation and edition maintenance representations. -
Coordinates with:
C.32.PADfor a project architecture decision about a named composite project Work,C.32.ADRfor its ADR-like projection, andE.17withE.24.PUBfor publication of an ordinary framework answer. -
Coordinates with:
F.18,G.2,G.11,E.21,E.23, andE.19only when naming, source synthesis, refresh, improvement, or admission is current for the selected answer.
E.4.PFAD:End
Domain and Local Principle Frameworks: Whether and How to Author and Publish
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative.
Problem frame
Use this pattern when a group needs to create a domain principle framework or local practice framework grounded in FPF: for example a hydroponic-cucumber framework, a neural-network architecture framework, or a Codex-process framework.
This pattern explains how to author or revise an FPF-grounded framework. Start by writing one paragraph that names the intended reader, the domain or local situation, the first useful action, and the ordinary stop or wrong-turn return. Add a non-use boundary only when an independently grounded competing use is plausible for that reader and changes action. That paragraph is enough to enter the first-hour route before the framework architecture, durable names, or publication package are settled.
Use this pattern when the work creates or revises the framework. Use E.11 or E.17 when an existing framework remains unchanged and the work only changes how readers or agents find or access it.
Problem
Domain and local framework authors often have strong source material and urgent local needs, yet three errors can undermine their use of FPF. They copy FPF terms without settling the domain ontology. They publish a framework carrier before deciding the framework architecture. Or they produce a useful checklist that is local process guidance but not yet an FPF-grounded pattern framework.
A working framework needs more than a good table of contents. It needs source-grounded pattern selection, architecture decisions, direct assertions of material relations, names, worked cases, quality evaluation, and refresh conditions. Include the relation or edition records a current maintenance use requires, for example for dependency, compatibility, migration, deprecation or supersession.
A DPF gives its intended users a source-grounded pattern language for recognizing typical problem situations in the domain, avoiding known failure modes, and applying SoTA solution moves with visible boundaries and refresh conditions. Its ontology and vocabulary support those moves. Known failure modes include beginner mistakes and experienced-practitioner failures caused by stale, local-only, or non-SoTA practice.
Forces
Solution
Use the first-hour route to help a reader new to the framework decide whether authoring should begin. This route answers that question before asking for proposal, dependency, naming, quality or publication material.
- Name the intended reader, recurring working problem and useful result. Identify the user by the work they need the framework for. Follow E.8:4.2.3 when the distinction between doing the work and assisting with it changes whose decision or capability the result demonstrates. Ask what the performer agreed to deliver, who receives or accepts it, and what choices the practitioner must make. If an existing result meets that order, show how to obtain and use it. When the reader must devise the connection among methods, E.4.CM:4.1–4.3 starts with the decisions still open to them.
- Recover candidate contributions from that problem and its sources before fitting them to the current Library: recurring practitioner problems, reusable moves or Methods, first useful results, ordinary stops or wrong-turn returns, and source or refresh boundaries. When the use spans several problem families, recover their consequential connections and the practice field or fields they serve. A field can include known Methods as well as missing ones; novelty of elementary operations is not its identity test.
- Inspect what FPF Core, existing domain or local frameworks, and current sources provide.
- Treat a role or competence account, several independently reusable contributions, a Card or mantra spanning material relations, a plausible field and refresh boundary, a representative use across contributions, or visible action or result loss under one broad owner as a cue to compare, not as proof of a DPF.
- In one recognizable situation, compare each contribution with the exact current FPF or admitted-DPF action, first useful result, stop or wrong-turn return, and source or refresh obligation. Include a non-use boundary only when
F.19's grounded-contribution test admits it. Preserve whether it is carried by an exact current owner, supplied by an exact external result, an action-bearing remainder, or unresolved. A shared topic, role label, carrier form, missing product name, or missing PatternIDs cannot close the question. - When the receiving use depends on connected Methods, establish what the intended reader receives as a whole under E.4.CM:4.4 before treating the constituent suppliers as a complete answer. A remainder can be a connection, a Method, a whole field or several fields. Keep the framework question open when exact ownership is unresolved or closure would erase a material relation, field or refresh responsibility.
- Distinguish content reuse from the architecture answer. Existing coverage establishes what can be retained; it does not freeze the boundaries of the frameworks carrying it. When the recovered families or changed uses put those boundaries in question, compare their current organization with a revised distribution of existing and new contributions under E.4.PFAD:4.2. A new or revised DPF can inherit or receive existing Methods. Several fields can warrant a Suite; one large pattern, an application or an umbrella title does not settle their organization.
- Test a cheaper search, curated reading route, access-only result or stop against the same receiving use. Take it when the comparison closes every contribution and no later-used field, edition, relation, direct-subject, publication or access decision remains. A sufficient narrow result needs no survey of unrelated fields.
- If a reusable problem-solution language still looks useful, state its field promise, connected problem families, material relations and representative first use. Compare new and revised frameworks with contribution to an existing one; use the field-scale test in :4.0. Retain a narrow useful result as a seed, contribution, guide, direct source use or another result whose kind fits its use when that test does not support a framework.
- Sketch provisional pattern candidates at the scale of the recovered problems. One to four can be enough to test a small proposal; neither that range nor the size of one carrier limits the number of Methods or fields to recognize. Keep independently reusable problem–move–result contributions and their whole-use relations visible before choosing their publication grouping.
- Consider five outcomes: a new or revised framework, a contribution to an existing framework, a non-framework product, a thinner publication or access route, or no new maintained product now. Would the choice settle an edition, dependency, initial pattern placement or relation, direct-subject identity or change rule, or publication or access decision that later work will use and whose rationale another author or reviewer needs?
- If no, take the useful contribution, thinner route, other result, or stop without a DRR. If yes, use
E.4.PFADto state which of the same five outcomes was selected and its framework-specific consequences in oneE.9DRR.
The five architecture outcomes are alternatives, not successive stages. A separate organization-design proposal is useful only when a named review use needs candidate organization claims. A separate dependency description is useful only when a named next authoring use needs a stable account of dependency availability and relevance. Neither is a prerequisite for recognizing or answering the architecture question.
When a DPF answer is selected and authoring begins, grow the seed only as far as the next use requires: a source-pack stub; provisional public names; the first pattern candidates through E.8; ordinary assertions of the material relations among them; optional E.4.PFR rows for a named maintenance use; a publication or access consequence; and the first quality and currentness route. Add the effective ReferenceScheme, ClaimScope, qualification window, or a selected BoundedModelUseStructure only when those distinctions change interpretation for the receiving use.
Stop when the result answers the commission. A direct answer can complete a request for guidance, and an inspectable seed can complete a request for a seed. Continue to usable patterns, assembly and the applicable evaluation when a usable DPF was commissioned. A rough DPF seed is inspectable when its reader, problem, useful move, source basis, provisional patterns and relations, edition/dependency boundary, publication or access consequence, and reopen condition are visible. Do not present it as a reliance-bearing DPF until the decision account is adequate for the intended authoring use, the pattern bodies are usable as normal FPF patterns, and the package is evaluated through E.4.DPF.DA.
Precision and object boundary after the first route. The first-hour route is Plain application guidance for one run-independent framework-authoring U.Method. This E.4.DPF episteme is a U.MethodDescription only because its EntityOfConcern is that independently admitted Method and its claims substantively describe how to carry it out under A.3.2. E.8 does not grant that membership. The Method, this description episteme, any U.WorkPlan, every dated authoring U.Work, and every result remain different objects.
If this account separately claims dated authoring or project U.Work, recover every precise performer's A.13 core and independently admit that Work through A.15.1. Cite F.6 only when the account also needs precise assignment-bound attribution through the same obtaining A.13 assignment. The complete tests remain in those patterns. The first-hour and complete authoring routes require neither a Work claim nor performer or assignment evidence.
The Work may use this description through an identified A.6.1 application and bindings. Recover any local system-role classification, capability, authority, responsibility, maintenance, access, Method, Work, or result claim through the pattern that defines it.
The first useful output closes the immediate question. It may be one of the following:
- a cheap route or stop with no DRR
- one
E.9framework-architecture answer selecting one of the five outcomes in steps 8–9 - an optional organization-design proposal whose candidate claims need separate review
- a post-existence architecture-description use
- an optional dependency description needed by a named next authoring use.
These results are selected by their conditions, not by list order, and they do not form a mandatory lifecycle.
Choose the source route from the current question and the result it needs.
Choose the constructive route by the missing result. C.39 supplies general construction, connection, explanation and change of Methods for a receiving use. E.4.CM applies that way to the framework author's public account of one composite Method or several independently usable Methods. F.0.2 compares source ontologies for a semantic result.
Developing the Method can expose a missing distinction, and comparing source ontologies can change the proposed action. The work need not follow a fixed sequence from a complete ontology to a complete methodology.
Establish framework scale before edition authoring
Run the candidate-recognition move in E.4.DPF:4 before a public product name, PatternID, pattern body, or pattern count is available. One Card, file, admitted MethodDescription, role account, or absent PatternIDs neither proves nor disproves framework scale; each can only supply evidence for the same semantic test below.
Do not infer a new DPF or LPF edition from the current authoring slice. One useful pattern is usually a seed, candidate, or contribution, so pattern_count = 1 is a strong diagnostic: ask whether the candidate supplies a connected pattern language rather than one useful result under a broad name. A few related patterns around one problem family may likewise form a useful selected problem-family pattern set inside an existing framework. In FPF, pattern nest remains the separate E.8 name for a publication and specialization placement grouping.
Run the same semantic test at every pattern count. A new edition needs a pattern language broad enough for its declared field or practice: a coverage map, selected problem-family pattern sets and material relations, a representative application, an internally usable first-edition set, an account of what the edition omits and where to return to its sources, and a credible edition, change, and refresh boundary.
Fail a candidate when those contributions are missing or do not work together for the named first use, not because its index has one row. Adding a second thin pattern does not cure that failure, while an unusually compact candidate still has to pass every part of the same test.
Develop the account of the field that the selected pattern language must explain. From the intended work and the relevant subject treatments, recover the characteristic entities, relations and operations, the grounds on which claims about them hold, and how practitioners obtain or revise them. Where the declared practice includes constructing its descriptions, units or operations, explain that construction instead of assuming it has already been supplied in the task data. Explain how knowledge constrains the choice and construction of a way, and how its use can challenge that knowledge; use C.20 when the framework describes a discipline.
Bring this account back to the proposed repertoire. Show why its grounds select, connect or change the contributions when a consequential condition changes. Use the result to retain or revise the problem families, pattern allocation and external reliance below. Publish the explanation through the account required by E.11.PFP:4.7 and its actual suppliers. It may be concise, distributed or interleaved with worked reasoning; it need not be one composite Method, a compulsory workflow or a separate ontology.
A successful application of the selected patterns supplies its bounded evidence, not the unexplained field account. Reuse an adequate current account and qualified source comparison rather than reconstructing them merely for a new edition.
For each promised family of work, follow the decision from the information available before the example chose its units, methods or model. A recurring department-level method and the record of one performance can use the same form. If the reader must choose the relevant wholes and model distinctions, show how to tell whether a change in the receiver's work calls for a different selection. Explain why the original choice still serves when the same range and distinctions suffice. A field rule can already supply the needed choice in a local practice. Its local adoption alone leaves the alternatives admitted by that rule unchanged.
Before authoring a new edition, the E.4.PFAD architecture answer states:
- what field or practice the public name promises to cover, the intended reader, the first use, and the ordinary stop or wrong-turn return;
- the recurring problem families, characteristic failures, and useful result families that the first edition includes or deliberately leaves outside;
- the candidate selected problem-family pattern sets and the material relations among their patterns;
- one representative application that crosses the patterns and problem-family sets needed for the first use;
- the selected first-edition patterns, every same-framework prerequisite needed for that use, and every relied-on external edition;
- what the sources and evidence support, including whether each load-bearing claim is actual, proposed, or still untested, and what must be realized or tested before a stronger claim is made; and
- where each contribution goes: into the new edition, back to an existing FPF or DPF, into an LPF or another available result, with its kind and supplying product identified, into direct source use, or into an explained decision to add no new maintained product now, together with the observation that would reopen the question.
During source comparison, recover useful wholes and their consequential joins as well as constituent operations. Source-coverage evidence answers what the sources contributed; it does not by itself establish recurrence, audience need or worthwhile public inclusion. Carry a missing composite, a meaningful variant or an anticipated use as a contribution to compare, rather than removing it because its parts already exist or publishing every source construction.
When audience preparation, the cost of learning and continued application, or a shared change boundary alters the framework answer, use E.4.PFAD:4.1.1. Its selected arrangement can include a method account for competent readers, pedagogical companions and short-use routes; E.8:4.2.1.1 keeps their content responsibilities distinct. An unwritten companion does not supply the learning it promises.
Before placing a proposed narrower contribution, apply E.8:4.1.3 to it and the broader available contribution in one recognizable situation. Keep or merge a warranted difference that changes the reader's action or result; omit or merge a true duplicate; repair or reject an unwarranted difference.
If something else answers the question, distinguish an available result from a MethodDescription, direct-source evidence, and an unavailable result; state maintenance only when it changes that use. This decides one contribution, not whether the package covers its public promise.
The first-edition set is internally usable only when it contains every selected pattern and every prerequisite from the same framework needed for the named first use. Keep relied-on results from an FPF, DPF, LPF, or separate non-framework product external when they are not members of this framework.
For each external result, identify the result and the content relied on, and state the result's direct kind, supplying product, edition or current state, receiving use, discovery route, and any currentness or availability condition that can change the use; say that it remains external. When the receiving use also needs a separate availability or compatibility result, identify that result and the basis on which it applies. When an edition dependency obtains, also name its direction, reason, and refresh condition.
If these facts are missing or the result does not answer the promised use, keep the family as a gap or omission; do not hide it behind the word closed.
When a keep, merge, removal, profile move, or external reliance materially changes the stable set for a promised problem family, obtain a current E.4.DPF.DA D12DomainProblemFamilyCoverageAdequacy result for the resulting exact DPF or LPF edition. Reuse a matching current result when that edition and its basis are unchanged; authoring history is not part of the D12 result.
Several sources may describe related Methods, their performance, support and development through structures that do not line up one-for-one—for example Method composition, Work order, model use, provider capability and cultural transmission. When those differences affect the framework architecture, use C.32.MWA to produce one readable synthesis for the E.4.PFAD answer; that synthesis does not choose whether to create a DPF or another result.
Use E.23.CDI only when the selected architecture includes developing capability for a named Work family, and use its result instead of copying its action sequence here.
When the DPF promises professional Method coverage, consume one exact accepted E.4.PFAD answer before treating a source list or pattern list as the authoring boundary. The five connected claim groups are a projection of that compact eight-part answer. E.4.DPF uses the projection without creating a second input record. Keep the accepted answer, its accepting decision, the E.9 DRR, the later framework edition, and any publication carrier distinct.
The five groups remain recoverable by value, filled only to the grain that changes the declared first use:
- Practice claims and first use: every bounded practice claim or promised contribution names its subject and scope, distinguishes existing practice from a proposal, and states what is supported, refuted or unresolved. Recover the practitioner, difficulty, sought result, first use, stop or wrong-turn return, qualification window, and receiving decision. Include only non-use boundaries admitted by
F.19's grounded-contribution test. Qualify each claim separately. - Project and Method positions: direct project subjects, use and environment, materially different solution forms, and Methods under their actual operational, system-change, solution, Method-of-interest, or Method-development relations. Incumbent Work, development or trial Work, candidate-practice Work, and intended Work keep different identities and truth status.
- Selected structures and correspondences: only the Method, Work, subject, transformation-flow, capability/provider, description, contribution, Method-development, and cultural structures whose correspondence, conflict, or non-isomorphism changes the answer.
- Pressures and evidence: constraints, conflicts, failures, environment or interest changes, and observed, source-supported, estimated, contradicted, and missing links remain distinct from causal history and temporal unfolding.
- Contribution, subtraction, gaps, and reopen: what current FPF and admitted DPFs already supply, each receiving pattern and domain filling still needed, exact external results, what is omitted or missing, and the observation that reopens the architecture.
One accepted answer may carry supported incumbent-practice claims, unresolved or refuted claims about existing practice, and candidate-practice proposals. Every selected question and resulting authoring disposition points to the bounded claims it consumes.
Independently obtaining A.13 agency claims and actual incumbent, development or trial Work keep their status; they support candidate-practice coverage only through the required relation and evidence. Public coverage is asserted separately at the scope supported by the exact edition and its evaluation.
For an obtaining practice claim, name actual recurring difficulties and representative actual Work. When the claim relies on a precise Agent performer, recover the A.13 core: exact admitted System, local agential kind and criterion, classification, obtaining assignment, and needed scope, working situation, and window. Add an agency-characteristic profile only when a Grade, autonomy or profile claim, a criterion-dependent characteristic, or a named assurance use consumes it.
A.15.1 then independently admits actual Work from its performance history, Method, extent, and containment; only after admission does F.6 add any precise assignment-bound attribution through that same assignment. A missing F.6 relation leaves Work membership intact and the attribution unresolved. State the evidence limits.
For a proposed practice, recover intended use, any incumbent Work or Method and available problem evidence, candidate Methods and architecture, and realization conditions. Reuse construction, source, derivation or validation grounds that suffice for the claimed use. Where required support is missing, either obtain the needed domain result at justified cost or retain the exact unresolved claim; binding requirements still limit use.
Plan a representative trial only when it is required or its attainable contribution justifies the full burden, retaining acceptance, failure and reopen conditions. A supported prospective design can be published as such; actual candidate-practice Work, Agents and practice coverage require their own grounds.
Turn the accepted input into one bounded authoring disposition for every selected question. Name the receiving pattern, the exact bounded practice claim or claims consumed, and the domain Method, evidence, constraint, direct relation claim, obtaining case, or planned trial still needed; or return a seed, external result, gap, omission, or reopened architecture question.
If the PFAD answer omits a required group or claim-to-question binding at the grain needed by first use, return that bounded PFAD gap instead of inventing the value. One clear question stays with its subject pattern. Use C.32.MWA only when correspondences or conflicts among several selected structures change the answer, and use its completed result rather than copying its action sequence.
E.4.DPF.DA evaluates the resulting edition once. D1DomainScopeAndUseAdequacy preserves the reader, first use, stop or return, qualification window, truth boundary, and any locally warranted non-use boundary of every public practice contribution. D4CoreDependencyAndDomainBoundaryAdequacy tests FPF subtraction, domain filling, and exact external dependencies. D5PackageFormLayeringAndRelationAdequacy keeps answer, accepting decision, DRR, edition, package, publication, and carrier separate.
D7PracticeUtilityAndProblemResolutionAdequacy uses the recognizable difficulty, practical move, and receiving result for each bounded claim without upgrading observed, planned, or missing evidence. D8HeterogeneousCaseAndTransferAdequacy examines what justifies the claimed transfer. A common rule can cover a class of uses. A claim about what practitioners actually did in another setting needs evidence of their work under the relevant conditions. D11DomainSoTAAlignmentAdequacy uses current domain sources, pressure evidence, limits, and reopen triggers.
D12DomainProblemFamilyCoverageAdequacy alone integrates the exact edition's several bounded public promises while retaining their different truth status; it cannot turn a prospective contribution into current practice coverage or erase an obtaining incumbent contribution.
The same case or source may support several coordinates, but each coordinate is judged once in the D1-D12 aggregate. Do not add a second project-Method architecture checklist, proof-of-revisit requirement, or second pass over the same evidence.
The input can be compact prose and a few selected structures. It is not a universal record schema, fixed view set, mandatory diagram count, source-chapter destination map, project lifecycle, or proof that a Method works or transfers. If no accepted answer identifies which practice questions change first use, or a required domain filling is missing, keep the affected contribution as a seed, gap, or omission and return to the exact E.4.PFAD or domain-source question. Do not infer Method parthood from a required contribution, transformation, Work enactment, capability, provider contribution, or cultural change merely because a source lists them together.
Keep the mapped objects distinct. A Method may be described by a MethodDescription, and a pattern may contain such a description only when A.3.2 applies. Selected patterns and their material relations make up the pattern language; a framework edition contains one version of that language; an independently identified physical or digital carrier may yield a particular publication or access-facing expression under E.24.PUB, with any reusable form constraining that expression separately; and an access route may help a reader or System reach the edition or a named carrier. Publication, availability, and actual access remain separate claims.
Semantic synthesis can support proposed claims and distinctions in the architecture. Methodological synthesis constructs the proposed ways of acting and their connections. Neither construction by itself establishes that a Method works or transfers.
Use product here only as Plain management wording for a deliberately identified result or service boundary. It helps a team state intended use, identity or current state, access, later change and retirement rules, and any maintenance that obtains. It is not one FPF technical kind and it creates no U.Product.
Before making a product-boundary claim, name the direct subject—the thing the claim is about—and the relation that carries its identity, edition, current state, provision, publication, availability, or maintenance.
Constitution or publication establishes only the claims made by those acts; maintenance and future Work need their own evidence. If the direct kind or relation is not settled, keep the management boundary as a proposal and return that question instead of inventing a common object kind.
A framework edition is an exact episteme. Treat its Readme, architectural reference, ToC, pattern bodies, coverage account, relation or edition note, and refresh route as named publication units in the same managed boundary when they share the edition's declared readers and use, edition boundary, access, and change rule. Being outside the pattern set or in another file does not by itself create another product or a maintenance claim.
Make a separate adjacent product only when people need to change, cite, or use its direct subject independently. Look for an independently useful identity, edition or current state, named users and use, an intensional rule for what belongs, access, a later-review or retirement rule, or cross-framework reuse or reliance. A separately established maintenance relation may also matter, but product identity does not require it.
A registry, guide, evidence package, companion, catalogue, tool reference, access service, programme, or another direct subject may justify that boundary; the label does not settle the subject kind.
When the direct subject is independently used or changed, keep it separate and point from the framework to its exact edition or current state. An annex may carry a declared snapshot or projection, but it returns to the authoritative subject and does not fork it. When no independent boundary is useful and ordinary framework use needs the material, keep it as a named support publication unit of the framework edition.
When programme is used, start with what continues. If a subject pattern admits the programme as a System or another exact arrangement, name it. Otherwise name the current programme-description episteme and any provider System, maintenance relation, accepted commitment, or service state that independently obtains.
Bounded inquiry projects remain separate Work occurrences, and their results remain separate epistemes. A maintained inquiry evidence package is its own editioned episteme. The management boundary may coordinate these subjects and relations, but it does not turn them into one indefinitely continuing U.Work or one generic Product. If the persisting arrangement is still unclear, return that architecture question.
A combined presentation carrier stays neutral. Each constituent keeps its identity, edition or state, form, access, later-change and retirement rules, and any separately established maintenance relation; the outer navigation names the exact constituents. Apply E.11.PFP only to FPF, DPF, or LPF constituents. An adjacent non-framework result uses the form and indexing discipline selected for its direct kind. DRRs, build manifests, quality runs, digests, logs, and campaign state remain process or maintainer evidence unless a selected reader use gives a direct subject its own product identity and publication or availability route.
When recurring domain wording prevents reliable use of the DPF patterns, apply the shared restoration method in E.10.ARCH and keep the domain entry beside the patterns that use it. Identify a separate local profile only when several entries have a named maintained use. A filled table can express the profile. Its reusable layout rules supply the publication form. A DPF with no demonstrated recurring wording problem needs neither.
Use E.4.DPF for the authoring route and its optional proposal or dependency branches, E.4.PFAD for framework-decision content, E.8 to author patterns, and E.4.PFR only when a named maintenance use needs a relation or edition record. Use E.11.PFP for the common framework publication form, E.24.PUB for publication occurrence, form, carrier, audience, bounded use, availability, and access, E.11 for practical entry, and E.17 for a source-backed publication face. Use E.4.DPF.DA and E.21 for package and pattern evaluation, E.23 for improvement, and G.11 for currentness. Each result or relation follows the predicate and evidence in its owning pattern.
If one receiving use needs reusable conditional unfolding, select one exact A.22.CGUS ConstraintGovernedUnfoldingStructure separately from this MethodDescription. Recover its A.22 identity, separately identified constituents and obtaining relations, applied constraints, more than one admissible continuation, and explicit stops or returns; keep any demonstrative walkthrough as a separate C.2.1 episteme. Otherwise keep the route Plain.
When separately admitted dated authoring Work first constitutes a framework episteme or a revised framework episteme, recover every precise performer's A.13 core and independently admit the Work under A.15.1; add F.6 only when precise assignment-bound attribution is also current. Recover the local inception claim separately through A.15.PROD.
C.2.1 identifies each authored framework episteme by its ClaimGraph, EntityOfConcern, and effective ReferenceScheme. An obtaining EpistemeEditionRelation, the authoring change or inception claim, the package architecture, and any EpistemePublicationRelation remain separately revisable. Publication occurrence, publication form, presentation carrier, framework truth, edition continuity, and package membership use their direct predicates and evidence.
The complete authoring account keeps the domain or local use frame, source basis, selected architecture, names, pattern drafts, direct assertions of material relations, publication or access, quality, improvement, and currentness returns recoverable. Keep any relation or edition records required by a named maintenance use recoverable too, without turning their order into another object.
When the selected architecture answer is to create or revise a DPF and authoring begins, keep claim-bearing epistemes, particular expressions, reusable publication forms, presentation carriers, and access routes separate.
Keep the accepted answer in a developer decision carrier written with the E.9 decision-record method and checked by E.9.DA. It carries the source basis, selected architecture answer guided by E.4.PFAD, initial pattern split and direct assertions of the material relations among those patterns, publication or access consequence, alternatives, rationale, consequences, first action, and reopen condition.
Publish the user-facing framework through a carrier named for the individual framework, as one assembled expression or as expressions of a split set of publication units. An exact access-facing artifact supplies the carrier only when its independently identified actual state or realization yields the selected expression; identify the service or route through which readers reach it separately.
Keep a source pack, E.4.PFR record, quality result, package evaluation, access manifest, or service description separate when its independent use and change require that boundary. C.2.1 framework-episteme identity, EpistemeEditionRelation, package architecture, E.24.PUB publication occurrence, expression, form, presentation carrier, access route, and actual access or use remain distinct; process state remains outside the user carrier.
A cheap route or stop remains outside this arrangement. Its result is the route or stop.
Plain vocabulary for adoption:
The labels SPF, TPF and xPF, and the strings FoundationalPrinciplePatternSet and ZPF, remain source aliases until F.18 settles a durable public name and any admissible short form. Use the full descriptive phrase "foundational principle pattern set" when that subject must be described before naming is settled. If an alias suggests a different framework identity, open the F.18 naming question before public use.
Keep the authoring apparatus proportional to the next receiving use. A first exploration may stop with a cheap route or no-framework answer and no decision record when it settles no later-used framework boundary.
A compact reliance-bearing framework may keep its Readme, architectural reference, pattern bodies, relation rows, source-use account, and quality route in one carrier when the same readers and stewards maintain them together. Split source packs, decision records, relation records, pattern files, quality results, skills, or access services only when independent editioning, confidentiality, transfer, automation, delayed feedback, expensive reversal, or another named reliance makes their identity separately useful.
Create an organization proposal only when candidate organization claims need separate review; create an authoring-dependency description only when a named next use needs stable dependency availability and relevance. More files or records do not make the framework more mature.
Prompt-shaped starter for SoTA harvesting and first candidate generation:
-
Domain or local use-frame declaration. State the intended reader, first use, stop or wrong-turn return, effective ReferenceScheme, ClaimScope, qualification window, and any non-use boundary admitted by
F.19's grounded-contribution test. Select a BoundedModelUseStructure only when its exact organization changes interpretation for this receiving use. Record each of these values through its direct pattern and predicate. -
Source basis and synthesis route. Select the applicable branch above. Use direct source reliance when one claim is enough;
F.1when source selection alone is current;F.0.2for one bounded comparison across source ontologies; andG.2only for the broadCG-Framepack and its downstream handoffs. Resolve derived lookups to authoritative editions, treat non-return as partial coverage, and classify earlier-DPF material as direct reuse, a receiving-DPF claim, or a proposed FPF improvement. Record adopted and rejected source payload, examples, currentness, and reopen conditions in the DPF source-use account. -
Cheap exit, optional proposal, or architecture answer. First test whether current FPF and sources close the immediate use through a cheaper route or stop without settling a later-used framework boundary. If so, stop without a DRR. Create the C.2.1 organization-design proposal described in 4.2–4.4 only when a named review use needs candidate organization claims.
When a later-used boundary makes the architecture question current, use
E.4.PFADto profile oneE.9DRR and select one of the five outcomes named in the first-hour route. If the answer selects a new or revised framework, state what field it promises to cover, its coverage map, representative application, first-edition set, external dependencies, omissions and returns, and which load-bearing claims are actual, proposed, or untested.Treat a one-pattern candidate as a strong warning and run the same framework-scale test used at every count. If the candidate lacks connected problem-family coverage, material pattern relations, a representative application, an internally usable first use, or a credible edition, change, and refresh boundary, keep it as a seed or contribution; the count itself does not decide. Keep the selected answer, acceptance, DRR, package architecture, direct relation assertions, any relation or edition records required by a named maintenance use, edition dependencies, authoring, and any ADR-like publication distinct.
-
Name and wording preparation. Use
E.10for kind discipline andF.18for durable names before public pattern heads or abbreviations stabilize. When a recurring domain wording failure blocks use, applyE.10.ARCHto write a local entry beside the affected DPF patterns. Create a separate profile only for a named maintained multi-entry use. A table can present its entries in a reusable layout. -
Architecture-use preparation. Before relying on all-in-one carriers, tables of contents, relation graphs, source summaries, search outputs, transformed views, or generated candidates as architecture evidence, apply
C.33to selected-structure recovery orC.34to structure-preservation comparison when that question is current. For an exact generated or discovered result intended to inform architecture work, useC.35to establish what kind of result it is, the organization concerned and the status of the actual or proposed claim, including any unresolved basis, next-use condition, and limit and return. -
Pattern drafting. Draft patterns with
E.8: recognition text, positive solution, worked cases, boundary, local anti-patterns, SoTA-Echoing, conformance checks, and relations. E.8 supplies authoring and publication-form rules; it does not make every pattern episteme aU.MethodDescription. Apply A.3.2 only when the episteme has one independently admitted Method as its EntityOfConcern and substantively describes how that Method is carried out.In a DPF, the pattern bodies render selected domain or local problem-situation architecture and solution-move architecture. When repeated first use benefits from an attentional aid, write a Plain local mantra by compressing that pattern's Solution without dropping the distinction that makes the move work or the stop, return, or redirect condition. Keep an established local name such as
mnemonic,watchword, orheuristicwhen it explains the aid better.Use
A.22.CGUSonly when an independently selectedConstraintGovernedUnfoldingStructurehas exact constituents, obtaining relations, constraints, admissible continuations, and stops; keep its demonstration separate. A thin skeleton, prompt seed, compressed design note, or memorable slogan detached from the Solution remains a pattern seed until anE.21evaluation finds the pattern adequate for the declared DPF use. -
Relation and edition discipline. State each material relation directly with its defining predicate. Use
E.4.PFRfor a relation or edition record only when a named maintenance use needs that representation. When dependency, compatibility, migration, deprecation, or supersession is current, keep the corresponding record recoverable. -
Quality cycle. Use
E.22to frame the evaluation purpose, quality floor, trade-off question, and expected improvement proposal when that frame is not already scoped. UseE.4.DPF.DAto evaluate the package as a DPF or local-framework package,E.21to evaluate individual pattern quality,E.23for repeated improvement, andE.19for selected admission or refresh questions, including affected-use replay underE.19:4.3.3when the candidate changes materially.If an evaluation result needs a carrier, publish or refresh that carrier through the pattern that defines its publication or currentness relation rather than through
E.22. -
Admission review. Use
E.19when the local process asks whether a pattern or framework slice is ready for admission. -
Support-unit and adjacent-product boundary. Use product only as the Plain management umbrella defined in
E.4:4.1. Group framework publication units only when they share the framework edition, declared readers and use, edition boundary, access, and change rule.For every proposed adjacent result, name its direct subject and test independent use and change, identity or current state, an intensional rule for what belongs, access, any later-review or retirement rule that changes use, and cross-framework reliance. State maintenance separately when it obtains.
Keep ordinary framework material as a support publication unit; keep an independently useful subject separate and point to its exact edition or state. Treat shared use and a combined carrier only as boundary probes.
-
Framework publication-carrier assembly and access-route check. Expose the selected framework episteme edition through exact publication and access relations. An independently identified physical or digital artifact supplies a publication- or access-facing carrier when its actual state or realization yields the selected expression; identify the service or route through which readers reach it separately and name any returned carrier.
When a Markdown carrier bears a publication containing the full
E.8pattern bodies, useE.11.PFPfor the common reader-facing title and edition cue, one logical index, and one practical-entry set with five-field ordinary entries and six-field selected cards. Show authorship, date, dependency, language, access, or a product-declared maintenance status, support window, or currentness window in the opening only when a product-specific publication rule names the reader decision or action they change.Keep the FPF heading hierarchy in that publication: the framework title and major publication units or Parts are H1, each PatternID and pattern title is H2, each canonical
E.8section is H3, and each nested section is exactly one level deeper. Navigation may surround a body, but it must not demote the body or merge two source heading levels.Keep the publication functions recognizable: in English use
Table of Contentsfor finding,<framework name> Readmefor short practical entry, and<framework name> Architectural Referencefor the full account of the language underE.11.DSGandE.11.PFP:4.7; translate them consistently in another publication language. Keep the account's declared section key separate from its displayed title. Do not create a parallelPattern Indexfor the same ToC function or rename the ReadmeReader Guide.Generated-source comments, source paths, source-set digests, machine identity blocks, build commands, and
do not editmarkers remain builder, package, manifest, or maintainer evidence rather than reader front matter. A compact card or summary remains entry guidance, while a distinct index remains a finding aid. Each returns to the ToC or Readme that locates its full pattern body, or directly to that body; do not present it as that body.After assembly and before calling the carrier released, current, or ready for its declared use, inspect the assembled carrier rather than only its sources: use the
E.11practical-use carry-through check for the public entries andE.4.DPF.DAfor package form, preservation of the published pattern bodies, and declared use. A successful build run shows that generation succeeded; it does not show that the published hierarchy, entry route, or pattern content survived. When the assembled publication claims accepted-source integration or continuity with its predecessor, useE.4.PFIPfor that comparison.Under E.24.PUB keep publication occurrence, selected episteme edition, audience declaration, bounded-use declaration, particular expression, any selected reusable publication form, and presentation carrier distinct; use the direct access pattern for actual access or use. Framework identity, package membership, truth, Work authority, and landing use their direct predicates and evidence. Domain or local frameworks publish through their own selected carriers.
-
Currentness route. Use
G.11for refresh plans, edition pins, source decay, deprecation, and supersession conditions.
Localize each repair before returning to wider framework architecture. A changed source payload first reopens the direct source use, F.1 source cut, F.0.2 comparison or G.2 pack used for that payload. Then reopen only the dependent assertions, examples or relations. A change to a relied-on FPF or domain-framework edition, or a material change to the Core claims used, first reopens the affected E.4.PFR dependency, compatibility, and migration relations.
Repeated misuse of one pattern first reopens that pattern's E.21 result and its E.23 improvement loop; a repeated domain wording failure may also reopen its local E.10.ARCH entry. A failed publication or access route first requires E.11, E.17, or the carrier relation that exposed it. A local mantra that no longer preserves its pattern Solution requires comparison with the exact Solution in that pattern body; A.22.CGUS becomes current only if the repaired aid must present a wider conditional unfolding.
Use E.4.PFAD only when the evidence changes selected framework-family, pattern-split, relation-structure, publication-form, presentation-carrier or access-route architecture, or dependency-boundary decisions. Use G.11 when edition currentness, source decay, telemetry, deprecation, or supersession must be orchestrated across those local repairs.
For an all-in-one DPF publication carrier, assemble the content in a reproducible order. This order is a publication shape, not a new framework kind.
In an all-in-one Markdown publication that contains the full pattern bodies, the framework title and major publication units or Parts use H1, pattern bodies begin at H2, and their canonical sections begin at H3. The framework name and publication language may vary with the domain and readers; E.11.PFP's reader-facing sequence, pattern-row profile, publication-unit jobs, and the heading hierarchy do not. In English label those functions Table of Contents, <framework name> Readme, and <framework name> Architectural Reference; use one consistent translation in another language:
-
Public framework title: use a domain- or practice-specific framework name such as
<DomainOrPractice> Principles Framework;Principles Frameworkalone is only the head or kind phrase, not an individual framework name. Do not putlocal monolith,draft, process status, or file-layout slang in the public title. -
Short public edition line directly under the title: name the stable public edition designation and a public edition-record locator. Add a dependency, language, access, product-declared maintenance status, support window, currentness window, or other short cue there only when the product-specific publication rule names the reader decision or action it changes. Do not create a separate edition H1 or put a machine identity block, source digest, source path, or build marker before the ToC; keep detailed edition and relation records after the pattern bodies or in maintainer evidence.
-
Table of contents: place one search-oriented overview before the body collection and use
E.11.PFP's five-position pattern-row profile. Every pattern row exposes its PatternID and title and gives at least one recognizable working-question cue inKeywords & Search Queries. State the DPF's declared reference code and local-locator form where a reader can disambiguate citations.Keep the displayed
§position separate from PatternID; the current row order may be non-ascending by PatternID but must match the body order. UseStatusandDependenciesfor values that can change the reader's choice; do not copy first move, result, and boundary into additional mini-method columns. Pattern bodies remain the main language of use; support maps and any relation or edition records required by current maintenance remain reachable without becoming a universal first inspection sequence or prescribed use order. -
Framework Readme: for the intended reader, present recognizable first-entry situations and practical questions, the first useful result or blocker, the direct pattern or small plausible set, the ordinary stop or wrong-turn return, and any non-use boundary admitted by
F.19's grounded-contribution test. State briefly which selected domain or local structures this carrier exposes. -
Framework Architectural Reference: the full public account required by
E.11.PFP:4.7andE.11.DSG, embedded here in the all-in-one carrier. Explain the whole's purpose, constituent Methods and their connections, shared conditions, architectural reasons, source synthesis, serious alternatives, limits and worked use, including the structure families the carrier foregrounds, coarsens, defers or returns to sources and pattern bodies. Give this one account the qualified public title and theReferencesection key defined by E.11.PFP:4.7.1. A separate preface or independently managed account is not required, and readers can enter a known pattern directly. -
Package carrier structure-account: intended reader and use, selected source-structure denominator, recurring problem-situation structures, reusable solution-move structures, captured structure, deliberately coarsened, abstracted, omitted, or lost structure, source-return condition, and quality or epiplexity route. Integrate this explanation into the architectural reference, with a short Readme cue where it changes first entry; it does not require a second account.
-
Package boundary and subject-pattern routing: Core subject patterns reused, local terms bounded, and source, evidence, assurance, publication, and refresh exits named.
-
Pattern bodies: each drafted through
E.8, with its PatternID and title at H2, canonical sections at H3, and deeper sections retaining their relative levels; each carries recognition text, positive solution, worked cases, local anti-patterns, SoTA-Echoing, conformance checks, and relations, and each is evaluated or marked as a seed underE.21before the package is claimed for public, teaching, enterprise, or reliance-bearing use. -
Grounds for the claimed use and transfer. Establish which relations hold over the declared range, which choices the practitioner must make and how the pattern set supports them together. Use a contrasting application when it helps the reader recover or test a distinction the available account leaves unclear. A single argument can establish the full stated range.
-
Support maps or appendices: detailed architecture-bridge, source-use, precision or package-name lookup material placed after pattern bodies unless a short first-entry trigger table is needed. Figures and tables needed to understand the architectural reference remain with that explanation.
-
Source use and refresh map: source rows with adopted payload, rejected or bounded readings, the conditions that reopen a direct source use, F.1 source cut, F.0.2 comparison, or optional G.2 pack, and the conditions under which source currentness or refresh must be reconsidered with
G.11. -
Conditional relation and edition records: add
E.4.PFRrows only when a named maintenance use needs a stable representation of dependency, specialization, publication, source reuse, evaluation, generated-carrier, teaching publication-carrier, ethics, deprecation, supersession, or edition effects. Otherwise keep the direct assertion. -
Refresh dependencies: which source-use, pattern-quality, package-adequacy, edition-dependency, or publication-carrier claim must be reopened when sources, relied-on framework editions or Core claims, local use, telemetry, or evaluation change.
Every DPF publication or access-facing carrier named here is independently identified under its domain identity rule and bears a particular expression; any selected reusable form constrains that expression separately; an access route may help a reader or System reach the edition or a named carrier, but the route alone establishes neither expression bearing nor availability or actual access.
In an all-in-one publication carrier, the Readme gives the short practical entry and the architectural reference gives the full explanatory account, sometimes using a narrative rendering of the domain. Their representation relation remains inspectable when they say what they are telling, for whom, which structures they foreground, which structures are deliberately coarsened, abstracted, omitted, or left to source return, and where a reader returns for fuller pattern, source, evidence, or relation detail.
The return can go beyond summarizing one text in another. Its source may be actual or possible holon structure, an architecture description, a view, a source pack, a model, a graph, or a pattern set. In architecture-mediated narrative-rendering use, read the return chain as narrative rendering borne by an exact presentation carrier -> architecture description or view -> architecture as selected structures under its exact use frame -> wider source structures. When no narrative rendering is present, read the first step as selected publication expression borne by an exact presentation carrier -> selected source structures.
If entry begins at an access route, name the first expression-bearing carrier or response reached and follow the same chain. Each step states selected structure, captured structure, coarsening, abstraction, omission, loss, and return conditions. An architecture description is often already a coarsened representation of selected real, expected, candidate, or actual structures, so the DPF carrier keeps that second-step loss visible. This does not make every DPF a literary narrative or every carrier a narrative.
When a sequential narrative rendering is load-bearing, use A.6.3.NAR; when the publication expression deliberately keeps only a narrower-use coarsened rendering, use A.6.3.CSC; for structure capture and loss, use C.33; for same-enough or preservation claims, use C.34; for practical-use publication and access, use E.11, E.17, and the direct publication or access pattern; for package adequacy, use E.4.DPF.DA.
Keep process and build state out of the carrier. DRR text about the current version's development, handoff notes, ledger rows, review status, helper state, admission blockers, landing evidence, generated-source comments, source paths, source-set digests, and build instructions may shape or identify the package, but the publication carrier should contain only durable user-facing package content, source-use boundaries, relation records, quality routes, and refresh conditions.
A short source-use or relation record may appear in the user carrier when it helps readers and maintainers use the DPF; a DRR argument about the current version's development, review transcript, quality proof, or build manifest does not.
Publish durable architectural reasons that help readers understand, select, combine, or adapt the Methods and profiles, with the necessary explanation and source return, under E.8:4.2.3; retain the dated decision and its evidence in the development record.
For access-facing carriers and routes, keep the same framework edition identity, direct relation assertions, and any relation or edition records required by current maintenance visible. An exact artifact supplies the carrier when its independently identified actual state or realization yields the selected access-facing expression. Distinguish a continuing file object, an immutable snapshot value and an individual copy under E.24.PUB's carrier-domain rule; a service address does not settle their continuity. A service or other access route names any returned carrier separately.
When an implementation uses a skill pack, MCP service, endpoint, retrieval or search route, or assistant integration, classify it through those same predicates. If the route returns a generated or discovered result for architecture use, apply C.35 to that result; if it performs Work or triggers tools, use A.15 and the pattern that defines the local tool or Work relation; if it claims currentness, evidence, assurance, or decision authority, retain the direct rule and grounds for that claim.
Use G.11 for currentness, A.10 for an implicit or required-replayable source-to-use account, B.3 for a named assurance claim, and E.9 for a decision that needs its rationale record. Take the DPF architecture from the accepted architecture answer and the framework episteme. Use E.4.PFR only when a named maintenance use needs a stable relation representation.
Starter evaluation characteristics for a principle-framework improvement loop:
These are evaluation characteristics for selecting and framing improvement Work. They are not measurement programs by themselves. If the pass needs a DPF package adequacy result, use the predicate defined in E.4.DPF.DA; if it needs individual pattern quality, use E.21. For DRR adequacy, FPF-level Pillar adequacy, measurement, or architecture-characteristic evaluation, state the exact subject assertion under E.9.DA, E.2.DA, C.16, or the relevant architecture-characteristic pattern.
For evidence use, retain the relied-on claim, its independently established sources and support relations, and the conditions of bounded reliance. Use A.10 when that source-to-use account remains implicit or must remain replayable; assess currentness when it can change the use. Keep each local result under the pattern that defines or tests it.
This episteme's A.3.2 MethodDescription use and its result-and-use account are sufficient only when a reader can answer all eight questions.
- Which framework episteme edition is being authored?
- What problem-and-solution architecture does it render?
- Which sources and decisions shaped it?
- Which patterns and material direct relations were selected?
- Which relation or edition records does a current maintenance use require?
- Which publication occurrence, form, carrier or access relation exposes it?
- How does its quality improve?
- When does it return for refresh or repair?
If the account also claims dated Work or a result relation, identify that claim through its direct pattern; E.4.DPF requires neither claim merely to describe the authoring Method.
Author a Suite from its composite discipline
A DPF Suite is a larger DPF describing a composite discipline. Use the accepted E.4:4.2 and E.4.PFAD answer and C.20 to determine the constituent subdisciplines and their frameworks. Recover what those disciplines explain and make possible, and what must be understood or constructed at the encompassing scale. Develop the field account required by :4.0 for that composite subject. Explain the consequential relations among the constituents' knowledge and practices, including any shared distinctions, objects, operations or conditions that their separate accounts do not determine on their own.
Obtain the needed joint construction and return it to the constituents. Show how a changed whole-level condition changes the selection, interpretation or combination of their contributions, and how a constituent result can revise the encompassing explanation. Retain the constituent DPFs' own problem families, grounds, sources, cases, limits and assurance. Explain where one constituent can be used directly and where the composite work needs more; not every application requires every constituent.
Supply separately selectable missing Methods as complete E.8 patterns in their appropriate framework. The Suite account makes the composition understandable; its publication or a list of dependencies does not create that composition.
Return a proposed change of field boundary, part relation or architecture to E.4.PFAD. Reuse, a common audience and a guide entry do not by themselves justify moving a whole DPF into the Suite. General transdisciplinary Methods may belong in Core; an independent supplier remains a direct dependency or use. A specialized contribution keeps the conditions that distinguish its field. Reconsider existing allocations when new synthesis changes these distinctions.
The Suite edition identifies the constituent editions it describes and assembles. State what coverage is incomplete and which contributions are unavailable. Keep one unambiguous current home for each full Method; replace maintained links when that home changes. For a dependency or compatibility claim, name the editions, relied-on contribution, receiving use, conditions and reopen fact under E.4.PFR. Publication, availability, framework-series continuity and maintenance do not establish disciplinary parthood.
Select practical examples for the product
Before shaping a DPF or LPF Readme, distinguish three jobs. A compact locator points to a direct pattern when retrieval is enough. An ordinary practical entry shows how one direct pattern or one bounded direct route can answer a comparatively simple difficulty without a mantra. A Practical-Use Card is selected only for a recurring complex difficulty when a repeatable formula materially helps the reader retain a useful path through several direct pattern contributions, checks, and returns.
Apply the E.11 comparison to the same truthful content with and without a mantra. If the reader can choose, obtain the first result or blocker, and return just as reliably without it, keep the locator or ordinary entry. Do not infer card form from pattern count, heading depth, phrase length, an inherited label, or a desired quota. Do not avoid the comparison by calling a rich cross-pattern example ordinary.
Keep one declaration for the product. It assigns every selectable example key exactly one ordinary-entry or card form. When the product selects cards, the declaration gives one measurable rule for compactness plus mantra and compact-card maxima. Authors, builders, and validators use that same declaration and the shared visible grammar from E.11.PFP. A product may select no cards when locators, ordinary entries, or direct guide answers already support reliable choice and return; it then carries no card-form burden.
Choose examples for the product’s readers and their recurring questions. Apply E.11:4.1 and its practical-use check to the Readme as people will read it. Show how to take the question they bring into a direct method, how to recognize the part a displayed example can answer, and where to find the method for the remaining work. A reader with a sufficient answer can stop there. Use F.19:4.4 to decide whether the reader also needs an explicit correction of a catalogue or coverage-boundary reading. When both simple direct use and extended cross-pattern use matter to adoption, show representative examples of both without turning every useful topic or pattern set into an entry.
Do not copy the FPF key inventory, card count, whitespace-token limits, or optional @FPFReadme records, and do not create a rival local card grammar or second key registry.
Keep a pattern-local mantra used to recall one pattern's Solution, a Readme card mantra used to retain a longer cross-pattern path, and an independently admitted CGUS demonstration distinct. Claims about a Method, performed Work, project result, or stronger relation use their direct patterns and evidence.
Keep pattern addresses stable while publication order changes
Before public references accumulate, declare a short, stable code for the DPF and assign one local locator to each pattern. Together the DPF code and local locator form the PatternID. Within that DPF, each PatternID is unique. The code is only a short reference to that named DPF; it need not be globally unique and does not identify an edition. Settle a new durable public abbreviation through F.18.
If the public DPF code later changes while the same DPF continues, preserve old references through an explicit F.13/F.18 rename or alias relation and a reader return; otherwise say where old-code use stops. Do not silently rewrite old citations. Do not assign the old code to another framework where readers may encounter both sets of citations without the full framework name.
A PatternID lets readers refer to a pattern carried forward across editions. The identifier does not prove that two bodies are the same pattern and does not define the pattern's claims. Keep it while the recurring problem, distinguishing working move, useful result, and ordinary conditions for use, stopping, or returning still describe the same practical answer. A title, wording, Part, publication position, or authoring work package may change while that answer continues.
A PlannedCatalogEntry may reserve an unused locator, but it is not yet an addressable PatternRef or evidence that a continuing pattern exists. When a complete pattern is admitted and published, its author may adopt that locator as the pattern's PatternID. The publication may use it as a PatternRef only after the complete body exists and the continuity judgment has been made; the reservation alone establishes neither identity nor addressability.
When the practical answer no longer continues, decide the split, merge, replacement, or retirement explicitly. Keep an earlier PatternID only for the answer that continues. Give each new pattern a new unused PatternID, and never reuse a retired PatternID for unrelated content.
If readers still use an old public reference, publish one maintained migration assertion. It names the earlier PatternID, any current PatternID or PatternIDs, whether the practical answer continued, split, merged, was replaced, or was retired, the uses covered, and where readers continue or stop. Add a structured E.4.PFR row only when a named maintenance use needs it; otherwise the readable assertion is enough. If no migration assertion is maintained, say where use must stop.
The local locator may use the numeric or mnemonic segments allowed by E.8. Prefer a numeric locator when it mainly needs to remain a durable address among many peers. Use a mnemonic segment only when it names an enduring distinction likely to outlast the title and publication position and materially helps recognition. A mnemonic is an address aid, not a compressed definition. Do not restyle established PatternIDs merely to make the set look uniform.
Keep current publication order separate. The ToC and body collection use the same Parts and the same within-Part order, while a § or position field shows that order separately from PatternID. Non-ascending PatternIDs are valid. State dependencies, Method relations, use order, and replacement relations in their own claims; do not infer them from identifiers, adjacency, Parts, or authoring work packages.
To refer to a pattern, retain its complete PatternID and enough context to identify the DPF. The full address may already identify that framework within the selected publication set; add a framework name only when needed. A citation intended to select the body published in one edition also names that framework edition.
For a reference that must travel beyond its present context, apply E.8 H-10 and E.11.PFP:4.2: preserve the complete PatternID or SectionRef and restore any identifying context that will be lost. Declaring the existing DPF code as a document qualifier does not change its PatternIDs or make that code globally unique.
Select the current first result
Select the result whose condition is true now:
- Cheap route or stop. Existing FPF or source material closes the immediate use and no later author or reviewer needs a settled edition, dependency, initial pattern-placement or relation, or publication/access boundary. Use the route or stop without
E.4.PFADor anE.9DRR. - Framework-architecture answer. A choice among the five outcomes in steps 8–9 must settle a later-used boundary. Use the
E.4.PFADprofile and record the selected answer, including relations among initial patterns that change the architecture, in oneE.9DRR. PFAD supplies no separate result or relation. - Organization-design proposal. Candidate organization claims need their own review before an architecture answer is selected. Use the C.2.1 proposal episteme locally called
FrameworkOrganizationDesignProposal. The proposal is optional and is not a prerequisite for the architecture question. - Architecture-description use. The framework entity, architecture relation, and selected structures already exist, and the immediate question is how an architecture description may be used. Use
C.30.AD; itsArchitectureDescriptionUseCard@Projectname is retrieval-only. When project locality depends on a composite projectU.Work, identify that Work underA.15.6, recover every precise performer's A.13 core, and use A.15.1 for independent Work admission. Add F.6 only when precise assignment-bound attribution is also current. Keep the description-use relation separate. - Authoring-dependency description. A named next authoring use needs a stable account of dependency availability and relevance. Use the C.2.1 episteme locally called
FrameworkAuthoringDependencyDescription. It may cite the accepted answer and itsE.9DRR when that basis matters, but it is neither a prerequisite nor an automatic result of the architecture answer.
The condition, predicate, and receiving use of each result determine when it exists. The five results are alternatives rather than stages.
Make the intended result reviewable when a separate organization proposal is needed
First make the design target present. IntendedFrameworkResultDescription is an ordinary local use name for one exact current C.2.1 U.Episteme, not a root kind or card kind. C.2.1 identifies it by:
The current A.15.2 WorkPlan declares coordination claims for possible future DPF-authoring Work, the intended framework-result kind, and its acceptance target. The WorkPlan exists now. A dated occurrence of the authoring Work and the framework result may still lie in the future.
The intended-result ClaimGraph states the domain or local use frame, readers, first uses, purpose, declared relation-family coverage constraints, intended-result constraints, and acceptance conditions. The effective ReferenceScheme interprets those claims. One exact A.2.6 ClaimScope separately bounds which claims and uses are current; changing scope does not substitute for changing the C.2.1 identity triple.
Keep use qualification and empirical grounding in their defined neighboring relations. If interpretation for the receiving use depends on an independently selected BoundedModelUseStructure, cite that A.1.1 and A.22 structure as an optional neighboring use qualification; effective ReferenceScheme and ClaimScope retain their positions in the account.
If empirical grounding is current, state a separate C.2.1 EpistemeEmpiricalGroundingRelation to one A.1-admitted holon and the covered claim subgraph. The grounding relation, its evidence, the WorkPlan, any separately claimed dated authoring Work, any separately claimed composite project Work, and the description episteme remain distinct. Each precise performer has an A.13 core; Work claims use independent A.15.1 admission, A.15.6 when composite, and F.6 only for current precise assignment-bound attribution.
Each declared relation-family coverage constraint is one FrameworkOrganizationCandidateClaimNode with claimNodeKind=constraint. Its coveredRelationFamilyRefKindPairs[1..*] identifies each covered relation-family value together with its exact kind; admittedFrameworkUseDescriptionRef names the use for which that coverage matters; and coverageCriterionDescriptionRef states how satisfaction of this coverage constraint is judged. A current A.15.2 WorkPlan acceptance target remains a different position: cite it through designBasisRefs[] or its direct acceptance-target relation. It neither replaces the coverage criterion nor shares one union field with it.
Create one C.2.1 proposal episteme
The organization proposal uses the present intended-result description as its one EntityOfConcern:
FrameworkOrganizationDesignProposal is a local use label for that C.2.1 episteme, not a second U-kind. Its one EntityOfConcern, one constituting ClaimGraph, and one effective ReferenceScheme supply episteme identity. ClaimScope, reader and use descriptions, optional model-use structure, empirical-grounding relations, A.7 provenance, F.15 proposal-status assertions when current, publication, and edition continuity are neighboring claims or relations; none is another identity slot.
A changed ClaimGraph, EntityOfConcern, or effective scheme identifies another episteme. A changed grounding relation alone changes that relation, not the proposal identity.
The proposal and dependency-description schemas use the C.2.1:4.1.2 ranges for their complete interpreted claim content and operative interpretation rules. Their narrower content requirements remain those stated in this section and :4.5. A graph encoding expresses those claims through a recoverable correspondence; graph shape or a nonempty file does not satisfy the content range. A partial scheme can satisfy its range while leaving a proposed relation, modal status or dependency condition uninterpreted; resolve that gap before the dependent use.
Use F.9 for an exact cross-context local-sense translation with its own endpoints and predicate. Candidate organization claims are expressed by the local node form below, with logical, alternative, refinement, dependency, support, conflict, and answer-to-question relations as current. The claims and their semantic relations constitute the proposal's ClaimGraph; its encoding has separate node and edge identities.
Each candidate organization claim node makes the subject-level proposal recoverable:
FrameworkOrganizationCandidateClaimNode is a local ClaimGraph node form, not a U-kind and not an episteme. FrameworkOrganizationClaimNodeKindValue is the local C.2.1-compatible enumeration definition | constraint | property | assumption. A node with claimNodeKind=constraint classifies a proposed constraint claim; its proposedConstraintDescriptionRefs[] identify the exact constraint descriptions that the node asserts, while a non-constraint node may cite those refs only when they qualify that definition, property, or assumption.
A relation-family coverage constraint node also has non-empty coveredRelationFamilyRefKindPairs[], one admittedFrameworkUseDescriptionRef, and one coverageCriterionDescriptionRef; other claim nodes leave all three coverage positions absent. Each pair identifies one relation-family value and its exact kind without a union field or untyped companion list. A WorkPlan acceptance target, when current, is cited separately through designBasisRefs[] or its direct acceptance-target relation. Both constraint positions describe the proposed organization.
If an obligation, recommendation-as-duty, or prohibition is current, use A.2.8 -> U.Commitment with its actual duty bearer, direct predicate, modality, referents, scope, validity, and instituting basis. A system-role kind or assignment may be an applicability ground; the commitment names the duty bearer.
If a permission, exercise, non-violation, or permission-conflict claim is current, use the exact A.2.8.PER result with the participants, references, constructive ground, and qualifiers required by that selected object.
FrameworkOrganizationClaimStatusValue is the local enumeration candidateProposed | rejectedAlternative | unresolved. FrameworkOrganizationAspectValue is the local enumeration frameworkFamily | component | dependency | patternRelation | publication | access; a domain extension adds another value only together with its exact interpretation rule in the proposal's effective ReferenceScheme.
Proposedness is claim modality: it says that a relation signature, position, constraint, invariant, or dependency direction is being proposed for the intended result. Actual relation occurrences and actual U.Structure values use their direct admission predicates; a pattern-use boundary condition remains a separate position.
The proposal's effective ReferenceScheme maps each organization-aspect value, described position kind, and proposed relation signature to claims about the intended result described by the EntityOfConcern; distinguishes ClaimGraph edges from the subject relations those claims propose; declares how basis and design-question refs qualify each claim; and states that claim status is modal rather than actual.
For example, a claim node can propose that one pattern family depends on Core, that publication and access remain separate positions, or that one relation invariant is preserved, without asserting that the future framework or those relations already exist.
Preserve proposal, answer, structure, and architecture boundaries
Keep the two result positions separate. If reliance-bearing E.11.PUA support materializes an exact expected-result support object for this E.4.DPF application, its expected result kind is the C.2.1 proposal episteme locally called FrameworkOrganizationDesignProposal. The intended later framework edition is described inside the separate IntendedFrameworkResultDescription and the proposal's ClaimGraph. One expectation support object never denotes both results, and neither object says the result was produced without the exact current work/result or inception claim.
Return to a framework-architecture question is separate from claim modality. When a reliance-bearing use needs an addressable return condition, E.11.PUA boundary support may name E.4.PFAD as the pattern for the next question and state which candidate claim, alternative, unresolved position, constraint, or dependency makes the downstream-used architecture boundary current. That support is adjacent to use of the proposal; it is not a proposal component and creates no PFAD result.
Subject organization is recovered from the candidate claim nodes, proposed subject relation signatures, described position kinds, constraints, invariants, dependency directions, alternatives, basis, questions, and framework-architecture settlement conditions. An A.22 U.Structure over the proposal ClaimGraph is optional and admissible only when the organization of the proposal episteme itself is a separate current EntityOfConcern. A selected BoundedModelUseStructure is a separate optional use qualification, admitted only when that exact organization changes interpretation for the receiving claim.
Proposal admission depends on the proposal identity and required claim content; the optional structures describe the proposal or qualify its use. The proposal is reviewable when it contains candidate organization claims and proposed subject relation content; headings, topics, and ClaimGraph organization arrange that content.
Before realization, C.33 notes compare proposal content with a declared current comparator: design questions, present basis epistemes, candidate alternatives, a relation-family coverage constraint claim node for an admitted framework use, or an earlier existing framework edition. When coverage is the comparator, C.33 cites the exact candidate claim node and reads its covered family ref-kind pairs, admitted use, and coverage criterion. A separate WorkPlan acceptance target may appear in designBasisRefs[] or through its direct relation; the coverage criterion remains the comparator for the coverage claim.
The notes may report represented, omitted, hidden, or unresolved candidate organization content relative to that basis. Comparison with actual framework structures starts after the framework entity and relevant structures exist.
Architecture-description and viewpoint use begins after the framework entity and relevant architecture relations exist. Later E.9 answers guided by E.4.PFAD, plus any C.32, C.30, and C.30.AD results, use their direct patterns and admission conditions; the proposal, intended-result description, and optional meta-structure keep their original types.
C.30.AD's ArchitectureDescriptionUseCard@Project remains a retrieval cue. Actual project locality additionally requires one composite project U.Work under A.15.6 when such Work is claimed. Recover every precise performer's A.13 core and independently admit the Work under A.15.1; add F.6 only when precise assignment-bound attribution is also current, and keep its description-use relation separate.
Describe authoring dependencies when a named next use needs them
The dependency description is optional, minimal, and status-bearing. It records current dependency positions while later authoring results retain their actual status:
FrameworkAuthoringDependencyDescription is a local use label for one C.2.1 episteme, and each FrameworkAuthoringDependencyPosition is a local ClaimGraph node form. Only the enclosing description is the episteme; each position is part of its ClaimGraph. The current authoring WorkPlan, one ClaimGraph, and effective ReferenceScheme supply the description's identity. ClaimScope, reader and use descriptions, optional model-use structure, empirical grounding, provenance, assessment status, publication, and edition continuity remain separate.
dependencyPatternLocator is an ordinary non-semantic PatternID that identifies the pattern whose content defines, constrains, or tests the dependency. A dependency that is also a MethodDescription identifies its episteme and the admitted Method it describes separately and applies the full A.3.2 test.
FrameworkAuthoringDependencyKindValue is fpfEdition | sourceBasis | frameworkArchitectureAnswer | nameRoute | patternDraftSet | relationAndEditionRecords | publicationOrAccess | packageQuality | improvement | currentness. FrameworkAuthoringDependencyAvailabilityValue is available | missing. FrameworkAuthoringDependencyUseRelevanceValue is currentForNextAuthoringUse | retainedForLaterUse | relevanceUnsettled.
The minimum positions are one fpfEdition and one sourceBasis. For an available fpfEdition position, dependencyValueRef identifies the selected FPF edition; the next-use boundary names the Core claims required for the declared authoring use. A change to that edition or a material change to those claims reopens the affected use.
Add another dependency kind only when the declared next authoring use relies on it or deliberately retains it for a named later use. A frameworkArchitectureAnswer position is optional: include it only when that use needs the accepted answer or its rationale, and point to the accepted answer and E.9 DRR rather than to a PFAD relation or record.
When dependencyAvailability=available, the exact dependency value and kind refs are present and the acquisition-condition description is absent. When dependencyAvailability=missing, those refs are absent and the acquisition-condition description is present.
Relevance remains independent: missing + currentForNextAuthoringUse blocks the next use and opens the stated return, while missing + retainedForLaterUse does not block current work. Record a condition on using an available dependency in the pattern that defines the dependency or in the next-use boundary, not in the acquisition position.
As authoring proceeds, a dependency description may refer to an accepted framework-architecture answer and its E.9 DRR; E.4.PFR relation records and edition dependencies; G.2 source packs; subject-home NameCards; E.8 pattern drafts; E.24.PUB, E.11, or E.17 publication and access uses; E.4.DPF.DA evaluation results; E.23 improvement results; and G.11 currentness relations.
Identify each dependency value, relation occurrence, evaluation result, edition relation, and receiving use separately, and apply the pattern that defines or constrains that object or use. The framework edition, package architecture, publication occurrence, publication form, carrier, dated authoring Work, and each dependency value retain their direct identities.
Archetypal Grounding
A hydroponic-cucumber framework begins with crop-production concerns, horticulture and greenhouse-control sources, local examples, and a dependency on a selected FPF edition for named Core claims used in its explanations and revision guidance. Its first all-in-one publication carrier is for domain users, while relation records, source packs, and quality evaluations remain separately recoverable.
A neural-network architecture framework may draw on dataflow architecture, model components, training and inference concerns, evaluation practice, and recent architecture-analysis work. The framework can describe layers, blocks, flows, optimization constraints, and interpretability concerns. For each resulting pattern, choose the smallest source route that preserves the relied claims and limits; use G.2 only when the framework needs a broad, refreshable SoTA pack and downstream Part G handoffs. Draft the pattern with E.8, and record material relations with E.4.PFR when a named use needs them.
A workspace-specific Codex process framework can contain prelanding and baton-handoff patterns. It should state its local context, the selected FPF edition and the Core claims required for its named process uses, process sources, local carriers, and refresh route. A useful local checklist keeps that status until it has source grounding, pattern bodies, direct assertions of material relations, and quality evaluation. Add a relation or edition record only when a named maintenance use needs it.
An enterprise local practice framework for architecture review starts from the organization's review setting, internal policies, proprietary examples, and approval path. It can depend on a selected FPF edition for required Core claims and on a domain-framework edition for its required content, but its confidential evidence, any local records about exact system-role classifications or assignment occurrences, training plan, and rollout telemetry stay local. Access, custody, maintenance, responsibility, authority, and approval remain separate direct claims.
Enterprise local-practice slice:
Replayable authoring slice:
Pattern-address and reorder slice
A Systems Engineering DPF edition gives SYSE.22 a stable address and places it after SYSE.2 because that order helps its readers. A later edition may move SYSE.22 without renaming it when its recurring problem and working answer continue; the ToC and body order change together, while old citations still resolve through the same PatternID.
If a later repair splits that working answer, only the continuing answer keeps SYSE.22; the other answer receives a new unused PatternID, and readers of the old reference get a short migration assertion or an explicit stop.
Local-mantra authoring slice
After the HC.NutrientMonitoring Solution is stable, its authors use the local mantra: Name the crop stage and root-zone condition; establish that the measurement is usable in its current calibration range; compare it with the stage-specific range; change the control setting only within the declared operating boundary; return when crop stage, sensor validity, or operating boundary changes. The formula helps a grower or crop-system architect keep the pattern's operative distinctions and return condition in attention. It remains Plain wording inside HC.NutrientMonitoring.
If a seminar instead needs to show alternative continuations for invalid measurement, out-of-range nutrient condition, control saturation, and crop-stage transition through one named wider unfolding structure, the authors open A.22.CGUS and build a demonstrative walkthrough. They do not obtain that structure merely by extending or repeating the local mantra.
Optional organization-proposal slice
A team intends a new clinical-method DPF, and a named review use needs candidate organization claims before an architecture answer is selected. It creates one current U.WorkPlan for possible future DPF-authoring Work, then one C.2.1 IntendedFrameworkResultDescription whose identity is its exact intended-result ClaimGraph, that WorkPlan as EntityOfConcern, and its effective ReferenceScheme; ClaimScope remains separate.
FrameworkOrganizationDesignProposal uses that description as its EntityOfConcern and proposes candidate pattern-family, dependency, publication, and access relations in one ClaimGraph. The proposal is the current result. No future framework entity, actual architecture, architecture description, dated Work, or production relation is asserted.
Coverage and acceptance slice
The proposal's medication-review coverage criterion names the pattern families whose representation is necessary for that declared use. One constraint claim node names the covered relation-family refs with exact kinds, that admitted use, and the coverage criterion. The authoring WorkPlan separately cites an acceptance target for review completion. C.33 uses the coverage node as comparator when evaluating proposal coverage; the WorkPlan target does not replace the criterion.
Empirical-grounding and use-frame stress slice
The intended-result description has a separately obtaining EpistemeEmpiricalGroundingRelation to MedicationReviewTeam@Hospital-A, an A.1-admitted holon, covering the exact supported claim subgraph. The holon is not an episteme identity slot.
A request to rely instead on a consortium first rechecks the empirical-grounding relation and evidence, effective ReferenceScheme, ClaimScope, and any independently selected BoundedModelUseStructure. Changing only the empirical ground changes that relation; changing the ClaimGraph, EntityOfConcern, or effective scheme identifies another episteme. F.9 opens only if an exact cross-context local-sense translation is current, not merely because the maintaining organization changed.
Optional authoring-dependency slice
A named next authoring use needs a stable dependency account. The selected FPF edition is available and relevant now, so its fpfEdition position has exact value and kind refs and no acquisition condition. The next-use boundary names the Core claims needed for that authoring use. The accepted architecture answer is cited through its E.9 DRR because this use needs that rationale; it is not a mandatory PFAD dependency position.
A publication carrier is missing but retained for later use, so its position has no value refs, has an acquisition-condition description, and does not block current pattern drafting. A missing source pack marked currentForNextAuthoringUse blocks the next use and opens the stated return. Availability never stands for relevance.
Framework-evolution slice
A new controlled-environment study changes the admissible nutrient range used only by HC.NutrientMonitoring. Because this example selected a broad, refreshable G.2 pack, first revise that pack and preserve the displaced source reading. Use E.4.PFR to identify the nutrient pattern, its source-reuse relation, and its dependent examples as the affected set; use E.21 to evaluate the revised pattern body; use E.23 for repeated improvement of that pattern edition; and use G.11 for currentness, telemetry, and deprecation or supersession of exposed editions.
Unaffected climate-control and harvest-feedback patterns remain current. E.4.PFAD stays closed while framework family, pattern split, relation structure, publication-form, presentation-carrier or access-route architecture, and dependency boundary remain unchanged; a change to one of those decisions makes PFAD current again.
Bias-Annotation
Scope: Use this pattern to select, author, assemble, and refresh FPF-grounded DPF or LPF editions and their first-use publication forms, presentation carriers, and access routes. Lifecycle, product-ontology, research-method, and general publication questions return to their owning patterns.
The first recurring drift is source-summary confidence: a summary feels sufficient because it names the right domain terms. Choose the smallest route that keeps the relied claims, rejected readings, limits, and source editions recoverable, then carry them into pattern Solutions and examples. Use a G.2 pack only when the question needs its broad, refreshable source frame and downstream handoffs.
The second recurring drift is publication-carrier-first authoring. Publish after the architecture decision, direct assertions of material relations, and source-return notes are recoverable, together with any relation or edition records required by a current maintenance use.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Using the exact authoring Method and MethodDescription while keeping dated Work, results, receiving uses, editions, relations, package architecture, and publication objects explicit adds overhead before a local framework becomes durable. In return, the source basis, Core change impact, relation semantics, production and membership claims, and currentness debt become reviewable at their own boundaries.
The pattern also makes local publication more useful. Readers get a coherent publication or practical-use carrier, while maintainers can still inspect the framework edition, source pack, relation records, decision records, and quality route.
Rationale
Domain and local frameworks are FPF-grounded framework editions for declared domain or local use frames. They need domain source work, FPF authoring discipline, architecture decisions, direct relation assertions, quality loops, and refresh routes. Add the relation or edition records needed for a current maintenance use; a direct assertion closes the task when no such representation is needed.
Its contribution is one framework-authoring Method plus Plain selection and branching guidance. This episteme qualifies as its A.3.2 U.MethodDescription because it describes that admitted Method as its EntityOfConcern. E.8 supplies the pattern-authoring and publication-form discipline; A.3.2 supplies MethodDescription qualification. When an A.22.CGUS is current, admit it separately with exact conditions, continuations, stops, and demonstration. Every produced or selected result needs a receiving use and the pattern that defines, constrains, or tests that result or use relation.
SoTA-Echoing
Source identity and currentness support replay and targeted refresh only. A later publication date, maintained repository, institutional status, or wider adoption cannot raise these comparisons; G.11 reopens only the smallest receiving rule when changed evidence can alter the selected answer.
Relations
- Uses for methodological construction: C.39 for general construction, connection, explanation and change; E.4.CM for applying that way to the framework's developed public account of Methods.
- Uses: direct source reliance when one source closes the question;
F.0.1for source-local meaning;F.1for a relevance-based source cut;F.0.2for one bounded semantic-synthesis result; andG.2only for the broad, refreshableCG-Framepack. Identified pack claims and provenance may enter F.0.2, but the pack does not establish its result. - Uses:
A.3.1for the framework-authoring Method andA.3.2for this independently qualified MethodDescription; A.13 for every precise performer's core and same obtaining assignment;A.15.1for independently admitted dated authoring Work;F.6only for a current precise assignment-bound attribution;A.15.PRODfor any local inception or production-completion claim;A.6.1for an actual application and its bindings;E.8,E.10, andF.18for pattern drafting, wording discipline, and names; andE.10.ARCHfor local domain wording restoration when a recurring problem has been shown. - Coordinates with:
E.4for family membership and the proportional support-unit/adjacent-product boundary, andE.4.PFADfor architecture decisions; usesC.32.MWAwhen several practice structures need one synthesis andE.23.CDIonly when capability development for a named Work family is current. - Coordinates with:
C.2.1andA.2.6for framework/result episteme identity, effective ReferenceScheme, empirical-grounding relations, and ClaimScope;A.1.1/A.22only for an independently selected model-use structure;A.22.CGUSonly for an admitted conditional unfolding;E.4.PFRfor separately identified relation records and for dependency, edition, compatibility, deprecation, and supersession effects;C.30.ADfor post-existence architecture-description use and its retrieval-only project card name; andE.24.PUBfor publication occurrence, form, and carrier. - Coordinates with:
C.33for recovery of selected architecture-relevant structure,C.34for use-bounded structure preservation, andC.35to identify the generated or discovered result, establish its kind and status, and determine admission for its intended architecture use. - Coordinates with:
E.22for quality-evaluation framing when needed,E.4.DPF.DAfor DPF package adequacy,E.21for pattern-quality evaluation,E.23for repeated improvement,E.19for selected admission or refresh questions and material-edition affected-use replay, andG.11for currentness. - Use next when current:
E.11.PFPfor the common framework publication form,E.11for practical-use discoverability,E.11.DSGfor the full architectural reference of the framework and its relations to Suite or Library accounts, andE.17for publication discoverability rather than framework authoring.
E.4.DPF:End
Evaluate a DPF or LPF Package for Its Declared Use (Package-Adequacy CharacteristicSpace)
Type: Evaluation (E) Status: Stable Normativity: Normative unless marked informative.
Problem frame
Use this pattern when a framework author, reviewer, steward, or AI agent must decide whether one Domain Principle Framework or Local Practice Framework package is good enough for one declared domain or local use.
Primary EntityOfConcern: one exact authored framework episteme edition checked for one declared package use. Name the visible publication form, presentation carrier, access-facing presentation carrier, or access route being inspected without turning it into the framework edition or package architecture.
The first useful result is one aggregate C.2.1 result episteme with all D1–D12 claims, its local status, and the smallest repair or a statement that no repair is proposed. The Solution gives the additional assurance detail only when a receiving use relies on it.
Use E.4.DPF.DA for an ordinary DPF package evaluation. E.2.DA asks whether FPF-level objects realize the FPF Pillars for broad FPF use. A DPF package must serve one declared domain or local use frame while depending on FPF Core without redefining it.
To recover that frame, identify the effective ReferenceScheme, ClaimScope, reader, intended use and qualification window. Include an independently selected BoundedModelUseStructure only when its organization changes interpretation. Add a non-use boundary only for a named competing use or plausible observed confusion. Use E.2.DA when the package changes or claims FPF-level Pillar adequacy.
Use E.21 for the quality of individual DPF pattern bodies. Use this pattern for the package as a whole: domain scope, source basis, Core dependency, the framework publication form borne by its selected carrier, pattern-set coverage, relation and edition records, local publication, evaluation route, refresh route, and adoption utility.
Problem
Authors often produce DPF packages quickly from source material, prompts, external literature, local practice, or generated candidates. Some are good enough as seeds; some can answer a domain question for an AI agent; some have publication carriers ready for public use; some are only source summaries arranged under pattern headings.
Without a DPF-specific adequacy evaluation, teams tend to rely on one of three substitutes:
- they apply
E.2.DAand ask whether the package is "FPF-like in general", even though the package is meant for one domain; - they average
E.21scores of individual patterns and miss package-level failures such as missing source packs, broken dependency direction, poor first entry, or stale edition records; - they inspect section presence and conclude that an all-in-one carrier, map, or seed package is adequate because it has patterns, a table of contents, a Readme, an architectural reference, maps, and sources.
When a team relies on these substitutes, a reader may adopt a fluent local framework that does not state its domain boundary, does not preserve rival source traditions, duplicates FPF Core ontology, hides relation functions, has no refresh route, or cannot tell a practitioner what typical problem is live, which known failure mode to avoid, and which SoTA solution move to try first.
Forces
Solution
Begin with the ordinary assessment:
- Pin the authored framework episteme edition and declared package use, then name the effective ReferenceScheme, ClaimScope, working reader, intended use, qualification window, evidence basis, floor, and any independently grounded non-use boundary that changes the result.
- Run
PFM1,PFM1a, andPFM2–PFM12where applicable and give each one a pass, fail, or not-applicable-with-reason disposition. - Judge every
D1–D12coordinate with one ordinal value, short rationale, exact evidence locus, and smallest repair or a statement that no repair is proposed. - Constitute one aggregate C.2.1 result episteme carrying those coordinate claims, protected trade-offs, the local
DPFPackageAdequacyStatus, and the first repair or no-proposal disposition. - State the next usable action, stop or repair, and reopen condition, then name any separate receiving use such as E.19 admission or refresh, assurance, publication, F.10 status use, or E.23 repair. Add a non-use statement only when a plausible reader has an independently grounded reason to confuse those uses.
The first useful result is that aggregate episteme and its local status for the declared use. Use seedOnly or repairBeforeDPFUse when the justified coordinate values support that status.
If evidence has not been inspected or cannot substantiate a coordinate value, keep the account as assessment material. State which coordinate claim remains unsubstantiated and what to examine next. Assign 0 only when its defined condition is established; a complete E.4.DPF.DA result requires all twelve justified values.
For a new or substantially revised DPF, ask four further questions:
- What work and connections does the field promise cover? Recover them from the intended practice and its sources. Follow the first use and a representative use that needs a connection between problem families. The same use may serve both purposes. Which obligations do their grounds cover, individually or together? For the rest, identify what the intended reader can obtain and use from the edition or its suppliers under the promised conditions. Record what is established, what the evidence rules out and what is still unknown about coverage of the field (
D12DomainProblemFamilyCoverageAdequacy). Preserve any first use the evidence supports. - Recover the relations on which the domain practice depends. Does the answer apply at each relevant composition scale, carry from one model to another, or depend on several contributions working together? Check shared conditions, including order or simultaneity when timing affects the result. Reuse a completed
C.32.MWAresult where structures must be reconciled andE.23.CDIwhere capability development affects the claim. - For the named first use, are all required patterns from this DPF present? For every relied-on external result, are its identity, direct kind, supplying product and edition or current state, receiving use, discovery route, material currentness or availability, and externality explicit? For each important source-backed claim, can a reviewer find the source and tell whether the evidence supports, suggests, or only motivates it?
- When the declared use includes a public presentation carrier, does that carrier bear the framework publication form defined by
E.11.PFPwhile keeping its DPF-specific body and references underE.4.DPF? KeepPFM1responsible for practitioner entry and navigation; usePFM12only for the remaining common-form and edition-projection questions. Form conformance does not prove field coverage or package adequacy.
These questions test the package and its evidence. They do not prescribe another Method to perform or turn use of a Method result into an edition dependency.
Assurance and object boundary after the ordinary route. Evaluate one exact authored framework episteme edition for one declared package use through a DPF-specific adequacy characteristic space. The evaluation takes its form from E.2.DA. Its subject is the DPF or LPF edition.
The evaluation asks whether the selected framework edition, together with its separate package architecture, pattern set, source basis, architecture decisions, relation records, edition dependencies, publication and access uses, quality evidence, and refresh route, realizes FPF-grounded domain value for one declared use frame.
Keep the evaluation objects separate:
- the authored framework episteme edition of concern, identified under C.2.1;
- its package architecture, E.4.PFAD architecture decisions, E.4.PFR relation records and edition dependencies, selected pattern set, source-use results, publication units and occurrences, publication forms, presentation carriers, access-facing presentation carriers, access routes, and actual access or use relations;
- the effective interpretation rules (
ReferenceScheme) for the package-adequacy claims, A.2.6ClaimScope, working reader, intended use, qualification window, any independently grounded non-use boundary, and an optional independently selectedBoundedModelUseStructureonly when its organization changes interpretation; - this E.4.DPF.DA characteristic space and evaluation specification;
- one exact semantic package-adequacy-evaluation
U.Method; - an ordinary evaluator action outside Work admission or, when dated assessment
U.Workis asserted, references to the actual evaluator System identified through A.13 and one independently valid A.15.1 Work account. Only when the result expressly represents precise assignment-bound attribution does it also cite the same obtaining A.13 assignment and applicable F.6 relation occurrences. Independently, cite an A.6.1 application only when the assessment uses an identified operation declared by a separately admitted Mechanism and the receiving claim depends on its bindings; - twelve ordinal coordinate-result claims about the same framework edition;
- one aggregate C.2.1 result episteme carrying those claims, the local package-adequacy status, protected trade-offs, first repair or no-proposal disposition, reopen condition, and any grounded non-use boundary;
- witnesses, independently established direct evidence-use relations, and any A.10 account that represents their source and bounded-reliance basis, plus an optional evaluation record that packages references without performing the assessment or granting authority; and
- any F.10 status use, E.19 admission or refresh decision, assurance, publication, later improvement Work, and changed framework edition.
In the configuration and coordinate-result fields below, EffectiveReferenceScheme carries an exact C.2.1:4.1.2 EpistemeInterpretationScheme value. Its rules and domains must interpret the selected evaluation claims, including their scale and evaluation-rule meanings. The framework edition and cited sources retain their schemes; recover any correspondence the evaluation uses. Scheme membership establishes neither a justified coordinate value nor an actual assessment or operation application.
Use this compact input/action/result separation:
These names are local record and claim shapes, not new U-kinds. DeclaredVisiblePackageFormOrUse is open plain wording for the exact form or use being checked; it neither types nor identifies the framework or package.
The separately typed reference fields keep publication units, particular expressions, reusable forms, independently identified carriers with actual expression-bearing relations, access routes, and actual access or use relations distinct.
If the visible material has no independently admitted single package entity, do not make a file set or list into one: keep the exact framework episteme edition as EntityOfConcern and cite its package architecture, records, contents, publication and access relations, forms, carriers, and routes separately in the configuration and evidence basis. A file boundary, manifest, directory, table order, publication, carrier, callable service, or endpoint establishes neither package architecture nor membership.
The characteristic table and this specification describe how to evaluate; the semantic Method, evaluator action, dated Work, A.6.1 application, and result keep their identities. A practitioner may make an ordinary package-adequacy judgement without classifying it as U.Work or asserting an A.6.1 application.
Such an application exists here only when the evaluator uses an identified operation declared by a separately admitted Mechanism and the receiving claim depends on its bindings. Work admission neither creates nor requires it.
If the account instead claims dated assessment Work, recover the actual evaluator System through A.13 and cite one independently valid A.15.1 Work account. Only when the result expressly represents precise assignment-bound attribution does it also cite the same obtaining A.13 assignment and applicable F.6 relation occurrences.
F.6 identifies neither assignment nor performer. Missing or failed F.6 alone leaves independent Work admission intact. If the same finding invalidates a premise consumed by A.13 or A.15.1 admission, recheck that premise and its dependent conclusions.
The evaluator System acts. A local evaluator system-role classification is an optional neighboring claim. The local account exposes assignment or F.6 references only for that attribution branch. Constitute the coordinate claims and aggregate result under §4.3, and establish each later receiving use through its own relation under §4.5.
Each coordinate value is an ordinal content-evaluation quality ascription about the same framework episteme edition under the declared ReferenceScheme, ClaimScope, use, and qualification window. It is not a C.16 measurement result, measurement output, average, vote, maturity stage, or status use.
The aggregate result episteme has its own C.2.1 identity; empirical grounding, witness presence, evidence use, publication, and evaluator identity remain neighboring relations or objects rather than identity slots.
Ordinal scale
Default floor is 4 for public, teaching, enterprise, operational, or reliance-bearing DPF use. A fast seed or exploratory prompt output may use floor 3 only when its limits, missing evidence, next repair, and reopen condition are explicit.
Required coordinates
Every E.4.DPF.DA result includes every coordinate below, including a result for a seed: assign the value that the seed earns. A bounded diagnostic may borrow selected questions while stating its limited scope; it does not claim an E.4.DPF.DA result or local status.
In this pattern, known failure modes means beginner mistakes and experienced-practitioner failures caused by stale, local-only, or non-SoTA practice. The check covers both groups.
For D12, recover the field account required by E.4.DPF:4.0, and E.4.DPF:4.0.1 for a Suite, before treating the selected patterns or their worked cases as the coverage boundary. Compare the promised subject with the public explanation and the actual supplying contributions. Follow a consequential question about how its units, relations or operations are obtained or revised, and a change that can alter the choice or connection of Methods. Reuse adequate existing evidence.
Distinguish declared preparation and a reader's ordinary construction by a supplied Method from an explanation or operation that the reader had to invent to supply the framework's promise.
For example, the package can support delivery of an approved report while a dispute about that report calls for an authorised arbiter. If a timely decision from the arbiter is available to that reader and the order is to obtain it, the external result can suffice. An order to design the arbitration method leaves a different choice with the reader. Where the agreement requires the arbiter's decision, an instruction to accept any disputed report as submitted permits a decision the agreement forbids. This establishes a defect in the procedure. To establish what happened in one dispute, use evidence for that dispute. A decision already issued and usable in time can suffice even if the arbiter is away today. An order for a new hearing requires provision for that hearing. Where the evidence leaves open whether the required decision can reach the reader in time, keep that use open. One established failure of a required use rules out coverage of the whole promised field even while another use is still unexplored. Keep whatever the grounds establish about the original field promise and about independently usable parts.
Retain successful bounded uses while returning a missing field-level contribution. This judgement concerns what the current edition explains and enables; it neither demands an exhaustive catalogue nor follows from having enough examples.
Result row shape
The aggregate E.4.DPF.DA result episteme carries twelve coordinate-result claims and may present them with this table shape:
For values 1..4, explain why the lower adjacent value would understate the evidence and the higher adjacent value would overstate it. For 0, explain why 1 would overstate the evidence and what would raise the value or reopen it. For 5, explain why 4 would understate the evidence and what would lower the value or reopen it.
Each coordinate claim is one ordinal content-evaluation quality ascription about the same framework episteme edition and keeps recoverable the effective ReferenceScheme, characteristic, scale value, evaluation rule or probe, ClaimScope/use/window, assessment account and any asserted Mechanism-operation application, short rationale, evidence locus, and repair or no-proposal.
Prose or a table can carry the complete twelve claims and their aggregate C.2.1 result. A bare verdict, checklist count, unsupported table, average of E.21 pattern values, or favorable status label does not supply that content. The aggregate is distinct from a C.16 measurement result, performed assessment, admission, and authority.
DPF-wide package-form checks
Run this subpass when the declared use depends on an all-in-one DPF publication carrier, selected-host set, card set, skill-pack or index carrier, returned response artifact, MCP or other service route, retrieval or search route, assistant integration, or another reader-facing form.
Inspect the particular publication expression, its selected reusable form and conformance when applicable, and the independently identified physical or digital carrier whose actual state or realization yields that expression. Inspect any service or route separately. Do not substitute editable sources, a manifest, or a successful build run.
These checks do not replace the twelve coordinates; they supply package-level evidence mainly for D1, D2, D4, D5, D7, D8, D9, D10, D11, and D12.
When the declared package use includes accepted-source integration or continuity with a predecessor publication, use a current E.4.PFIP conclusion as evidence for the affected coordinates. Keep the PFIP conclusion and the package-adequacy result separate: the first reports publication integration or continuity, and the second judges package adequacy for the declared use.
PFM1 owns practitioner entry, navigation, and the publication placement of the ToC, Readme and full architectural reference; it does not prescribe a reader's use order.
PFM1a owns the product-native key/form declaration, the return from an entry to the reader’s question under E.11:4.1 and any warranted coverage correction, and the content judgement that a mantra materially improves each selected cross-pattern card while a plausible direct example remains sufficient without one.
PFM12 owns the remaining common-form and edition-projection agreement.
If one observation bears on PFM1, PFM7, or PFM12, record it once, point every affected disposition to the same evidence and repair, and lower an affected coordinate only once for that defect. Give the checks different dispositions only when they discriminate different defects with different repair actions.
A failure in this subpass lowers the affected coordinate even when individual pattern bodies pass E.21. Repair the package carrier, relation record, first-entry route, dependency record, or support-map placement; do not copy the package-form proof into pattern bodies.
Evidence basis and where to check it
Use these sources and patterns to inspect the evidence needed for the declared package use:
When a coordinate is below floor, return a finding or repair proposal. When a coordinate is at 4 and improvement is requested, search for a substantive non-dominated improvement.
Do not raise a value by adding proof apparatus, more maps, more citations, or quality-status prose unless the package becomes easier to use, more source-grounded, more accurately bounded, or more refreshable.
Local result status and receiving-use boundary
DPFPackageAdequacyStatus is a local admissible-use claim carried by the aggregate result episteme. It reports the package evaluation result for the declared scope.
A receiving process may use it for an F.10 status use, E.19 admission or refresh decision, assurance, publication, work authorization, or improvement Work only through that use's relation and decision rule.
Archetypal Grounding
A personal-development DPF is generated in one short run. It may have useful principles and pattern seeds. The evaluator can make an ordinary package-adequacy judgement without asserting U.Work.
If a dated evaluation is instead admitted as Work, recover the actual evaluator System through A.13 and cite one independently valid A.15.1 Work account. Add the same obtaining A.13 assignment and applicable F.6 relation occurrences only when this result expressly represents precise assignment-bound attribution; F.6 identifies neither assignment nor performer, and missing or failed F.6 alone leaves independent Work admission intact.
Separately cite an A.6.1 application only if the evaluation uses one exact operation declared by a separately admitted Mechanism and the result depends on its bindings; Work admission does not supply that application.
The aggregate C.2.1 result can then state local status seedOnly, with high coordinate claims for first-entry utility and low claims for source currentness, heterogeneous probes, relation records, or refresh. The prompt output, evaluator action, any Work, any attribution, any Mechanism-operation application, result episteme, and local status remain distinct. The status gives the package-use result and points to the next repair. It does not itself admit the package.
A domain DPF all-in-one publication carrier contains domain patterns, a source-use map, a Core-bridge map, relation records, and heterogeneous acceptance cases for several user situations. E.4.DPF.DA asks whether those cases require the pattern set to solve different domain problems, whether source rows changed pattern obligations, whether maps are reachable during work, and whether the package's local evaluation pattern can feed E.22 and E.23 without becoming a hidden Core dependency.
A proposed systems-management DPF promises help with service launch, cross-team coordination, incident response, and feedback-based improvement. D12 checks whether the selected pattern sets and their actual relations serve all four problem families, whether one service-launch case needs patterns from more than one set, what remains omitted, and where later authors return to the sources.
One source describes an incident flow in which one operation follows another. Another describes monitoring, coordination, and resource provision contributing at the same time. The evaluation preserves both readings. A completed C.32.MWA result may supply that evidence, but the evaluator neither repeats the Method's actions nor treats use of its result as an edition dependency.
The named first use must include every required pattern from this DPF. For each relied-on external result, the assessment names its actual kind and supplying product, the receiving use and discovery route, material currentness or availability, and the fact that it remains external.
PFM11 then checks whether the carrier tells readers what it exposes and omits. PFM1 checks practitioner entry and navigation; PFM12 checks only the remaining common-form and edition-projection agreement. A shared observation and repair are recorded once. None of these form checks substitutes for D12.
A hydroponic-cucumber DPF has excellent crop-control sources but no relation records and no first-entry carrier. E.21 may find that individual crop patterns are good, while D2DidacticEntryAndAdoptionAdequacy stays below floor until a usable first-entry route exists. D5PackageFormLayeringAndRelationAdequacy falls only when the needed relations cannot be recovered or a named maintenance use needs the missing records; direct assertions can otherwise supply those relations.
Near miss: A DPF all-in-one publication carrier puts a large lookup map before the pattern bodies. The map is correct but cold readers do not know when to open it. Place the lookup material after the bodies. Link to it from pattern relations, low-value repair actions or first-entry text at the point where readers need it.
A full architectural reference explaining the Methods, shared conditions, sources and worked connections has a different job: it remains in the front matter and may include the figures needed for that explanation. PFM2 distinguishes those functions rather than rejecting the account for its length.
Near miss: A DPF has polished Readme and architectural-reference prose, but neither says what selected domain structure the publication/access expression exposes, what it deliberately coarsens or abstracts, or where a reader returns for fuller source and pattern detail.
If the carrier is based on an architecture description, view, model, or graph, it also hides the fact that the intermediate source already selected and coarsened structure on the route source structures -> architecture -> architecture description or view -> publication/access expression. D1, D2, D5, D7, D8, and D11 fall because the carrier may be pleasant but its structure-capture claim is not inspectable.
Whole-account calibration: missing condition, false relation, and expert recovery
A constructed observation-planning profile offers three contributions: P observes current representative performance, U examines adaptation to unfamiliar work, and D examines delayed performance under recorded practice and support conditions. They answer different questions. A qualified earlier observation may supply an input to another question, but the three contributions are not mandatory stages.
The promised first use is to select a supported observation set under the actual common resource condition, explain how the contributions relate, and find a direct return for a narrower question. Each activity requires seven observer-hours that cannot be shared. Individual and pair feasibility is stipulated.
The promised planning result, including the common resource limit, determines what the assessment must cover. The profile’s headings do not set that scope.
Three accounts present the same substantive headings and direct source returns. A is compact but omits the common resource limit. B supplies the eighteen-hour limit and explains its whole-combination consequence. C is more expansive but retains A's omission. The useful comparison concerns what the reader can decide from the available account.
Separate prepared-agent readers of A, B, and C, without the authors' private context, returned these first and changed-condition results without an intervening corrective hint. They contributed arithmetic and ordinary logic from their preparation and found the public source returns.
A/C readers did not fail to understand a supplied relation: the decisive common fact was absent. The bounded result therefore supports the omission diagnosis, useful direct entries, and changed-condition planning. It supplies no human learning, reading-time, or cognitive-load comparison.
A separate contrast holds the facts fixed: eighteen available hours, seven non-shareable hours per activity, and individual and pair feasibility. Two defective accounts now state that feasible pairs suffice to promise all three. The facts are available here. The claim about combining the three activities is false. The adequate B account stays unchanged.
Fresh prepared readers rejected the false conclusion and returned 21 > 18 and 14 ≤ 18. They explained how they supplied the correction from prior knowledge.
Their numerical answers match the adequate-account reader's answer, but the material contributions differ: B supplies the valid whole-set relation; the defective accounts supply a relation their readers must correct. Inspecting the answer alone would hide that defect.
The reader's source-use explanation, checked against the actual account, makes it visible. This is expert recovery despite a false explanation, not evidence that the explanation is adequate for a less-prepared audience.
Use the example to answer the existing coordinate questions:
- D2 asks whether the reader can reach the promised first result from the public account and its usable returns.
- D7 asks whether the account changes the actual planning decision or yields the precise missing input or repair.
- D8 asks what the changed condition and contrasting failures reveal within the tested breadth.
- D12 asks whether the selected contributions work together for the public promise, including common constraints and important omissions.
These are bounded diagnostic contributions, not a complete D1–D12 evaluation or a local package status. A complete evaluation follows this pattern's specification. The example neither adds a coordinate nor fixes a universal number of accounts, tasks, or readers.
For the planning use in this example, a professional reference requires a usable explanation of how the contributions combine. It does not require the framework to become an instructional course or depend on the domain profile used to illustrate that combination.
Bias-Annotation
Scope: Limited to evaluating one exact FPF-grounded DPF or LPF edition for one declared package use. It is not a whole-FPF evaluation, a universal product score, an admission decision, or a publication template.
Five errors recur in package evaluation. The first is whole-FPF overreach: a DPF package is judged as if it had to cover every domain. Declare one domain or local setting and evaluate adequacy for that setting.
A second error is to judge the package from fluent text or good individual patterns while overlooking missing source, relation, edition and refresh structures. Evaluate the package coordinates.
The third is quality-proof leakage: evaluation results, review status, or package-architecture development evidence are copied into user-facing pattern prose. Move that evidence to this evaluation's result, E.21, E.19, E.11, or the applicable publication-evidence locus, and keep the user-facing move, boundary, and architectural reasons needed to understand, select, combine, or adapt the pattern in its body.
A fourth error is to present the package as a transparent list of principles without asking which domain structures were selected, coarsened, abstracted, omitted, or already transformed through source structures -> architecture -> architecture description or view -> publication/access expression before the publication carrier was written. Make the architectural reference explain that selection and loss, with the necessary entry cue or access return, and check it through PFM11.
The fifth is assurance by duplication: the evaluation copies the action sequence of C.32.MWA or E.23.CDI and treats completion as package proof. Use the completed result only where it bears on a named coordinate, then run the package's own probes.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Package evaluation examines connections and field coverage that individual pattern checks cannot establish. Its cost includes recovering and assessing the needed evidence; reuse adequate results and scope additional work to the package judgement being sought.
A DPF seed can be useful without being ready for public use. Readiness for public use does not extend the scope of a domain package to FPF Core.
D12 adds one field-coverage judgement and PFM12 one incremental publication-form check. Neither starts a second evaluation pass, copies another Method's steps, or charges a PFM1 observation twice.
Rationale
A local edit can improve one pattern while harming the whole language, which E.2.DA evaluates. DPF packages need an analogous evaluation at their own scope: a package can have good patterns while failing as a domain framework. Domain source grounding, relation architecture, first-entry adoption, package publication, and refresh are package-level effects.
FPF asks whether the whole framework remains broadly first-principle and cross-domain. A DPF asks whether one bounded domain or local framework is strong enough for its declared use while preserving dependency on FPF Core and a route back to its domain sources. D12 makes that field-scale question explicit. Follow the field practice to establish the relations a diagram or chapter sequence represents, including timing when the result depends on it.
SoTA-Echoing
Currentness checks remain separate from adequacy values. G.11 reopens only the affected comparison, coordinate, case, or boundary when changed evidence can alter the answer; a new edition, current catalogue entry, maintained tool, or recent paper alone changes no value.
Relations
- Builds on:
A.19.ECSfor evaluation characteristic-space construction. - Specializes by object:
E.2.DAsupplies the adjacent form for complete multi-coordinate adequacy evaluation, but this pattern changes the evaluated object from FPF-level object to DPF package edition. - Coordinates with:
E.4for framework family; the E.4.DPF authoringU.Methodand MethodDescription;E.4.PFADfor architecture decisions;E.4.PFRfor relation records and edition dependencies; and the separate package architecture. Their coordination list and document order identify neither the authoring Method nor the evaluation Method. - Coordinates with:
C.2.1for framework and result episteme identity and the effective ReferenceScheme;A.2.6for ClaimScope;A.1.1andA.22only for an interpretation-changing selected model-use structure;A.3.1andA.3.2for the semantic evaluation Method and its description;A.13for the actual evaluator andA.15.1for one independently valid Work account when dated assessment Work is asserted;A.2.1andF.6only when the result expressly represents precise assignment-bound attribution through that same obtaining A.13 assignment;A.6.1only for an actual application of an operation declared by a separately admitted Mechanism and its bindings;A.10for evidence use; andG.2andG.11for source packs, source currentness, and refresh. - Coordinates with:
E.21,E.19,E.22, andE.23for individual pattern quality, admission review, evaluation framing, and repeated improvement. - Uses:
F.19for precise plain-language checks and ordinary wording repair;E.10for cues and unresolved meaning routes. - Coordinates with:
E.24.PUBfor publication occurrence, form, and presentation carrier;E.11.PFPfor the common framework publication form;E.4.PFIPonly for a separate accepted-source integration or predecessor-continuity conclusion; andE.11,E.17,F.18,C.33,C.34, andC.35for first entry, publication or access use, naming, preservation, correspondence, and generated- or discovered-result admission to an intended architecture use. - Coordinates with:
B.1.5,A.22,A.22.CGUS, andC.30.ADfor the relations that support the practice. Use a completedC.32.MWAresult where practice structures need reconciliation andE.23.CDIwhere capability development affects a named coordinate. - Coordinates with:
E.2.DAonly when a DPF package change claims FPF-level Pillar adequacy or proposes Core amendment effects.
E.4.DPF.DA:End
Pattern-Framework Relation and Edition Discipline
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative.
Use this when a named framework-maintenance, edition-impact, comparison, publication/dependency-repair, or refresh task needs a stable row for a specific relation across patterns, framework editions, publication or access carriers, source packs, decisions, generated carriers, or quality results.
Begin with the assertion. State the subject assertion in ordinary C.2.1 form. Name the subject or claim, identify the relation function and the ClaimGraph that defines or constrains it, and give the polarity and current fact or condition. Stop there unless an identified maintainer or tool consumes standardized relation form.
Primary object. One already identified subject assertion, optionally represented by one PatternFrameworkRelationRecord@Context for a named framework-maintenance use.
The assertion, relation row, pattern description, relation kind or occurrence, framework edition, publication occurrence, form, carrier, access route, source use, Work, evidence, assurance, and currentness result remain distinct.
Primary reader. A framework author or maintainer who must state one relation or edition claim now and decide whether a named maintenance use justifies a reusable row. A tool may consume the row. That fact alone does not make the tool an actor in the subject claim represented by the row.
Ordinary authoring stays light. Edition and framework-maintenance consumers can compare relation functions, inspect compatibility and dependency effects, preserve blocked stronger readings, and reopen only affected uses.
When to stop or use another method. If a readable subject assertion closes the task, use C.2.1 and stop.
Use E.11.PUR for pattern-use recommendations, E.17 and E.24.PUB for publication, C.33 and C.34 for selected-structure capture and preservation, C.35 for exact-result admission to an intended architecture use, and the exact subject pattern for the direct relation.
For direct source reliance or bounded source selection and synthesis, use the source route in E.4.DPF:4. Use G.2 only for a broad, refreshable CG-Frame pack.
E.4.PFR does not define a generic governance relation, pattern owner, mandatory relation-record layer, workflow, runtime route, API call, build dependency, or performed Work.
Problem frame
Pattern frameworks need several relation functions. One pattern may specialize another. A local framework edition may depend on a domain-framework edition or selected Core content in an FPF edition. A publication occurrence may expose a selected set through a carrier. A skill pack or MCP-backed service may provide access to that set. A generated graph may suggest candidates. A quality result may evaluate a pattern version.
Those claims differ in subject, predicate, identity, use, evidence, and change behavior.
Problem
Two opposite failures are common. Flattening everything into "related patterns" or "dependency" hides relation function and change effect. Requiring a PFR row for every load-bearing relation turns a readable assertion into a record-first ontology and encourages the mistaken claim that one pattern contains the defining content for the other content or governs it.
Forces
Solution
Select the lightest lane that changes the named receiver's next action. Extra structure alone does not make a representation better.
Lane 1: ordinary subject assertion
Name the subject or claim, identify the relation function and the ClaimGraph that defines or constrains it, and state the assertion’s polarity and the current facts or constituting history.
A pattern id, heading, field, file, or carrier may locate that ClaimGraph but does not own the subject and creates no governance relation.
If no actual formal-premise use or criterion selection is claimed, omit Lane 3. In that case, stop after the subject assertion unless a named maintenance receiver needs Lane 2.
Create no PFR row, actual-use predicate assertion, candidate universe, basis analysis, scope/time placeholder, edition pin, witness wrapper, evidence or assurance result, accepted-use record, or relation occurrence merely to make the sentence look complete.
Lane 2: optional relation-specific maintenance row
Choose the smallest stable form the named receiver consumes.
Use PatternFrameworkRelationRecord when a cross-relation comparison or maintenance index needs common endpoint, relation-function and use fields.
Use FrameworkEditionDependencyRecord when an edition-impact or refresh receiver needs the relied-on content, dependency reason, direction, and refresh fields.
Choose one by default. Either form faithfully represents an already identified assertion and creates no relation.
Choose each view for the task that consumes it. One practitioner may need both views, or different practitioners may need one each. Both cite the same subjectAssertionRef and derive their overlapping values from it. Use genericRelationRecordRef when the consumer needs to follow the dependency record to the generic row.
Rebuild both views when the assertion changes. Never maintain duplicate facts independently.
relationFunctionClaimRef, when present, resolves the defining or constraining ClaimGraph used by the subject assertion. It is not an owner field, pattern-authority assertion, provenance claim, or actual-use predicate.
subjectAssertionRef is present only when the receiver needs stable reference to the C.2.1 assertion that the row represents.
Source and target remain the endpoints for this relation function. Row order creates no direction.
Dependency, specialization, publication, source reuse, quality, access, preservation, admission, and refresh retain their existing semantics. A PFR row translates none into derivation, evaluation, evidence, assurance, permission, authority, Work, or relation occurrence.
Use these companion forms only for their named maintenance receivers:
blockedStrongerReading and blockedRuntimeOrBuildReading are optional. Include a rejected reading only when it passes F.19:4.4's independent-ground, plausible-reader and contribution test, and state the smallest clear correction. Leave the field absent when no such correction is needed. A tabular consumer may keep an empty column.
The dependency record mirrors one already stated direct dependency and cites that assertion through subjectAssertionRef. It names one dependent edition, one relied-on edition, the exact content from that relied-on edition, the named use, direction, reason, and refresh conditions as one unit.
If one edition has several direct dependencies, write one record per relied-on edition or use a keyed collection of those records. Never pair parallel edition and content lists or read them as a cross-product. A useful aggregate is only a projection over those direct records, not a second maintained truth.
The record contains no compatibility boundary. compatibilityClaimRefs, when present, points only to a separately stated pairwise compatibility claim because a named maintenance receiver needs that connection. The reference does not create or complete the compatibility claim.
Populate genericRelationRecordRef when a receiving task requires a link from the dependency record to the generic row. When both views exist, derive their overlapping values from subjectAssertionRef and refresh them together.
Deprecation and supersession likewise remain separate assertions. A package manifest may index them when its named maintenance use needs those refs.
The manifest is a package-like index for a domain principle framework or local practice framework when authors need one. It indexes whichever generic relation rows or dependency-specific records its named operation consumes. Either list may be empty, and one indexed form never requires its duplicate. When the operation needs both linked views, the manifest may index both without making either a second semantic source.
The form of FPF uses E.4.FPF and its FPFEditionRebuildabilityRecord. A manifest entry, relation row, identifier, citation, or file path creates neither the referenced object nor any relation.
Relation functions keep their own semantics
The direct assertions, not a PFAD or PFR row, state the architecture-bearing relation facts. The E.9 DRR records their selection and rationale. E.4.PFAD adds no relation or second decision result.
An optional PFR row may point to the assertion it represents for maintenance, but it neither creates that relation nor becomes a condition for accepting the answer.
There is no Subject-pattern relation.
When earlier prose says that one pattern contains the defining content for or governs a value, claim, boundary, relation, record form, or use, recover the subject assertion and identify its relation function. Preserve genuine non-pattern ownership, legal or institutional authority, source attribution, evidence, and other direct relations.
Edition and package discipline
A local practice framework may depend directly on a domain framework, on Core claims in an FPF edition, or on both. A domain framework may depend on Core claims. Identify each supplier for the work that actually uses it.
Identify that FPF edition and its required Core claims separately, as E.4 specifies. E.4.FPF governs the FPF edition.
Core does not depend on domain or local frameworks. Incorporate an accepted transdisciplinary contribution through a deliberate Core amendment whose content no longer depends on those frameworks.
Framework-edition dependency obtains for one dependent edition, one relied-on edition, exact content in the relied-on edition, and one named use only when the dependent edition's current content or result for that use requires the relied-on content. Removing it or changing it in a way relevant to the use would invalidate the dependent content/result or require that use to be reopened.
Name the decision, calculation or other use that requires the supplying claim. In a chain of editions, trace the particular content through each use. If an intermediate edition uses the changed rule only for a different operation, the chain alone gives no reason to revise the final user's operation. If the intermediate result needed by that user relies on the rule, check whether the result still serves that use. Honor any separate commitment to check a new edition when adopting it.
Several grounds may support the same decision. Establish which grounds must hold together and which can each support the decision on their own. If either of two accepted routes suffices, losing one leaves the result available through the other. Losing both reopens it. Trace any premise they share. A proposed substitute becomes a ground for the decision once it is qualified for this use and in effect. A separate promise of conformance to a named edition still requires checking that edition.
Domain@Duses selected Core relation semantics inFPF@Cas required constraints on framework review. Without those semantics, or after a relevant change to them, the affected review guidance cannot remain current without recheck.Domain@Dtherefore depends onFPF@Cfor that content and use. The edition labels are illustrative. An actual assertion identifies the claims used, as in §4.2.
E.5.3 tests the permitted direction and acyclicity of the identified imports. G.11 governs the edition pin, currentness and refresh condition. A permitted direction, joint publication or a shared use of two editions is insufficient to establish a dependency.
Compatibility answers whether the two editions in the pair can support an overlapping use despite a stated difference or interface. State it separately and only when current:
Domain@DandFPF@Care compatible for framework review across relation-semantics interface I. Difference X changes no admitted review operation within boundary B; reopen when I, X, B, or either edition changes.
If that basis is insufficient, state the unresolved pair, overlap, or impact and make no positive compatibility claim. A dependency record may cite the independently stated compatibility claim only when a named maintenance consumer needs the link.
Both claims may obtain for the same pair. Neither is shorthand for the other. Deprecation and supersession are also separate claims and are indexed only when current.
Do not import binary compatibility, runtime import, build, module-call, API permission, or performed-work semantics. An edition label alone establishes no dependency, compatibility, deprecation, supersession, or refresh result.
Keep these neighboring objects distinct:
- a selected pattern-set result and its stable public identity when one is needed;
- a publication occurrence and an access carrier;
- a relation row, a dependency pin, and an edition status;
- a source-pack pin, a quality result, a refresh plan, and a first-entry carrier.
They are not members of one type. Listing an access means—for example, a skill package, MCP endpoint, API route, or assistant integration—records only the exact access claim that obtains. It creates no framework dependency, method order, tool permission, or authority.
Use G.5 for selector-facing set-result declaration, E.17 for a source-backed publication face and return to source, E.24.PUB for the publication occurrence and audience availability, G.11 for currentness, and C.33 or C.34 when a carrier is used as architecture or preservation evidence.
Lane 3: exact actual rule-content use
Use derivedUsingRuleContent(dependentContent, baseContent) only when one identified derivation claim used the exact nonempty base subgraph as a formal premise under a declared inference rule or application to produce the exact dependent ClaimGraph.
Use evaluatedAgainstRuleContent(dependentContent, baseContent) only when one identified criterion-selection claim selected the exact base for one bounded evaluation claim concerning that dependent ClaimGraph.
RuleContentBasisFindingDefinition@R7 declares both predicates. Definition, constraint, applicability, consultation, influence, source use, evidence, evaluation Work, result, sufficiency, assurance, reliance, permission, and publication are independent.
Each positive or negative actual-use assertion is an ordinary C.2.1 episteme. It names exact subject S, dependent graph U, base subgraph B, mode, exact derivation or evaluation-and-selection claim identity, bounded receiving use, effective scheme, and only independently current scope, time, interpretation, source, or witness qualifications.
A same-scheme use invents no Bridge. The serialized form is a representation, not a relation occurrence or new kind.
Optional high-cost basis analysis
Open a basis analysis only for a named automated candidate comparison, reproducible cross-edition replay, same-subject conflict whose resolution can change the cell disposition or named receiver action, or bounded reliance/assurance receiver.
One analysis is one C.2.1 episteme identified by <AnalysisClaimGraph, BasisAnalysisQuestion@QGroup, effective ReferenceScheme>.
The question includes every independently current discriminator: S, U, derive/evaluate mode, bounded receiving use, the identities of the actual-use claims, the effective scheme, optional exact ClaimScope, and the exact temporal-policy branch. Include the receiving edition whenever changing it can change candidate applicability, the cell disposition, or the named receiver’s action.
The analysis ClaimGraph carries a finite candidate universe containing only bases whose inclusion or exclusion can change the cell disposition or named receiver action. It carries a closure claim only when the enumeration rule, source boundary, completeness evidence, qualification window, and exclusion argument are exact.
Each candidate is a finite nonempty set of semantic-base subgraphs used conjunctively. Each CandidateEvaluation keeps exactness, applicability, acceptance, witness, sufficiency and minimality independent, with supporting claim refs and a reconsideration condition for every unresolved or negative axis.
Establishment follows RuleContentBasisFamilyAlgebra@R8 in A.6.0 §4.6b. Exactness, applicability and sufficiency are required. Acceptance and witness are required only by the receiving contract. The candidate is established when all required axes are true. Any false required axis defeats it even if another is unknown. If no required axis is false and at least one is unknown, establishment is unresolved.
Optional minimality is a separate answer and never filters a sufficient candidate. Duplicate graphs under one scheme collapse to one semantic atom while retaining source qualifications. Independently sufficient bases remain separate alternatives; jointly necessary bases remain one conjunctive alternative.
Derive-mode sufficiency requires derivation of the exact proposition under the named inference rules without undeclared premises. Evaluate-mode sufficiency requires the criteria and facts needed to obtain the bounded result under the named evaluation rule, including a negative result.
These candidate tests establish neither R7 actual-premise use nor actual criterion selection. A minimal-family request adds a separately qualified minimality answer.
Compatibility is pairwise, not a candidate property. Every overlapping established pair whose resolution can change the cell disposition or named receiver action receives an exact result naming both alternatives, overlap, supporting claims, and, when conflicting, incompatible consequences plus a bounded E.9 decision.
Candidate axes, pairs, conflicts, and receiving-edition distinctions enter the analysis only under that same effect test.
The temporal partition is maximal and non-overlapping under the selected policy and this candidate set. A no-time-dependence policy yields one atemporal cell. Changes to scope, temporal policy, candidate inclusion, or applicability reopen only the assertions and cells whose disposition or named receiver action can change.
Apply conflict precedence, then exactly one disposition in each cell:
An unknown axis on a defeated candidate does not make establishment unresolved. Unknown optional minimality does not open the family disposition.
The compatibility conclusion concerns the required pairs. A further claim that three or more conditions can hold in one realization needs grounds for the joint realization. With no required pair, all required pairs are compatible by vacuous truth. The closed-empty branches differ in the supported need claim, not in how many candidates were found.
The basis answer is non-permissive. A downstream A.10 bounded-reliance claim or B.3 assurance result cites the exact analysis edition or cell-answer subgraph and supplies the evidence, attempted use and disposition required by its own rule. Include currentness and any live rival when they change that use or conclusion.
Neither the bounded-reliance claim nor the assurance result grants permission, gate passage, decision, Work, actual use, publication, or authority. When a reverse consumer lookup is needed, derive it without inserting it into the upstream ClaimGraph.
Bootstrap and stopping rule
A direct C.2.1 assertion always precedes optional PFR representation. The selected generic row or dependency-specific record cites the assertion and its exact defining or constraining ClaimGraph only when the named maintenance receiver needs that form.
Neither representation can provide circular evidence for its own semantics.
After each assertion, ask which named next action consumes more structure. If none, stop. At a stop there is no pattern for the next question. When reconsideration is needed, state the condition or question and name a candidate pattern whose entry accepts it. Do not model the pattern as receiver or destination.
Archetypal Grounding
Ordinary subject assertion without PFR
In one CGUS position, selectedConstituentRef designates result-42. The exact C.11 ChoiceResult definition classifies that record from its stated disposition, selected option, comparison basis, rule, and stop-probing reason. A.22.CGUS constrains the position locator and selected-constituent reference.
The assertion about result-42 is the first useful output. It adds no owner field, PFR row, actual-use predicate, or basis analysis.
If a later Core relation-function maintenance replay must enumerate every CGUS position whose constituent-kind assertion cites a defining ClaimGraph, add one compact row for this position with the governed use, subject assertion ref, and exact C.11 definition ClaimGraph ref. The named replay receiver needs a PFR row. The importance of the relation alone does not require one.
Framework edition dependency
Start with the readable dependency assertion. The edition labels below are illustrative. FPF@C1 contains the cited Core claims:
CodexProcessFramework@L1depends onFPF@C1for local process authoring: its revision guidance applies E.8:4.1.2, item 6, which requires repair of stale direct consumers in the same authoring increment; its evaluation guidance applies E.21:4.3's rule for an adjacent-value rationale in each coordinate result. Removing or materially changing either claim reopens the corresponding local guidance. Changing the selected FPF edition also requires checking these dependencies.
A local review manual uses the evaluation guidance to judge scores. If the direct-consumer repair rule changes, revisit the local revision guidance. The manual's score judgement can still use the adjacent-value rule if its grounds hold. If that rule changes, trace the change through the local evaluation guidance to the manual. Adopting a different pinned Core edition still calls for the edition checks specified above.
Suppose the manual accepts either a signed assessment under the named edition or an independently validated repeat assessment on the same scale. Either can support the score decision while it remains valid. Withdrawal of the signed assessment matters immediately if the repeat is only planned. If both assessments use the same faulty scale, agreement between them cannot rescue the score. A request specifically for the signed assessment under that edition still needs the signed assessment.
Choose the representation from the receiver's job. A cross-relation comparison may use one generic PFR row. An edition-impact or refresh receiver may use one dependency-specific record. This receiver needs the relied-on content and refresh fields, so it uses only the dependency record:
When a cross-relation task needs the generic view as well, add one PatternFrameworkRelationRecord and give both forms the same subjectAssertionRef. The two tasks may have different recipients. Set genericRelationRecordRef if the dependency task must follow the link to the generic row.
In the generic row, relationFunctionClaimRef points to the E.4.PFR:3.4 dependency predicate, dependencyOrEditionEffect states the E.5.3-constrained direction, and refreshOrSupersessionCondition cites the G.11 refresh condition.
Derive the two records’ shared endpoints, use, direction/effect, and refresh condition from the subject assertion. A change to that assertion refreshes both views together. Neither carries an independently maintained copy of the dependency fact.
If this same pair also has a supported compatibility result for an overlapping use, state that C.2.1 assertion separately. Add its ref to compatibilityClaimRefs only when the named edition-impact receiver must traverse from this dependency record to that claim.
The dependency record proves neither dependency nor compatibility, and E.5.3 does not own either edition.
Source and decision reuse
Suppose a maintainer needs to find the nutrient-monitoring guidance to reconsider after a source-pack selection changes. The following row indexes its bounded source use for that maintenance task:
A hydroponic framework may separately carry a dependency on Core content in an FPF edition, publication relation to its all-in-one carrier, access relation to a grower-assistant skill pack, specialization relation for a narrowed authoring pattern, and quality relations for evaluated drafts. Each remains a different assertion and optional row.
Genuine overlap conflict
A named automated replay receiver whose contract requires acceptance and witness has two exact, accepted, witnessed, independently sufficient bases for the same subject, use, scope, and time cell, and their consequences conflict. Lane 1 can state the conflict but cannot give that receiver a stable closed family-plus-pairwise result.
The basis analysis retains both alternatives, records the exact pairwise conflict, returns established-conflict, and leaves unrelated work available. It selects no winner, grants no permission, and changes no actual-use fact.
Establishment, minimality and a total answer
For a receiving derivation of p under ordinary premise introduction, both {p} and {p,q} are sufficient without any undeclared premise. Both can be established when their required axes are true. Only {p} is inclusion-minimal among their sub-bases. If the receiver requires a minimal family and that additional search is incomplete, the minimal-family answer stays qualified while establishment of these two bases survives.
In evaluate mode, criterion “value ≥ 80” and an applicable exact value 70 suffice to obtain fail under that evaluation rule. With the other required axes true, the candidate is established even though the evaluated object fails.
The criterion/value set's sufficiency says nothing about whether a particular evaluation selected and used it. Lane 3's actual-use predicates need their own facts.
Bias-Annotation
Scope. The assertion-first discipline is Universal within this pattern's subject: every claimed relation starts as a readable assertion with its subject, relation function, basis, polarity, and current facts.
The reusable row, dependency record, package manifest, and high-cost basis analysis are limited tools for a named framework-maintenance use whose next action needs them. None is a prerequisite for ordinary relation prose.
The first drift is reading too much into relation words. Terms such as depends, uses, supports, governs, or profiles are treated as if they settled relation function. Recover the assertion about the subject and the relation it states. Use F.19:4.4 to decide whether a stronger reading needs an explicit correction.
The second drift is to require a standardized row before an assertion can count. Keep the direct assertion as default and demand a named receiver for every row.
The third drift is treating a framework edition like a software package. Compatibility, endpoints, and access carriers are useful, but framework relations are not build imports, module calls, APIs, runtime routes, permissions, or performed Work.
The fourth drift is to mistake definition, citation, evidence, or later sufficiency for actual formal-premise use or criterion selection and to give every actual-use claim a complete analysis. Keep actual use exact and open analysis only for its high-cost receiver.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits. Readable assertions remain cheap. Named maintenance receivers gain stable rows for relation comparison, edition impact, publication/dependency repair, and refresh.
Dependency and pairwise compatibility can change independently and reopen only affected claims. Actual use, candidate bases, conflicts, and downstream reliance become inspectable without adding governance kinds or owner relations.
Costs. Any selected generic row or dependency-specific record must resolve an existing assertion and exact ClaimGraph rather than citing a heading. High-cost analysis requires complete candidate and pairwise work. Existing record-first schemas and owner fields need repair.
Limits. E.4.PFR neither defines every subject relation nor decides source acceptance, evidence, assurance, permission, actual Work, publication, or currentness. A row cannot repair missing subject semantics. A basis analysis cannot manufacture a candidate, actual-use claim, or authority result.
Rationale
Descriptions of FPF pattern ecosystems state predicates, constraints, dependencies, publication and access arrangements, quality relations, and source uses. The patterns do not act on one another. A sequence of pattern descriptions may describe a method or constrain a separately admitted transformation-flow structure, but displayed order alone performs no Work and admits no Transformation or flow.
The assertion-first rule follows C.2.1 identity and A.22.CGUS declarativity. It avoids building a second ontology of pattern ownership while retaining exact relations already supplied by the Core. The PFR row is a projection for its named maintenance use. The subject assertion and defining or constraining ClaimGraph supply its semantics.
The three-lane split also controls cost. Use Lane 1 whenever it closes the task. Lane 2 adds standardized form only for a consumer. Lane 3 separates actual-use truth from optional basis analysis and keeps its output non-permissive.
SoTA-Echoing
Relations
- Builds on: C.2.1 for subject-assertion and analysis-episteme identity, A.6.P and A.6.RCD for exact relation recovery and reusable predicate definition, A.6.0 for
RuleContentBasisFindingDefinition@R7, A.6.5 for the boundary that keeps predicate parameters outside SlotSpec, and A.6.6 for claim-scoped basedness. This pattern's §3.4 defines framework-edition dependency. E.5.3 constrains only its allowed direction and Core acyclicity. - Coordinates with: E.4 for framework-family architecture, G.5 for
JointUseSetand other selected-set results without importing their membership into edition relations, E.4.FPF for FPF form and carriers, E.9 with the E.4.PFAD profile for a framework-architecture answer, C.32.PAD only for an exact project architecture decision, E.11.PUR for recommendation, E.11 for discovery, E.17 for a source-backed publication face and return to source, E.24.PUB for the publication occurrence, form, carrier, audience, bounded use, and availability, and F.18 for names after the value is settled. - Coordinates with: G.11 for edition pins, currentness, and refresh after a dependency is established, never for the dependency predicate. Use E.22, E.2.DA, E.4.DPF.DA, E.21, and E.23 for quality framing, evaluation, and improvement. Use G.2 for source use, E.9 for DRR and selected-answer reuse, and the exact separate acceptance decision and its authority relation or local rule for accepted-decision reuse. Use A.10 and B.3 for downstream evidence, reliance, and assurance, C.33 and C.34 for selected-structure capture and preservation, and C.35 for exact-result admission to an intended architecture use.
- Does not replace: any direct subject pattern, C.2.1 assertion, publication or source decision, evidence or assurance result, authority or permission claim, Work, Method, Transformation, transformation-flow structure, or registered edition.
E.4.PFR:End
Principle-Framework Publication Integration and Preservation
Type: Method pattern Status: Stable Normativity: Normative unless marked informative
Problem frame
Use this pattern during assembly of accepted changes into a candidate FPF, DPF, or LPF publication when the maintainer must answer either of these questions:
- Did the candidate incorporate every accepted source contribution with its intended meaning and use?
- Did the candidate preserve the useful content and selected structure of the predecessor publication outside accepted changes?
Use it especially when a publication form is replaced, split, merged, added, retired, or assigned another bounded use. A clean build, matching source files, or a readable candidate cannot answer the second question.
The primary reader is a framework maintainer or integrator preparing one candidate publication. The primary EntityOfConcern is one candidate FPF, DPF, or LPF edition being assembled for one declared publication use. The method returns a preservation conclusion about that edition. Files and build results are construction means or evidence.
Begin here. Identify the edition and the publication question. For incorporation of a named contribution, follow its meaning into the expressions and receiving uses on which that conclusion depends. For predecessor preservation, compare the expressions whose continuity is claimed. For a full edition these sets cover the candidate, all accepted contributions and its predecessor under the declared publication use.
The first result is a preservation conclusion for the declared use. It names losses, repairs, accepted content changes or retirements, unexpected additions, blockers, and unresolved correspondences or content decisions. Keep the complete comparison checkable. Ordinary retained content needs no positive prose row.
A maintainer can incorporate an accepted change while preserving unrelated predecessor content and assess a changed form while distinguishing the form, carrier, edition, and content.
Related work. Use E.8 to author one pattern, E.24.PUB to identify publication, expression, and bearing relations, E.17 to select reader-facing forms, E.11 to design the public entry, E.4.DPF.DA to evaluate a DPF or LPF package, and E.2.DA to evaluate whole-FPF adequacy.
A responsible maintainer uses the local construction and release methods for repository operations and decisions to accept, admit, release, or land a publication.
For a whole first publication, compare its accepted sources, inspect its complete inventory, check changed public entries and direct consumers, and apply the package evaluation. A request about incorporation of one contribution has the narrower comparison stated above.
Problem
Framework integration has two independent failure modes.
An accepted source contribution may never reach the candidate, or it may reach the wrong public entry or consumer. A candidate may also contain every accepted addition while losing useful predecessor content unnoticed, without any accepted decision to change it. The source-to-candidate comparison can detect the first failure and cannot answer the second.
Publication plurality makes the second failure harder to see:
- one framework edition can be exposed through a sequential monolith, extracted pattern texts, reader-entry and Architectural Reference forms, cards, diagrams, or retrieval forms;
- two forms can serve the same broad use without being versions of one another;
- a text diff can expose changed sentences but not a lost diagram relation, card field, retrieval cue, or admitted operation;
- splitting or merging forms can make a predecessor form disappear even though its content still needs a disposition; and
- a carrier can remain byte-identical while its selected edition, bounded use, or expression relation changes.
The recurring mistake is to use a proxy such as source parity, build success, carrier continuity, heading coverage, or textual similarity as proof that the complete predecessor publication survived.
Forces
Solution
A full edition with a predecessor requires both comparisons below. For a bounded commission, perform the comparison that answers it and give the corresponding conclusion. An incorporation judgment about one approved change can stand on its own. An edition-wide preservation judgment requires the predecessor comparison.
- accepted source to candidate: whether each accepted source contribution was incorporated into the named part of a candidate publication-form expression without changing its accepted meaning or use; and
- predecessor to candidate: whether the candidate preserves the complete predecessor publication outside accepted content changes.
The predecessor-to-candidate question has two branches.
Use a one-to-one expression comparison for an eligible predecessor/candidate pair.
Use an allocation comparison for a non-one-to-one publication-form change, including every split or merge even when a narrower one-to-one pair also survives.
These comparisons and predecessor branches are operations of the method. They introduce no FPF kinds. When a later use needs the conclusion as a reusable episteme, identify it under C.2.1 and state the later reliance separately.
A publication-form expression here is one independently identified expression under E.24.PUB that follows the selected reusable form. Fix its domain and equality rule before comparing it.
The form-mediated relation can remain true while its expression witness changes, so a preservation comparison identifies the predecessor and candidate expressions rather than relying on that relation alone.
Bound the candidate and accepted inputs
For the commissioned comparison, identify the following inputs. A full edition uses the complete sets. A question about one contribution includes the expressions and dependencies needed to settle its use.
- the candidate FPF, DPF, or LPF edition and declared publication use;
- the accepted source contributions covered by the incorporation claim,
- each candidate
PublicationExpressionRelationoccurrence on which the commissioned comparison depends, its selected edition, independently identified expression and bounded-use declaration, and any selected form and itsPublicationFormConformanceRelation. Retain a currentPublicationFormExpressionRelationclaim with its expression witness; - the corresponding carriers and publication occurrences only when their identities affect the comparison; and
- every predecessor expression whose continuity, replacement, retirement, split, merge, or use change is claimed.
Complete the accepted input set for the declared assembly or comparison.
If one source contribution changes a public entry, required input, result, field meaning, action order, stop, return, or another consumed interface, either include the affected public entries and direct consumers in this candidate or leave that contribution out until they can be updated with it.
For either comparison, recover the reader’s assigned work independently of the candidate wording. For incorporation, follow the accepted contribution into the use it was meant to serve. For preservation, follow the former reader’s path through the candidate with the information and help now available. This applies to a one-to-one replacement as well as a split or merge. A new audience or a new commission calls for its own usability judgement.
Before assigning a receiving passage, recover the content architecture that makes that receiver appropriate. Method placement and architectural-account placement answer different questions.
Place a Method by its subject, generality and required contributions under E.4 and E.4.PFAD. Place an architectural account by the named whole it explains under E.11.DSG:4.1. An account connecting Core and domain languages can belong to a Library account even when the general Method it explains belongs in Core.
Retiring a separate carrier supplies neither placement decision. If the receiver is not justified, return that architecture question before treating a content match as an accepted allocation.
For each accepted source contribution in the comparison, record the candidate publication-form expression and the passage, field, relation, cue, or selected structure intended to incorporate it.
A source contribution with no corresponding predecessor content is an accepted addition. Changing or retiring predecessor content needs an accepted content decision; changing or retiring a publication form is not such a decision.
Use E.24.PUB to keep the framework edition, individually identified expression, reusable publication form, expression and conformance relations, carrier, bearing relation, audience, bounded use, and publication occurrence distinct. Use the smallest explicit statement that supports the comparison.
Compare accepted sources with candidate expressions
After assembly, inspect every accepted source contribution in the named part of its candidate publication-form expression.
For each contribution, ask whether the candidate preserves the selected claim, action, result, boundary, relation, structure, or other content that made the contribution acceptable. Classification by filename, heading, or copied wording is insufficient when the receiving use changed.
Classify each accepted source contribution with one of these outcomes:
- incorporated as accepted;
- incorporated with an accepted content change;
- missing or only partly incorporated;
- placed where the intended reader or consumer cannot use it; or
- blocked because the accepted source contribution or intended candidate expression part cannot be recovered.
You can retain an ordinary positive outcome in the checkable comparison without writing a prose row. This comparison establishes source carry-through only. Unrelated predecessor content needs its own comparison.
Compare one eligible expression pair
One preservation comparison follows one eligible pair: one predecessor PublicationExpressionRelation occurrence and one candidate occurrence, with both expression identities fixed.
Where form-mediated continuity is claimed, retain the corresponding PublicationFormExpressionRelation occurrences and identify the expression witnesses being compared.
A pair is eligible only when both expressions have the same declared bounded use and either:
- retain publication-form identity under the FPF pattern that defines or constrains that form; or
- are identified by an accepted one-to-one replacement or continuity decision.
The same broad use alone does not pair two forms when several forms serve that use.
Before concluding preservation, select one complete, form-appropriate comparison inventory. The inventory names every predecessor claim, instruction, boundary, field, relation, cue, or selected structure required for the declared use.
Beside each entry, record why it matters—the FPF pattern that defines or constrains the form, or another accepted comparison basis—and any candidate correspondence.
The inventory aids comparison. It does not make the named kinds members of one new kind.
For comparable text expressions, deletion and replacement spans expose independently actionable predecessor claims, instructions, and boundaries. These spans are one comparison technique. Completeness depends on coverage of what the declared use requires.
For a card, diagram, retrieval form, or another expression without shared span coordinates, inventory the claims or selected structures required for the declared use. Use the FPF pattern that defines or constrains that form and, when applicable, C.33 to state captured and lost structure or C.34 to state a bounded structural correspondence.
Traverse the entire predecessor inventory. For each named item of content or selected structure, record one outcome:
- matched by content or selected structure in one or more named candidate expression parts;
- intentionally changed or retired by an accepted content decision;
- accidentally lost; or
- blocked because the correspondence or decision cannot be established.
Classify candidate content or selected structure without a predecessor correspondence as an accepted or unexpected addition.
If no applicable FPF pattern or accepted comparison basis makes a complete inventory selectable for the declared use, stop at missing form-comparison basis. Carrier identity, visual similarity, or a successful build cannot complete the comparison.
Allocate a non-one-to-one form change
An accepted publication-form addition, retirement, split, merge, or use-change decision names the affected predecessor and candidate expression occurrences and any narrower one-to-one continuity that survives. It authorizes the form change. Predecessor content needs its own disposition.
Run a separate allocation comparison for every accepted form change that leaves an affected predecessor expression outside an eligible one-to-one pair, and for every split or merge even when narrower pairs survive.
- Select a complete inventory for every predecessor expression named by the form-change decision.
- For each named item of predecessor content or selected structure, use the receiving subject and publication function established in section 4.1, then record one or more corresponding candidate expression parts, an accepted content-change or content-retirement decision, an accidental-loss result, or a blocker. A textual match locates the inherited words. Follow the instruction from the intended reader’s entry with the premises, help and other parts then available. For a preservation claim, compare the candidate with the work entrusted to the predecessor’s readers. Can they still make the same choice and obtain the promised result? If a different audience has been given a different job, assess its use on its own terms. When continued service to both audiences was promised, establish both results. A reused task placed after a figure that supplies its answer can now serve a different use. Give the architectural grounds from section 4.1.
- Allow predecessor content to appear in several candidate expressions and several predecessor inventory entries to correspond to one candidate expression part.
- Inspect the complete inventories of the named candidate expressions and classify candidate content or selected structure without a predecessor allocation as accepted or unexpected additions.
- Reuse an eligible-pair result as correspondence evidence when applicable, but do not let it replace the allocation traversal.
Keep every named expression, carrier, edition, and publication occurrence separate throughout the allocation. Use the allocation comparison to reason across these objects. It requires no collective publication kind.
Without the form-change decision, stop at missing publication-expression continuity decision.
Without a complete selectable predecessor inventory, stop at missing form-comparison basis.
Without an outcome for any named item of predecessor content or selected structure, return accidental loss or a blocker. Retiring a form never retires its content by implication.
Complete the affected-publication comparison and return
Identify every affected occurrence of the exact expression relation and its selected reusable form where applicable. Include any form-mediated relation whose continuity is claimed. If carrier or publication identity affects the comparison, include its bearing and publication relations.
Run every applicable eligible-pair and allocation comparison. Check each shared public entry or direct consumer once across these comparisons and keep the separate expression results that depend on it.
Return a bounded preservation conclusion with:
- accidental losses and the expression parts that need repair;
- accepted content changes or retirements that account for a predecessor difference;
- unexpected additions;
- unresolved candidate correspondences or content-change questions;
- blockers and the comparisons they prevent; and
- the declared use for which the conclusion holds.
The complete traversal remains checkable. Unchanged predecessor content and selected structure receive no prose rows. A build result, successful accepted-source comparison, pattern-quality result, or package-adequacy result cannot substitute for that traversal.
For the whole first edition, compare its accepted sources, inspect its complete inventory and affected public entries and direct consumers, and use the applicable package evaluation. A maintainer asked whether a named contribution reached a particular expression can complete that comparison with the sources and dependent uses it requires.
When an unchanged edition is republished through a different form or carrier, use E.24.PUB, E.17, C.33, and C.34 for the changed publication claims.
Use this pattern for accepted-source incorporation or a judgement about continuity of a framework publication or a specified part of its use. The conclusion covers the comparison actually made.
Archetypal Grounding
One-to-one text revision. A DPF maintainer adds an accepted pattern and updates one existing pattern. The source-to-candidate comparison confirms both changes. A text comparison with the predecessor monolith then finds an unrelated non-use boundary deleted during paragraph cleanup. That boundary has no accepted retirement decision, so the preservation conclusion reports accidental loss even though the build and accepted-source comparison are clean.
One form split into two. One predecessor pattern-text form is split into two candidate forms. The accepted split decision names all three expressions. The complete predecessor inventory allocates a shared working-situation claim to both candidates, three predecessor claims to one candidate pattern card, and one obsolete tool instruction to an accepted content-retirement decision.
A predecessor non-use warning appears in neither candidate. Retiring the old form does not retire the warning, so the missing line blocks preservation until the warning is restored or its content change or retirement is accepted.
Architecture diagram. One FPF architecture-diagram form is replaced one-to-one by a candidate diagram for the same orientation use. The diagrams have no useful line diff. The maintainer uses the FPF pattern that defines or constrains that diagram form to select named framework parts, dependency edges, and legend distinctions for the inventory. They use C.33 to record captured and lost structure and C.34 to record the declared correspondence.
One predecessor dependency edge has no candidate correspondent, so similar layout cannot support a positive conclusion.
Local practice framework. An LPF candidate replaces a repeated general explanation with a reference to a new FPF method. The accepted-source comparison passes. The predecessor comparison checks the LPF's local recovery cue, concrete work-size limit, and tool-specific stop because the general FPF method does not include those local actions. Replacing these local actions with the reference would lose them.
A thematic account changes carrier. A Library removes a separately published account that connects general development Methods, human learning and organizational change. The general development construction belongs in Core. The human and organizational mechanisms remain in their DPFs. The account explains their combined use, so its receiver is the Library Architectural Reference. Copying every paragraph into Core would preserve the words but change which whole Core claims to explain.
The allocation first establishes these receivers, then checks the complete transferred account, the retained domain Methods and the updated current returns. A different passage explaining only the connections among Core patterns can remain in the Core account.
Bias-Annotation
Scope: Limited to accepted-source integration and predecessor continuity for FPF, DPF, and LPF publication expressions. The complete-traversal rule applies across text, diagrams, cards, retrieval forms, and split or merged publications only when a selectable inventory and feasible semantic inspection exist for the declared use.
This method is not a universal software-merge, data-migration, or digital-preservation method.
The sources below come mainly from software transformation, merge analysis, and digital preservation. They show how to compare meaning and how the comparison depends on the form. A principle framework remains a different subject from software or an archival object. The applicable FPF pattern defines or constrains what matters for each form, and the maintainer selects the inventory for the declared publication use.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Framework integration becomes more reliable because a maintainer can distinguish a missing accepted change from an accidental loss of predecessor content and can compare text, diagrams, cards, and split forms using an inventory suited to each representation.
The cost is additional work whenever the claimed continuity affects use. Each eligible pair needs a complete form-appropriate inventory, and each split or merge needs a complete allocation traversal.
The cost stays bounded because retained content needs no positive prose rows and the method is used only for accepted-source integration or publication continuity.
A later claim may rely on the conclusion only as allowed by the FPF pattern that defines or constrains that claim. The conclusion carries no process authority.
Rationale
FPF, DPF, and LPF publications share the same integration problem. Accepted source carry-through and predecessor preservation answer different questions. When both are claimed, neither implies the other.
E.24.PUB distinguishes edition, expression, form, carrier, audience, use, and publication occurrence. E.17 explains how to provide several reader-facing forms. C.33 and C.34 define captured, lost, and corresponding selected structure. Redefining those distinctions here would create a second publication ontology. The integration method uses these existing distinctions together.
The one-to-one and allocation branches remain separate because they answer different correspondence questions. A pair compares two expressions. An allocation traverses several named expressions without inventing a collective expression or treating a form decision as a content decision.
SoTA-Echoing
Compare the properties and correspondences that matter for the declared use. The easiest visible proxy may omit them. The sources support this distinction in different cases. Semantic merge informs the one-to-one text case. Traceability and migration inform the split-form case and direct-consumer closure. Digital preservation informs the diagram case and the separation of edition, form, content, and carrier.
The method adapts that stance to principle-framework publications and keeps its scope narrower than general software merge, data migration, or digital preservation.
Reopen this source use when current publication-preservation or structured-transformation practice supplies a cheaper complete semantic comparison, shows that trace reuse can replace rather than only support allocation traversal, or demonstrates a non-framework use that warrants a broader pattern scope.
Relations
- Builds on:
C.2.1for framework-episteme identity andEpistemeEditionRelation, andE.24.PUBforPublicationExpressionRelation,PublicationExpressionBearingRelation,PublicationFormConformanceRelation, andEpistemePublicationRelation, together with the form-mediatedPublicationFormExpressionRelationandPublicationFormBearingRelationwhen making that broader claim. Identify the expression witnesses so that a later reader can recover them. These relations keep the selected edition, expression, reusable form, carrier, audience, bounded use, and publication occurrence distinct. - Coordinates with:
E.17for several reader-facing forms and visible omissions,E.11for public entry and discovery,C.33for captured and lost selected structure, andC.34for declared structural correspondence. - Coordinates with:
E.8for material-revision authoring and direct-consumer repair. A maintainer usesE.4.PFIPto compare an assembled principle-framework publication, not to replace pattern authoring. - Entry from FPF and DPF authoring:
E.4.FPFrefers maintainers here for integration or continuity of an FPF publication edition.E.4.DPFdoes the same for a DPF or LPF publication edition. Those two patterns continue to supply their respective authoring and publication-form guidance. - Evaluation use: A bounded preservation conclusion may be used as evidence under
E.4.DPF.DAonly when package adequacy for the declared use depends on publication integration or continuity. UseE.2.DAto evaluate whole-FPF adequacy andE.21to evaluate individual pattern quality. - Later reliance: Use
C.2.1to identify the preservation conclusion as an episteme. UseA.10when the source-to-use account remains implicit or bounded reliance must remain replayable, andG.11when currentness needs qualification. Those later claims are not part of the comparison.
E.4.PFIP:End
Four Guard‑Rails of FPF
Status: Stable
Problem frame
Use this pattern when authoring or revising Core concepts or their Tooling and Pedagogy forms. Four recurring risks require different checks:
- Implementation language — concept prose acquires tool-specific terms.
- Notation lock-in — one diagram style becomes the only way to express a concept.
- Reverse dependencies — a local convenience makes Core depend on an outer layer.
- Disciplinary bias — domain-specific assumptions enter rules intended for several domains.
These risks undermine Pillars P-1 Cognitive Elegance, P-4 Open-Ended Kernel and P-5 FPF Layering. Use E.5.1–E.5.4 for the corresponding detailed rules.
Problem
Core definitions can become tied to transient tools or one notation, acquire dependency cycles, or present domain-specific assumptions as universal. Authors need constraints that preserve reusable meanings while allowing concrete examples and different forms of expression.
Forces
Solution — the Four Guard‑Rails
Every Core, Tooling, and Pedagogy artefact must obey these four guard-rails at its own layer.
Scope. These guard-rails regulate concepts, claims and their relations. Abstract Core definitions exclude incidental tools, formats and workflows.
A rule about a named publication form, protocol or grammar may retain the syntax necessary for that subject under E.5.1, with its edition/use boundary.
Linting and automation remain outside Core and are optional unless a separate operational contract requires them. They preserve the governed conditions rather than define the concepts.
For the detailed rule behind each guard-rail, see E.5.1 – E.5.4.
Archetypal Grounding (System / Episteme)
Conformance Checklist
All CC‑GR duties are conceptual. Any automated checks are informative only and live in Tooling/Pedagogy.
Consequences
Rationale
The four guard-rails turn abstract Pillars into constraints on wording, notation, dependency and bias. Grouping them here gives newcomers one entry to the applicable rule, identifies any needed amendment, and supplies shared requirements for optional automated checks.
They instantiate Pillars P-1, P-2, P-4 and P-5 and follow the precedence order defined in E.3.
Relations
- Comprises:
- Depends on:
- Constrains: every Core, Tooling, and Pedagogy artefact and all DRRs.
E.5:End
DevOps Lexical Firewall
Status: Stable
Problem frame
Use this pattern when Core prose mentions an implementation detail, such as a file format, build command or runtime flag. Determine whether that detail is incidental to the concept or necessary to the conformance rule for a named publication form, protocol or grammar.
Problem
An incidental tool dependency can force a Core concept to be revised when the tool changes, even though the concept does not. It can also make earlier editions harder to interpret. A blanket ban on concrete syntax causes a different loss when the syntax defines how to test the subject’s conformance.
Forces
Solution
Apply the Lexical Firewall to the subject the text governs:
-
Keep abstract definitions independent of incidental machinery. A Core definition SHALL NOT make an unrelated tool, file format, command or deployment arrangement necessary to identify or use its conceptual subject. Describe that subject directly. Place the implementation in Tooling.
-
Retain the syntax needed to test conformance of a concrete subject. When a rule governs a publication form, protocol or grammar, it MAY state the exact syntax needed to apply that subject's conformance condition. Name the subject and its edition or use boundary. The exception applies only where replacing the syntax with a conceptual alias would remove an operative condition.
E.8 H-2 specifies the characters that separate a heading identifier from its title. An editor can explain the separator in prose. To decide whether a submitted heading conforms, inspect the characters in that heading. A correct explanation of the rule can accompany a heading that fails it. An unrelated build command belongs with the tool that uses it.
-
Keep executable teaching in its teaching or Tooling source. A Core concept needing a runnable illustration cites the teaching or Tooling artifact by conceptual name. Runnable teaching programs, concrete execution paths and commands remain there. A subject-defining syntax fragment under item 2 is not a runnable illustration merely because it appears in a code span or block.
Mathematical expressions follow E.5.2: explain their interpretation, operation and prerequisites, with semantic mapping when compared or translated.
-
Use aliases where they preserve the rule. An incidental technical term can be defined in a Tooling Glossary and cited by conceptual alias. Preserve necessary spelling under item 2 when the concrete subject's condition would otherwise become untestable.
Non‑normative automation. Machine checks MAY exist in Tooling. They are advisory and MUST NOT be imported into the Core.
Archetypal Grounding (System / Episteme)
Four placement decisions
Conformance Checklist
Consequences
Rationale
Keeping incidental implementation details outside abstract definitions supports P-1 Cognitive Elegance, P-2 Didactic Primacy and P-5 FPF Layering. The subject-sensitive boundary keeps abstract concepts portable while allowing a form-specific rule to state the syntax that its reader must use.
Relations
- Parent umbrella:
pat:constitution/guard‑rails(E.5) - Constrains: every pattern in Conceptual Core
- Instantiates pillars: P‑1, P‑2, P‑5
E.5.1:End
Notational Independence
Status: Stable
Problem frame
Use this pattern when expressing FPF content or using another expression to carry that content. Keep the conceptual meaning recoverable while choosing an expression that supports the intended work. A diagram, calculus or sequence of learned signs may help a reader reason or construct a result.
A single expression needs an explanation of its interpretation. A semantic mapping is needed when expressions are compared or one is translated, substituted or relied on as carrying the other's content.
A representation language supplies conventions for forming and interpreting expressions. A particular table, drawing, formula or vocal phrase is an expression read under those conventions. When a decision depends on which forms the language admits or which readings they can carry, use E.5.2.L:4.1. An ordinary choice of expression can be made here.
FPF concepts must remain understandable across disciplines, tools and future notations. If their meaning can only be recovered from one diagram style, file syntax or markup dialect, other users must reconstruct it before they can apply the pattern.
Problem
A definition can depend on unstated conventions of one glyph set or diagram grammar. Translation then changes its claims or leaves them unclear. Treating every visual form as an illustration leaves a useful construction or reasoning operation out of the description when the expression carries that operation.
Forces
Solution - Keep meaning portable while choosing a useful expression
-
Explain the meaning and operative use. Normative content SHALL state the concepts, claims and conditions needed for its use, using prose and mathematics as appropriate. Explain the interpretation of meaning-bearing signs and relations.
A diagram, calculus or learned vocal-gestural expression may carry a reasoning or construction step. When it does, explain the operation and its prerequisites locally or cite the guidance for that subject. The pattern about that subject or the applicable Method governs the reasoning or construction. The reader obtains the subject result by applying that Method.
When the intended operation depends on a change of representation language or reasoning medium, use A.6.3.RT to construct and compare the expressions. A faithful translation whose purpose is textual restatement can use A.6.3.CR.
-
State the semantic mapping. When expressions are compared or one is translated, substituted or relied on as carrying the other's content, their semantic mapping SHALL be stated.
Name the source expression and the expression being compared with it. Name their representation languages when the rules matter to interpretation. State the correspondences for the claims and conditions needed by the intended use. If a relevant distinction is lost, state the loss and limit the equivalence claim accordingly.
First check that the display, its conventions and the other information available to the recipient fit at least one possible situation. If they conflict, resolve the conflict before relying on the display. For a value or a yes-or-no decision, start with the answer the recipient needs. Could two situations consistent with everything the recipient gets yield the same display and different answers to that question? If so, locate the missing information or limit the use. A display can determine whether a limit was exceeded even when it gives only a range for the value. To rely on it, show that every value still possible gives the same decision. Count any independent information used in that argument among the prerequisites.
The recipient may instead need any one acceptable choice. A choice that works in every situation still consistent with the display can be made now, even when the complete lists of acceptable choices differ. If every situation has a solution but they share none, another observation may show which one to choose.
Show how the intended reader obtains the determined answer or makes the supported choice with the preparation and resources available. Use A.6.3.RT.OE when the expression needs to support a calculation, comparison or edit. Use B.5.MR to compare models when the substitution also changes what they say is possible about the same subject.
-
Reference the conceptual role. If the Core cites a diagram, refer to its conceptual role, such as a boundary schematic, rather than making a file or syntax name part of the concept.
-
Keep conceptual prefixes neutral. Use E.10.P for the prefix registry and required anchors. A conceptual label's meaning MUST NOT depend on its expansion into a serialized name or URI. If a tool supplies such an expansion, describe the relation between the conceptual label and the tool's name or URI in Tooling or Pedagogy, and mark it informative.
-
Identify what the form governs. A Core card or table can identify what the reader needs to record and how the entries relate. Another expression can carry those same claims. When the work calls for a file, publication or message in a specified form, check the submitted expression against the convention for that form. E.5.1:4 permits the exact syntax needed for this judgement. A correctly formed submission can still contain an incorrect claim. An independently specified tool or delivery procedure has its own conditions. When proposing a new syntax rule, explain what the users of the governed form need it for. A restriction needed solely by one validator belongs in Tooling with that validator. Put the implementation of the automated check there as well.
The first result is an expression whose meaning can be recovered, with an explanation of any reasoning or construction step it supports. Include a semantic mapping when the use compares it with another expression or depends on the content it carries from that expression.
If a claim changes, compare again with the named source expression. Then repair the correspondence or narrow the claimed equivalence and permitted use.
Use A.6.3.RT to compare content carried between expressions and A.6.3.RT.OE to construct an expression with which the intended operation can be performed. If the representation language cannot express a needed distinction, select another language or extend its conventions before claiming that the distinction is preserved.
Archetypal Grounding (System / Episteme)
A pattern describes a pump boundary by specifying which components belong to the pump. A table records each component's membership. A diagram places the same components inside or outside a closed line. The line denotes the selected boundary, with no claim about physical distance or drawing scale.
The table's reading convention assigns membership through a field value. The diagram's convention assigns it through enclosure. The table and diagram are the two expressions being compared. Their semantic mapping relates entries to labelled components and membership values to placement relative to the line. Both expressions preserve the membership claims.
The diagram can help the user examine connections across the boundary. That operation depends on the represented connections and their interpretation. The boundary's meaning remains available if the table, or another notation that preserves the membership claims, replaces the diagram.
In the episteme case, F-G-R assurance components can be described in prose and represented in a diagram. Explain which elements denote the assurance components and which relations denote their connections. A triple-store serialization stores this content using tooling conventions. Its storage names do not define the components.
An R-score table can also be rendered as a heatmap. State which score or interval each colour denotes and how the review threshold treats equality. For a cell that represents several scores, apply the review rule to every score still possible. If all would be flagged, flag the cell. If all would pass, the cell passes this threshold check. If some would be flagged and others would pass, consult the values or other available evidence for that decision. For an order requiring the numeric table itself, deliver the table even if the heatmap settles every flag.
A project instruction permits a motor to start only while both the clamp-closed and pressure-present signals are true. A local sketch replaces this with clamp closes -> pressure arrives -> motor starts, where each arrow means only that one event precedes the next.
The sketch shows an event order but omits the required overlap: it does not rule out the clamp reopening before pressure arrives.
Compare the source condition with the sketch's timing relations. Show that motor start falls within the overlap of the two true signals, or limit the sketch to showing a proposed event order and use the source instruction to decide whether starting is permitted. This comparison is needed because the sketch is used in place of the instruction.
Conformance Checklist
Consequences
The explanation and mapping let a reader follow named claims between the compared expressions. Stated losses show where a substitution needs a narrower use or a repair, as in the heatmap and timing examples.
Authors must explain the correspondences needed for that use. Keep the comparison to the expressions and distinctions the use requires. Reuse a mapping when it covers the expressions, interpretation rules and claims needed for the present use.
Readers may also need preparation for an expression's operation. State those prerequisites or refer to the guidance that supplies them, as required by §4.
Rationale
A notation can preserve meaning while changing which operations a reader can perform readily. Explaining interpretation, comparing expressions when needed, and retaining their operative use supports P-1 Cognitive Elegance and P-2 Didactic Primacy. Keeping conceptual meaning in the Core and implementation formats in Tooling preserves P-5 FPF Layering.
SoTA question and choice. How can framework content remain understandable across notations while an expression also supports reasoning or construction? Adopt recoverable interpretation of concepts, claims and conditions. Adapt this portability rule to an expression that carries an operation by explaining its use and prerequisites. Compare the content whenever another expression is used as equivalent.
A prose-first alternative treats diagrams and written calculi as secondary illustrations. It keeps verbal definitions accessible, but can exclude an operation from the normative account when the operation is performed through the expression. Reject that blanket restriction.
For example, in the Euclidean construction in A.6.3.RT.OE:5.1, the same segment participates as a radius and as a triangle side. The prepared expression helps the reader combine those relations. Its role needs an explanation of that operation, beyond a caption describing the picture. Either intersection of the two circles yields a valid triangle on the given segment. A request for the triangle on a specified side adds a condition to that construction.
A second alternative is to require one canonical notation. This can provide shared interpretation and manipulation rules within a practice whose readers have learned them. Retain that option for such a practice. Extending it across FPF would also require readers in other practices to acquire those conventions, even when another expression supports their task.
Compare these choices for the same content, operation and reader preparation. The selected rule accepts the cost of explaining local conventions and mapping compared expressions in exchange for allowing the expression suited to the work. It claims no universal advantage in learning time or performance.
Effect on this pattern. Section 4, item 1 and CC-NI.2 require guidance for an operation-bearing expression. Item 2 and CC-NI.3 require comparison when content is carried between expressions. The pump and timing cases in §5 show preservation and consequential loss. The heatmap case limits a coarsened expression's use. These moves keep interpretation and usable operations together. Subject Methods supply the reasoning or construction.
Source roles and limits. Macbeth's paper-and-pencil analysis (2011, especially pp. 16-18 and 31-42) supplies the constructive argument from prepared expressions and shared parts.
Dutilh Novaes's Formal Languages in Logic (2012, §§3.2, 5.2 and 6.1) supplies the account of learned manipulation and interpretation. On p. 202 she argues that temporarily disregarding meaning is neither necessary nor sufficient for a cognitive gain.
These are conceptual grounds for the selected line and its reader-dependent limits. They provide neither a universal notation-design procedure nor a comparative estimate of learning or performance across all readers and notations.
Reopen the choice if interpretation and mapping satisfy this rule yet a needed claim or operation remains unrecoverable, or if a canonical-notation alternative supports the same work for the intended readers with less total preparation and explanation. Reconsider the affected requirement. An incorrectly prepared expression instead calls for the local repair or source return in §4.
Relations
- Parent umbrella:
pat:constitution/guard‑rails(E.5) - Constrains: the expression of normative Core content, including project-local expressions used as carrying that content. The semantic-mapping requirement follows the comparison or intended reliance.
- Uses: A.6.3.RT when the intended operation depends on a change of representation language or reasoning medium, A.6.3.RT.OE for constructing an operative expression, and C.2.8 and C.37 when recovery or use must be assessed.
- Conceptual prefixes: E.10.P supplies the policy, registry and anchor requirements. E.5.2 keeps interpretation independent of tooling-specific expansion.
- Subject reasoning and construction: the pattern about the represented subject or the relevant Method governs the operation.
- Method descriptions: use
U.MethodDescriptiononly when A.3.2 applies: the episteme concerns one admitted Method and makes substantive claims about that way of doing. - Instantiates pillars: P‑1, P‑2, P‑5
E.5.2:End
Languages, Expressions and Interpretation
Type: Definitional pattern Status: Stable
Problem frame
Use this pattern when a decision depends on how someone can form an expression, read it, or use it to identify an entity. An engineer may need to select the notation for a calculation, compare two specifications, or identify equipment from a code. Start with the expression and the decision. When colleagues already understand an ordinary instruction in the same way and can act on it, that reading suffices.
Problem
Readers can agree on every printed sign and still disagree about the operation it specifies or the object it designates. Conversely, the reading of a record can change when the issuing organization reassigns a code even though everyone still uses the same language. Which system of conventions is in use, and what else must be known to answer the question?
Forces
A familiar name makes a language easy to use. A comparison may require the community, edition and conventions behind that name. Two teams may need agreement on only one construction. A release claiming compatibility across an entire notation needs correspondingly wider grounds.
Solution
A language (Language) is an identifiable repertoire of expressions with conventions that determine which expressions belong and, for composite expressions, which constructions are admissible. A grammar may specify the conventions. They may also be evident in established use. A formal language of strings can be defined by membership alone.
A representation language (RepresentationLanguage) is a language whose conventions also give its expressions permissible readings under specified conditions. A language of designations (ReferenceLanguage) is a representation language whose complete entries serve to identify entities under its resolution conventions and any context or bindings they call for. A prefix, separator or other part of an entry can help form a designation without identifying any entity on its own. Each more specific kind classifies the very same language. English contains personal names. An equipment-code language embedded in an English manual can be specified and selected independently.
Recover what the conventions establish
For a rule about admissibility or interpretation, fix the convention, participants and conditions. When an expression refers to an individual, establish what selects it in this use. A pointing gesture or a previous mention can identify it in ordinary conversation. An issued code requires the applicable assignment. A reader's act of interpretation, the act of issuing a code, and a physical change in the object each have their own evidence. Express the needed rule or fact in a claim. A separate relation occurrence needs its own identity and a practical use that depends on referring to it (A.6.RCD, E.24).
A malformed entry fails the formation test. A complete authoritative register can establish that a valid code had no assignment in a given interval. Several candidate assignments may leave the referent ambiguous. An unavailable or incomplete register can leave the assignment unknown. These conclusions support different actions, from correcting the expression to obtaining the missing register.
Choose what a reference to the language identifies
Identify the conventional system. Specify its community, period or edition where the distinction changes the answer. A specification can document that system, an expression can belong to it, and an expression can refer to something else. Two specifications can define the same language. An edition label or the name of a family alone cannot establish that two editions admit the same expressions and readings.
For languages defined solely as sets of strings over a fixed alphabet, compare membership. For a language with parsing or interpretation conventions, compare those conventions as well. When only selected readings matter, state the construction, inquiry and conditions on which the readings agree. This bounded result can be sufficient for the intended use.
A sublanguage has an identifiable repertoire within a language and inherits its formation conventions and any interpretations for the selected constructions. A profile selects provisions for a use. If those provisions define such a repertoire and preserve the conventions, the profile also specifies a sublanguage. A register concerns customary choices of expression for an audience and activity. A language serves as a metalanguage when its expressions specify or examine the expressions or conventions of another.
One language can express a repeatable method, its local adoption and a particular performance. An equipment code can designate a pump while a work instruction says what to do with it. A writer can also change register and retain the same commitment. To claim that a second model refines the first, establish that at least one design, performance or other completion satisfies the second model, and that every such completion satisfies the first under the chosen interpretation. Strict refinement also excludes a formerly admissible answer to the same inquiry. B.5.MR:4.2 gives the test.
Use representationLanguageRef when the receiver must identify the language. A notation designer choosing a system in which a reader can build and interpret new expressions and an integration engineer comparing languages of asset designations can reuse these same kind criteria. When a single record already supplies its one relevant interpretation by value, that record needs no separately identified language edition. Name the binding authority or give the necessary rule by value when that is what the receiver needs. C.2.1 determines the effective interpretation of a particular episteme, which may also fix modality, permissible inferences and qualifications. E.8.ML addresses a pattern language that organizes methods and their dependencies. A learned representation with a successful readout supplies evidence for that readout under the tested conditions. A claim about a language in the representation also needs an identifiable repertoire and formation conventions.
Use a language classification in a typed reference
When an interface needs to constrain what a language reference denotes, use the C.3 declarations below. Each row declares one U.Kind individual. Its KindSignature is a C.2.1 episteme whose EntityOfConcern is that kind; a table carrying the three declarations does not combine their subjects. The member language, its specification and its classification declaration remain independently identifiable.
For each declaration, the application slice selects the conventional system under examination, including its community, period or edition when needed. A formal string language also fixes its alphabet. Use the selected conventions rather than an implicit latest edition. The effective interpretation of the declaration combines the criterion in its row with §4.1's formation, reading and reference rules and §4.2's member-identity rule. The declaration's formality is a characteristic of that episteme.
Recover the candidate's identity and conventions from the governing source or established practice. Test admissibility before membership under C.3.2: a known candidate outside the declared domain gives not-applicable; insufficient support for a criterion on an admissible candidate leaves membership unknown. Unknown does not mean false. Classification establishes neither the truth of an expression nor a successful lookup or performance.
Each kind continues while its candidate domain and operative membership distinction continue under C.3.1. Rewording a declaration or applying it in another discipline creates no new kind. Changing a member language's conventions can identify another edition while leaving the classification criterion unchanged. No ExtentRule is needed for the reference uses here; supply one when a later operation consumes an extension.
The added criteria entail the preceding criteria for the same individual: ReferenceLanguage is a subkind of RepresentationLanguage, which is a subkind of Language. A reference to a member is a U.EntityRef whose referent satisfies the corresponding criterion. Its binding selects the language; classification does not create or reidentify it. Here C.3's local-kind classification supplies a reusable distinction for the stated uses. It does not restrict a language to a particular enterprise or locality.
Archetypal Grounding
A system and its designation. A maintenance planner finds equipment code P-17 on a drawing. The drawing and maintenance register use the same form. The designated pump depends on the issuing plant and the date of assignment. A later reassignment can change which pump a job concerns within the same code language. To route the job, resolve the assignment in force when the work is to be done. To establish what happened to the pump, consult the work evidence.
An episteme and its reading. Two reports use the same table language. Each lists two scheduled gauge readings, 8 in the morning and a blank at noon. One report instructs the reader to use zero for a blank when calculating the daily mean. The other omits blanks and gives the number of readings used. With these entries the first report gives a mean of 4. The second gives 8 from one reading. To report that the gauge actually read zero, establish that the noon measurement took place and yielded zero. A reliable witness can establish this. So can a record of that reading, including a blank when there are grounds that the recorder used it to record a measured zero on this occasion. An arithmetic default supplies no such observation. If a new edition of the table language prescribes a calculation for every report, the scope of the convention changes. A rule in one report governs that report.
Bias-Annotation
A published grammar makes formal conventions easy to inspect. In a workplace, people can learn a convention through use. Ask which new constructions they accept and how they read them when the answer matters. Shared spelling or agreement on a familiar example can conceal a difference elsewhere.
Conformance Checklist
- CC-LANG-1. A language claim identifies the repertoire and the conventions on which membership depends. The representation and designation specializations have the additional conditions in §4.
- CC-LANG-2. An equivalence or preservation claim names what can be recovered, for which uses, and how the comparison establishes it. Success on selected entries warrants the bounded result.
- CC-LANG-3. A resolved reference identifies what selected the individual in this use. Ambiguity, a valid unassigned code and an unknown assignment each have their own grounds.
- CC-LANG-4. A claim about an actual reading, assignment or performance identifies its occurrence and evidence when that fact changes the work.
Common Anti-Patterns and How to Avoid Them
Routing a job because the printed code matches can send it to the wrong equipment. Resolve the designation under the applicable assignment. Treating a more formal wording as a stricter rule can also change a decision without warrant. Apply B.5.MR:4.2 to the rules as models of the same inquiry. Establish that the revision has a feasible answer, preserves the earlier conditions for every answer it permits and excludes a formerly possible answer.
Consequences
A reader can locate a disagreement in formation, conventional reading, a document-specific interpretation, or an assignment. The appropriate repair may be a new expression, an explicit reading, correction of a binding or collection of evidence about the world. Carry only the distinctions on which the receiving decision depends.
Rationale
A shared definition lets a notation comparison and an equipment register reuse the conditions for identifying a language. Specifying what a well-formed expression may mean gives the designer a reason to permit a translation. Establishing the assignment of a designation gives the worker a reason to select an individual object.
SoTA-Echoing
For a spelling check, comparing the written signs can settle the question. RDF permits distinct concrete syntaxes to express the same abstract graph. In SMT-LIB, the signature and the sorting rules determine whether a term of the indicated sort can be formed. A structure interprets the symbols of that signature. For those uses, comparing spellings alone can miss equivalent readings or accept a construction whose declared sorts do not fit. Section 4 therefore asks which constructions are admitted and, for an interpreted language, which readings its conventions support. A shared convention for writing identifiers still leaves an issued identifier’s assignment to be resolved. Reopen the affected comparison when a declaration changes the admissible construction or reading, or a new binding changes the referent. W3C RDF 1.1 Concepts, §§1.3 and 2 · SMT-LIB standard, §§5.2–5.3
Relations
E.5.2 governs notational independence. A.6.2 uses language correspondence in comparing epistemes. A.6.3.RT applies when an operation depends on a change of representation language or reasoning medium. A.6.3.CR handles a faithful textual restatement, including an ordinary translation. C.2.1 governs effective episteme interpretation. C.30.AD.BA uses qualifications for built-asset designations. E.24 and E.24.UK govern direct relations and kind admission. B.5.MR compares the commitments of two models. C.39.WA:4.3 applies that comparison when explaining a whole subject.
E.5.2.L:End
Unidirectional Dependency between FPF Families
Status: Stable
Problem frame
Use this pattern to decide whether one FPF member may rely on another member, in the same family or a different one, for a definition, rule or procedure. Name the work that needs the supplier and what that work takes from it. A reader who follows a link for an optional illustration can use the original rule independently. For an assertion about two framework editions, establish the dependency under E.4.PFR:3.4.
Problem
A tutorial may teach a Core term using a helper. If the helper's output becomes the only way to tell what the Core term means, changing the helper can change the term for every Core reader. A cycle of imports also prevents an author from supplying one member on the declared basis independently of the others in the cycle. Pillar P-5 governs this separation.
Forces
Solution — One‑Way, Acyclic Imports
Read a dependency arrow from the member that needs content for a stated use to the member that supplies it. The strict partial order on distinct families constrains the permitted direction of cross-family imports. The actual imports connect particular members. A family may have several permitted suppliers and need only one for the work at hand.
imports expresses dependency. Specialisation is a separate relation (normative). When an artefact exposes an imports : [...] list (e.g., SignatureManifest.imports in A.6.0), treat each dependency claim in the list as an edge of its imports graph. The graph MUST be acyclic (a DAG) and MUST respect the declared direction.
imports MUST NOT be used to encode specialisation (e.g., ⊑ / ⊑⁺ between mechanisms). Specialisation relations and any specialisation-chain claims are declared separately under their defining rules. For mechanism declarations, apply A.6.1:4.8 to the comparison claim. Use C.29 when a separate mathematical-representation choice, transfer or reliance question remains.
Permitted suppliers for Core, Tooling and Pedagogy
Each row lists permitted suppliers. Use the ones needed for the member’s stated work. A guide can take a definition directly from Core. An intermediary is needed when its contribution is required for that use.
The acyclicity rule applies within each family as well as across families. Before adding an import, follow the existing imports from the proposed supplier. If a path reaches the receiving member, the new import would close a cycle. Keep Core meanings independent of Tooling, Pedagogy and domain or local frameworks (E.10 V-1, E.4). Linters and other executable checks of Core conformance belong in Tooling. Core states the conditions they check. For Core, DPF and LPF editions, apply the edition rules in E.4.PFR:3.4. An LPF may depend directly on a DPF, and either may depend directly on Core. E.4.DPF also permits a DPF to rely on a contribution from another DPF when it fits the receiving use. The discipline to which a Method belongs, the order of chapters and the level of abstraction of a model each need their own grounds (E.4:4.2.2, B.5.MR).
Extending the order. Classify a proposed tool or teaching publication by the use it supplies. It can join an existing family under that family's rules. For a genuinely new family, use E.4.PFAD and E.9 to decide both who may supply it and who may import from it. Under the current extension rule, place it on the receiving side of at least one existing family. Follow the proposed directions through the existing order. Reject a loop even if nobody has yet exercised any of the permissions. Specify any change in what existing families may obtain from one another and seek the amendment needed before an existing prohibition can change. Families without a direct or transitive ordering relation can remain incomparable. A permitted direction need not yet have an actual import. A proposal outside the extension rule requires a change to that rule before admission.
When a needed import would be forbidden, keep the contingent claim with the family that can warrant it. If Core readers need a transdisciplinary rule, supply the rule and grounds they can use independently. A guide can receive credit for the discovery. Copying a guide's convention leaves Core dependent on that convention whenever the argument still requires it. If two members would have to import one another, look for a common definition they can both use from a permitted supplier. Where two definitions must be given together, consider one defining account and check its imports. Printing two mutually dependent members in one file leaves their cycle intact.
Archetypal Grounding (System / Episteme)
A guide teaches the boundary invariant from Core and links to a profile checker for practice. Readers can establish a boundary from the published rule with the checker unavailable. One exercise accepts any report satisfying a separately published format. Another generator can supply it if its report meets the acceptance conditions. A separate exercise requires a report signed by the named checker in the agreed format. This exercise relies on the checker as well as the Core rule. If the available checker can no longer produce an acceptable signed report, use a compatible edition whose signature the recipient accepts for the named checker. If none can supply an acceptable report, revise that exercise. Readers can continue to work from the Core definition. Recheck their explanation if its definition or the argument used to teach it changes.
Conformance Checklist
Consequences
Rationale
Under P‑5 FPF Layering, an author can replace a tool or a teaching example while readers continue to use the Core claim on its own grounds. P‑4 Open‑Ended Kernel keeps a convention that depends on a particular practice with its proper supplier. P‑10 Open‑Ended Evolution allows a warranted proposal to change the family order. Until the competent authority enacts such a change, the current order governs every import.
Relations
- Parent umbrella:
pat:constitution/guard‑rails(E.5) - References family definition:
pat:constitution/fpf-ecosystem-family-architecture(E.4) - Instantiates pillars: P‑4, P‑5 and P‑10
- Constrains: Actual imports between FPF ecosystem members. A manifest, DRR or SCR may record an import. A required claim can still create a dependency when its source is unacknowledged.
E.5.3:End
Cross‑Disciplinary Bias Audit
Status: Stable
Problem frame
Use this pattern when you write or review a normative pattern and need to check its assumptions, explain an established rule to its readers, or decide where the rule applies. A familiar example can conceal who makes a decision, which parts have to work together or what a record actually establishes. The audit helps you find such assumptions. To recommend the rule in another discipline, establish what entitles a practitioner there to take the proposed action or draw the proposed conclusion.
Problem
Bias in wording, examples, unit choices or principle weighting can make a rule appear applicable where its assumptions do not hold. This undermines P-2 Didactic Primacy and P-8 Cross-Scale Consistency. The author needs to justify the emphasis, repair the imbalance or state a narrower scope.
Forces
Solution — Principle‑Taxonomy‑Guided Bias Audit
Examine the rule from all five classes in E.3. Gov asks who can decide or change what the rule relies on. Arch asks which parts, interfaces or invariants the result depends on. Epist asks what the terms refer to and what supports the inference. Prag asks whether the work is feasible and worth doing for its intended use. Did asks what the reader needs to recognize and do in their role, with the help available for the intended use. If independent performance is promised, check which choices the reader must make. For each, ask whether the emphasis is justified and whether an omission would change how someone uses the rule. E.3:4.1 assigns a principle to one class for precedence. Here the same rule is examined from all five. Give each concern the weight its effect on use warrants.
State what carries across disciplines. Locate the rule and its grounds wherever the author develops them. Which features of the situation does the inference require, and which may vary within its declared range? Name the action or conclusion it licenses and the conditions it relies on. For each discipline in which you claim it applies, explain what plays the relevant parts there and how its practice supplies the conditions. One argument can cover several disciplines when their definitions or work arrangements establish the same relation. A worked case may carry the whole argument or illustrate how an argument established elsewhere applies. A shared word, an attractive analogy or successful joint work can suggest a transfer to investigate.
Match the grounds to the conclusion. A derivation together with a warranted interpretation can establish what follows under specified conditions. For a predicted change in performance, use an adequate comparison in the receiving practice or explain why evidence from elsewhere bears on the effect under the receiving conditions. Account for material differences such as workload or case mix. To say who may act, recover the decision or rule that gives that authority. Where a condition is unknown, state what the recipient can already conclude and what they would have to establish for the further use.
Test the boundary. If a plausible alternative gives a different conclusion, try a case that decides between the two. A derivation can already settle the claim throughout the stated scope. If an instance meets the stated conditions and contradicts a factual conclusion asserted for every such instance, revise that claim. For a normative rule, distinguish breaking the rule from a valid exception to it. An unauthorized release can show a violation. An authorized exception can disprove the claim that every release requires the same approval. To retain a narrower rule, name a condition the practitioner can establish independently of the promised result and show why the conclusion follows within it. A case outside the conditions can expose a limit to use. It can also defeat a separate promise that every intended recipient will meet those conditions. For a claim about frequency or average effect, examine what the observations support about that frequency or effect.
Report the material result. The Bias-Annotation records the justified emphasis and any repair or scope limit that changes use. If a local rule is intended for one practice, explain the local condition that matters. For universal-core placement of a durable U-kind, A.8:3 and E.24.UK govern the projections into at least three foundationally different domain families and the other admission conditions. Workflow checklists, reviewer metadata and data-format advice belong outside Core.
Archetypal Grounding (System / Episteme)
When can an earlier check help settle a later acceptance criterion? Combine what it established with the other evidence available for the relevant item and state. First check that at least one situation fits the evidence. Conflicting records may need reconciliation. If all the possibilities that fit meet the criterion, the receiver can accept that aspect. If they all fail it, the receiver can reject it. If both remain possible, report the established part or obtain the observation that decides between them. An applicable risk rule may also authorize action on incomplete evidence. Other acceptance criteria and the authority to release the item keep their own force.
In mechanical inspection, the gauge reads 20.01 mm on an identified shaft. For this check the inspection procedure warrants an absolute error bound of 0.01 mm. The whole interval fits a receiving tolerance of 19.95–20.05 mm. If the measurement is valid for the shaft in its present state, the inspector can accept its diameter. A reading of 20.20 mm with the same bound settles rejection. Two checks can also settle acceptance together, one establishing the lower limit and the other the upper limit for that shaft in the same state. A requirement for performance under load still calls for evidence about the loaded assembly. Suppose the shaft passes the diameter check but a damaged keyway lets the assembly slip during its rated-load test. That outcome defeats the stronger rule that passing the diameter check guarantees performance under load. The diameter finding still holds.
In archival description, the receiving question might be which date the accession register gives for a photograph. A verified link from the photograph's identifier to its register entry and a legible entry giving the date settle that question. Dating the exposure itself requires grounds for interpreting the entry as a reliable record of the exposure date. An accession date can be entirely correct and still answer a different question. These two disciplines support the same conditional principle. Whether either team saves time by reusing checks is a further empirical question.
A reader who can follow the inspection example may still need help recognizing which archival date matters. Explain the receiving criterion in the archivist's vocabulary. The distinction between what the record establishes and what the next decision requires survives that change of wording.
Conformance Checklist
Consequences
Rationale
Using the Principle Taxonomy classes as the audit lenses keeps the guard‑rail concept‑driven. Review-board procedures, CI jobs, and workflow checklists belong in Tooling. This separation satisfies GR‑1 (Firewall) and supports Pillars P‑2, P‑7 Pragmatic Utility, P‑8.
Relations
- Parent umbrella:
pat:constitution/guard‑rails(E.5) - Depends on:
pat:constitution/principle‑taxonomy(E.3) - Constrains: All normative patterns claiming universality
E.5.4:End
Didactic Architecture of the FPF Specification
Status: Stable
Problem frame
Use this pattern when arranging the specification's explanation and practical entry. Readers differ in at least two respects:
- Disciplinary background — for example, systems engineering, knowledge science or ethics.
- Preparation — some readers need help seeing a relation in use. Others can enter through the definition or operation that answers their question.
Problem
The publication needs an explanatory order in which readers can find the grounds needed for later claims. A practical question may require entry into a later Part. Document order must not become a compulsory workflow.
Forces
Solution — Order the explanation and keep practical entry separate
Document order is distinct from first-practical entry
The document's macro-order organizes explanation. It does not prescribe one practical workflow. Apply E.11.PFP to the compact opening and navigation.
After the Table of Contents, the Readme offers recognizable working questions and direct entries. The Architectural Reference explains the whole pattern language, its cross-cutting ideas, connections and architectural choices, including introductory explanation.
Choose the opening from the reader’s question. What are they trying to understand, do or construct? Which whole, constituent or relation is the question about, and what must the model let them distinguish? Begin where the reader can obtain the first grounds or perform the first operation they need. Follow its contribution to the promised result across Parts when necessary. E.11 helps when several plausible entries would lead to different work.
Keep the main explanatory sequence recoverable:
- Readme and Architectural Reference. The Readme helps a reader find a first useful result. The Architectural Reference explains the connected language and its rationale. Its introductory System and Episteme stories build intuition before the kernel exposition.
- Part A — Kernel Architecture. Introduce the holonic ontology and acting-side externalization in A.12.
- Part B — Trans-disciplinary Reasoning. Explain the reasoning moves and their grounds for the kinds of things each question concerns. Use E.7 to show how the reasoning applies.
- Part C — Kernel Extension Specifications. Develop the calculi and extension patterns whose questions use those foundations.
- Part D — Multi-scale Ethics and Conflict Optimization. Introduce the reflective ethical and conflict questions with the needed holonic distinctions available.
- Part E — Constitution and Authoring Guides. Place framework governance and contributor guidance after the principal kernel and reasoning exposition so an ordinary first use need not begin with authoring rules.
- Part F — Unification Suite. Keep concept-set, sense and system-role unification methods in their own Part. Reach them directly when the question calls for them.
- Part G — Discipline SoTA Patterns Kit. Keep the discipline source-synthesis, comparison and refresh methods in their own Part.
Part H remains reserved. Examples and extended explanations belong with the pattern whose contribution they explain. Tooling guides and executable examples follow their separate family and lexical rules.
This publication order does not require readers to traverse every Part. Practical entries return to the pattern needed by the reader’s question.
Archetypal Grounding (System / Episteme)
Consider a reader who wants to understand how a whole can depend on its constituents. The two reading paths below show how an introductory account, a definition and an application can contribute to that question. The physical arrangement and the grounds for a synthesis require different relations among constituents. A reader who needs only the criterion for one kind can enter at its definition.
Conformance Checklist
Consequences
Rationale
A reader needs grounds for using the rule in the work at hand. A direct demonstration can establish them. A contrasting application earns its place when it reveals a condition that the first leaves open. The document order and E.7 make those grounds accessible while practical entries can cross Parts. This serves P‑2 Didactic Primacy and P‑1 Cognitive Elegance.
Relations
- Depends on:
pat:constitution/guard‑rails(GR-1 governs implementation detail and necessary subject syntax). - Constrains: Placement of Parts, patterns, and their explanatory examples.
- Instantiates pillars: P‑1, P‑2
E.6:End
Archetypal Grounding: Explain Rules of FPF Architectural Patterns through Cases
Status: Stable
Problem frame
Use this pattern when writing or reviewing an architectural rule that somebody must apply. Recover the question that brings the practitioner to the rule. What do they need to explain, construct, decide or do? Choose material that shows how the rule supports that work. A reader who came for a fact about one occurrence can use the fact directly.
Problem
An author can explain a related subject accurately while leaving the reader's question unanswered. The mistake can run either way. “Identify the objects and relations” omits which participants and relations matter to this decision. A fully worked instance can prescribe a particular kind of participant, arrangement or model where the reader was meant to choose one. More detail about that instance can leave the original choice just as obscure.
Forces
Solution — mandatory Archetypal Grounding subsection
Every architectural pattern SHALL include a section headed Archetypal Grounding. Derive its content from the practitioner’s question and the work they must be able to do. Compare that work with what the Solution actually supplies. Establish who must choose, perform or explain what, which kinds of thing the question concerns and which relations support the answer. A method can organize several constituent methods. A system can perform work by a method, and an episteme can report what happened. A fact about one of these can help answer a question about another when the connection is established.
Choose a fitting passage, case, comparison, or usable boundary of the rule. Show what the practitioner starts with, what the rule enables them to explain or obtain and how. If they must devise the way of working, show how to make and justify the choices that a finished example would supply for them. When a passage changes the kind of holon, its scale of composition, or the level of abstraction of its model, identify the correspondence at the point the reader needs it. On returning to the commissioned question, use the part of the argument that survives this move. A demonstration can contain the entire general argument when its reasoning covers the permitted variations. A finding confined to one case carries that reach.
When U.System and U.Episteme disclose different conditions of the rule, show the difference. Rules about methods, work, disciplines or relationships across kinds need grounding in those subjects. For a range, explain what holds at each relevant scale or level, what transfers between them, or what they support only together, according to the claim. C.39.SA helps to select the range. C.37.MM connects findings from several models. F.19 returns to C.39.QR when a clear passage answers a different question, and to C.39.GA when the reader needs the omitted general argument.
Archetypal Grounding for E.7
An author explaining B.1 wants a practitioner to decide whether several constituents make the claimed whole. The author first identifies what would warrant that decision for the kind of whole in question. In B.1:5.1 the arrangement of physical components supports a claim about an assembly. In B.1:5.2 a collection of sources can exist while their claims still lack the integration required of the proposed episteme. The author chooses the contrast because it exposes a consequential assumption: the way constituents make a whole must be established for its kind. Merely naming a system and an episteme would leave that decision to the reader.
Now change the commission: the reader of A.15.1:5.1 must establish which executions constituted a particular completed work. The same team and plan are compatible with an operation performed during that work and with an operation performed earlier whose result the team used. The author therefore grounds the rule in what was actually performed and how that execution belonged to the work. The plan and prior result have a supporting role. This choice follows from the question about performed work and can be used in another organization with other performers and records.
Conformance Checklist
Consequences
A practitioner can see which choices the rule supplies and which they must make for their own work. The author can keep a short direct demonstration, or add a contrast where the first application leaves a consequential alternative unresolved.
Rationale
P‑2 Didactic Primacy requires the explanation to support the practitioner's work. P‑1 Cognitive Elegance favours the smallest demonstration that carries the needed reasoning. Change an assumption of the example while keeping the commission fixed. If the reader's justified answer changes, make the dependency visible. If a claim is universal over a stated domain, a counterexample inside that domain can settle the question against it. A successful instance supports the general claim only to the extent that its argument applies throughout that domain.
Relations
- Implements macro flow:
pat:authoring/didactic‑architecture(E.6). - Selects relevant kinds and relations:
pat:kernel/holon(A.1), C.39.SA and C.37.MM. - Exposes assumptions:
pat:guard/bias‑audit(E.5.4) via CC‑AG.3. - Governs the grounding section:
pat:authoring/pattern‑template(E.8).
E.7:End
FPF Authoring Conventions & Style Guide
Type: Architectural (A) Status: Stable Normativity: Normative (unless marked informative)
Use this when
Use E.8 when you are writing, revising, or reviewing one FPF pattern and need to know what shape, voice, reader-recognition function, and assurance material the pattern must carry before it can be treated as mature FPF text.
For the whole account of an FPF, DPF, or LPF and its substantive profiles, reuse the twelve content functions through E.8:4.1.4 and E.11.PFP:4.7.
Those functions guide the explanation at each selected scale; the individual-pattern heading grammar remains the form for individual pattern bodies.
Use it especially when a draft is technically correct but hard to use: the cold reader cannot tell when to apply it, what action to take, what mistake that action prevents, which related pattern defines or constrains a specific outside claim, or which assurance material is informative rather than the first user-facing guidance.
Not this pattern when. Use E.9 when the main work is deciding why FPF should change and how that decision is distributed across patterns. Use E.19 when the main work is an admission or refresh review. Use the local domain pattern when the question is what FPF says inside that domain rather than how a pattern should be authored.
What goes wrong if missed
A pattern can satisfy a checklist and still be practically unreadable. A draft may open with package architecture instead of a recognisable working moment, bury its payoff, hide the pattern that defines or constrains a specific outside claim, or let assurance prose silently replace the reader-facing claim.
The result is a formally neat text that authors can defend but practitioners cannot reliably use.
What this buys
E.8 gives FPF authors a shared pattern shape and authoring discipline: recognition text first, assurance text second, canonical sections present, terminology kept stable, SoTA used as current practice grounding rather than decoration, and practical consequences visible before a reader has to reconstruct the architecture.
First useful move. Recover the question and the work the intended reader should be able to perform. Choose the first operation by the contribution it makes to that work. Put the working situation, that operation, its practical payoff, the ordinary boundary and the nearest heavier assurance condition into the recognition text.
Solution and working move. Solution gives the pattern's conditional answer to its Problem frame, Problem, and Forces: what the reader should do or decide, under which conditions, what result to seek, and when to stop or return.
A working move is ordinary reader-facing wording for one such action or judgement. Reserve U.Move, dated U.Work, and U.Transformation for claims that assert those admitted objects.
E.11.PUA governs use of one selected Solution to reach the first useful result. When alternatives are formally qualified under A.22.CGUS, call them continuation candidates; E.18.3 applies only when the selected CGUS uses a qualifying transformation-flow substrate.
Move wording in pattern prose. In ordinary prose, say recommend this pattern use, coordinate these uses, or show their total order when those are the claims being made.
When the durable governed object matters, use its exact published designation under E.11.PUR: PatternUseRecommendation@Context, PatternUseCoordination@Context, or PatternUseSequence@Context. The suffix is retrieval wording, and the sequence designation requires an admitted total order for the named use.
For any other claim, recover the relation under its governing pattern. State what cited content contributes and use E.10.MOVE when the current relation remains unclear.
Cheap stop. When the reader can recover the situation, first useful operation, payoff, boundary and assurance condition, the opening may be complete. Follow the remaining explanation far enough to establish that it supplies the work the reader was commissioned to do, including any choice left to that reader. A short direct answer can complete a short commission. Use conformance material to verify that guidance. Do not let it replace the guidance.
FPF-governed wording extension. Add heavier assurance, conformance, SoTA, or relation material only when it changes correctness or use: it repairs a false claim, stabilizes the primary EntityOfConcern, supplies a missing concrete contribution, grounds a practical payoff, or states an action-changing boundary.
Cite the pattern that defines or constrains the value at issue.
When an authoring pass claims quality improvement rather than ordinary drafting, keep these pattern responsibilities distinct.
E.22 frames the improvement-oriented quality-evaluation question. The object-under-improvement evaluation such as E.21 or E.9.DA supplies value meanings and stop meanings. C.16.Q repairs overloaded quality and evaluative-characterization wording. C.25 carries engineering quality-family endpoints when those endpoints are claimed. E.23 governs any repeated quality-improvement method.
Closing checklist rows or satisfying a review profile is not by itself quality improvement.
When a pattern claims practical payoff through a visible score or other proxy, name the intended value and the relation by which the proxy bears on it. If the proxy is being treated as the value itself, apply E.13 before admitting the payoff claim.
Quality or projection evidence placement. Development, quality-review, projection, assembly, and landing evidence belongs in its own evaluation, review, projection, or release carrier rather than in the pattern body. Keep it in a pattern only when that work is the pattern's declared EntityOfConcern and intended-reader use.
A Part E pattern may govern FPF authoring, review, evaluation, entry, or publication, but it does not narrate the development of its own current version. Judge placement by the sentence's use, not by a blacklist of words.
Pattern positions across coupled flows. During drafting, E.21 questions may guide a focused author-side check. An overall pattern-quality conclusion requires an independent E.21 evaluation for the declared use. An admission or refresh question uses the applicable E.19 review.
Keep their objects and evidence distinct even when an applicable flow relation connects drafting, review, publication, use, and later refresh. A publication may guide or constrain later Work. Assert the actual Work and its evidence only through the patterns that admit those claims.
Maturity rule. Section completeness is not pattern maturity. A pattern matures when its Problem frame, Solution, worked cases, boundaries, source/SoTA use, relations, consequences, and conformance checks all point to the same usable action guidance for the declared reader and use.
If the reader still needs the DRR, source notes, campaign handoff, or author memory to know what to do, the pattern is not mature for that use.
Primary EntityOfConcern in plain terms. The primary EntityOfConcern of E.8 is the authored FPF pattern: its canonical sections, reader-recognition function, wording discipline, examples, rationale, anti-patterns, SoTA-Echoing, and relations.
Primary working reader. The first reader is an FPF author or reviewer shaping pattern prose for later practitioners and managers. The downstream practitioner is the reader the pattern must ultimately serve, so the authoring guide must model the same recognition discipline it requires.
Pattern Kind In Plain Terms
An FPF pattern supplies action- or judgement-guiding content for a recurring working situation. In ordinary phrases such as “use this pattern” and “apply this pattern”, the acting participant is the agent applying the pattern. The pattern supplies guidance for its work.
Call the pattern content a U.MethodDescription only when it describes one independently admitted U.Method under A.3.2 and that distinction matters to the current claim. Keep the Method and its description episteme separate.
A Solution can guide future action or help choose a Method without establishing that any dated Work has happened. The intended reader, an actual performer System, local system-role classification, assignment, capability, responsibility, authority, result, and Transformation remain separate whenever those claims are current.
When a pattern or worked case does assert dated U.Work, first recover every actual performer's A.13 core: the admitted U.System, local agential kind and criterion, classification, same obtaining assignment, scope, working situation, window, and adequate core evidence. Add a characteristic profile only when its own receiving use consumes it.
Then independently admit the Work under A.15.1 from its performance history, enacted Method, time, and containing System. Add F.6 afterward only when the pattern also needs precise assignment-bound attribution through that same assignment.
A short practitioner sentence may omit identifiers unused by its receiving claim only when every relation the claim consumes remains recoverable.
Pattern application is ordinary shorthand for user-side use: the agent applying the pattern recognizes the situation and uses its Solution to choose the next action or judgement. The pattern body is description-side guidance. Assert a performer, assignment, dated U.Work, result, or U.Transformation only when independently established and material to the receiving claim. A Conformance Checklist checks the authored description and evidenced use but does not replace the Solution.
The pattern's main job is constructive action or judgement guidance. In the opening Problem frame and Solution, state the primary EntityOfConcern, first admissible move, first useful result or practical delta, and the conditions needed to understand that entry.
Develop the grounds and connections needed to understand, choose, adapt and combine the guidance in the full explanation. A first useful move is an entry into that explanation, not its completeness criterion. Auditability, conformance evidence and the rationale for packaging the pattern must not displace its subject.
The subject's own argument is part of the useful contribution: retain the premises, reasons, consequential inferences and limits needed by the declared reader, even when one principle grounds several different moves. Repeat a distinction from another source when the reader needs it to follow or use the argument here. Otherwise refer to the supplying explanation. State what this pattern governs. Do not surround it with an unbounded catalogue of other things.
Name an FPF object by its known kind and state the relation the sentence uses. For a neighbouring pattern, state its concrete contribution and cite the PatternID. Identify an exact episteme, ClaimGraph, edition, or relation assertion only when that identity changes the receiving use.
Put detailed discovery in its dedicated carriers, including README, ToC, E.11, and keep compact pattern relations late in Relations. Do not repeat the same boundary or reference family in small prose variants.
Use E.8 to keep the pattern's positive subject and action guidance first. Apply F.19 on the first draft and to each changed natural span. Its connected reading covers semantic completeness, the contribution and ordering of expressions, preservation of kinds and useful content, and local revalidation. Where a conclusion depends on holon composition or model abstraction, use F.19:4.3.1.1 on the explanation and on any mantra offered for separate use. If the grounds establish the conclusion only for a case or subclass and the pattern promises it more generally, use C.39.GA to restore the missing inference where the practitioner needs it.
For FPF authoring, E.8 adds the pattern-specific question: can the intended reader still recognize this pattern's EntityOfConcern, working situation, first action or judgement, first useful result, and action-changing boundary before optional modeling and assurance? A cold reader must recover that path from the pattern body. F.19 follows the connected passage back to the work the reader was asked to do, including a change of subject, composition scale or model abstraction across paragraphs. Do not copy its method, regression table, or result boundary into a local authoring profile.
If the F.19 reading leaves an FPF value unresolved, take the applicable E.10, E.10.ARCH, E.10.ROLE, A.6.F, F.18, or subject-pattern route and state that source's concrete contribution. Ordinary PatternID citation and “use this pattern” wording remain ordinary. Use an identity-bearing MethodDescription or dated Work claim only when the selected route requires it. Keep development and release evidence outside practitioner guidance unless rewritten as the user's action or boundary.
When an action-adjacent pattern classifies wording or another semio-facing object, connect that classification to the reader's current action. State the admissible use now and any independently grounded boundary that changes that use. Route any other live claim to the FPF pattern that defines or constrains it.
Semio-Echoing is admissible only for one grounded wording-use overread that changes the reader's action or boundary. Keep the pattern's own EntityOfConcern, positive move, and result primary. Use a thin cue to the subject pattern only when its contribution is needed. Do not add a generic counterreading catalogue.
Problem frame
FPF grows through patterns written and revised by authors from many disciplines. Without a shared structure, practitioner-facing use order, and semantic writing discipline, the framework would fracture or become formally uniform but harder to use, violating Pillars P‑1 Cognitive Elegance and P‑2 Didactic Primacy.
Problem
Structural drift, stylistic fragmentation, and revision by visible proxies rather than working use threaten five qualities:
- Comparability – readers cannot align patterns lacking common headings.
- Narrative cohesion – prose swings from dry jargon to informal blog style.
- Practitioner use across revisions – cleanup can erase the recognizable situation, first action or judgement, first useful result, ordinary boundary, or affordable stop while leaving a tidier-looking text.
- Semantic and relation clarity – generic heads, false agency, imprecise neighboring-pattern contributions, and drifting package or relation words can change what the prose asserts or what a reader may do.
- Reviewability after guidance – missing or misplaced grounding, boundary, SoTA, conformance, assurance, and publication-reference material can hide a defect or replace the positive guidance it is meant to verify.
Forces
Solution — One template, enriched by style principles
Canonical Pattern Template
Within each pattern, the canonical section headings SHALL appear in the order below.
For each canonical content section heading (1–12), the <Title> component in <FullId> - <Title> MUST start with the canonical section title or its explicitly accepted alias (case-insensitive match; canonical capitalisation preferred). An optional clarifier after an em dash is allowed (e.g., Solution — …).
The mandatory Footer marker (section 13) is the final sentinel and is governed by H-9 rather than the standard <FullId> - <Title> shape.
Extensibility.
Authors MAY add sections. Prefer expressing them as subsections under the nearest canonical section (e.g., 4.1, 4.1.1 under Solution). If an additional pattern-level section is necessary, it MUST NOT delete or reorder the canonical sections and its title MUST NOT shadow a canonical title.
Mandatory vs optional.
-
Canonical sections 1–13 are mandatory in every pattern.
-
Canonical sections carry content. Authors must not use omission placeholders as section substitutes. When a section is intrinsically small, write the smallest content-bearing grounding, misuse, boundary, or reduced-case statement that preserves the section's function.
-
First substantive authoring seed. The first non-empty authored body of a pattern SHALL already instantiate the canonical section frame by value: title line, header block, canonical sections 1–13, and the footer marker.
-
Seed is not maturity. The canonical frame is a minimum authoring seed, not a mature pattern claim. Before a pattern is used for public, teaching, enterprise, reliance-bearing, landing-input, release-input, or ordinary practitioner guidance, each canonical section must carry enough recognition, action guidance, worked material, source/SoTA use, boundary, consequence, and relation content for the declared use.
A maturity or overall pattern-quality claim needs an independent complete
E.21result for its declared use; an author-side provisional pass or focused repair check does not supply that result. A bounded repair or publication-preservation conclusion uses the checks required for its own question.A file with correct headings, thin bullets, scenario labels, or compressed DRR recap remains a pattern seed until the required content is present. A package that keeps it as a seed can mark it
seedOnly. -
Recognition openings and first-minute working guidance belong inside that canonical frame. Place any retained pre-template entry material there as well. Authors MUST NOT start with a separate opening memo and defer the canonical sections,
Conformance Checklist, or footer marker toE.19, assembly, or review repair.
Template:
- Title line:
<hashes> <FullId> - <Title>, with an optional(informative)note. Use the ASCII space-hyphen-space separator from H-2. - Header block: Type, Status; optional Normativity override.
- Problem frame
- Problem
- Forces
- Solution
- Archetypal Grounding (E.7; show how the practitioner can answer the commissioned question or perform the promised work in a fitting passage, case, comparison or boundary of use)
- Bias‑Annotation
- Conformance Checklist
- Common Anti‑Patterns and How to Avoid Them (grounded misuse, text-invited misreading, or a decision-relevant non-use boundary under
CC-SG.11) - Consequences (Explain what using the pattern changes for the practitioner named in the original commission and how far the grounds for that change reach. If a worked case also shows an effect on another participant, identify whose work it changes and under what conditions.)
- Architectural Rationale (
Rationaleis an accepted title for the same function) - SoTA-Echoing (current-best problem answer; by-value comparison at comparable effort; explicit trade-off and adopt/adapt/reject decision whenever external or internal practice changes the Solution)
- Relations
- Footer marker
Footer marker. End each pattern with a single visible sentinel heading line by itself: ### <PatternId>:End. This makes truncation detectable even when HTML comments are stripped or shown by editors. The footer marker is content-free. Do not place prose under it.
Note. Pattern boundaries are still parseable by scanning for the next pattern heading (## …), but a :End marker helps retrieval pipelines (and LLM prompts) distinguish “this chunk is the whole pattern” from “this chunk was cut mid‑pattern”.
Heading & ID discipline (human tooling + retrieval)
FPF is often consumed through full‑text search and retrieval (RAG). A reader or an LLM may see a subsection without its parent headings. Headings must therefore be self‑identifying.
The following grammar governs Markdown publication forms authored under this E.8 edition. Its heading levels, separator and sentinel spellings are necessary subject conditions under E.5.1; they do not make the represented conceptual objects depend on Markdown.
Markdown without raw HTML. Markdown publications in FPF Library, and other Markdown publications governed by this profile, MUST NOT contain raw HTML elements or comments, including empty anchors and HTML used for layout. Use Markdown headings, links, tables and images, with the mathematical notation declared by the publication.
Show literal HTML discussed by the text inside code spans or fenced code blocks. Such code, mathematical angle brackets and Markdown autolinks are not raw HTML markup.
The restriction covers complete publications, including framework editions, non-framework References, Readmes and other Library entry material. Preserve information when converting an existing presentation; removing a tag does not justify removing the content it presents.
H-1 (Heading shape). Every pattern heading and every subsection heading inside a pattern SHALL follow:
<hashes> <FullId> - <Title> (optional note of non‑normativity)
Exception. The Footer marker is a sentinel heading and is governed by H-9, not by the standard <FullId> - <Title> shape.
H-2 (Heading separator). The canonical separator is ASCII space-hyphen-space, as in <FullId> - <Title>.
Previously authored headings may use Unicode separators, as in <FullId> – <Title> or <FullId> — <Title>. Tooling SHOULD treat these as migration candidates, and authors SHOULD migrate touched headings to the ASCII separator in H-2.
H-3 (FullId). FullId is the complete address used by this heading grammar.
For a pattern heading it is the PatternID (e.g., A.2, E.10.D1).
For headings inside a pattern, append dot-separated ordinal section numbers after the colon (:) (e.g., A.2:4.4, E.10.D2:3).
Exception: the Footer marker uses the reserved sentinel token :End as defined in H-9.
The colon (:) is reserved for section paths and MUST NOT appear in PatternIDs.
Retained published addresses. When repairing an already published heading, preserve its established address and the contribution it denotes. An existing non-ordinal address is a compatibility exception, not a form for new addresses. Establish that exception from the exact earlier publication and its target; a permissive parser match does not establish it.
An absent address receives a new ordinal address chosen by the source author. A local component, clause or extension name remains its own declaration under H-10; it does not by itself supply the missing subsection address.
PatternID segments may be numeric or mnemonic. When the surrounding text identifies the framework, the complete PatternID identifies one pattern in that framework. The shape of its segments does not by itself state the pattern's title, meaning, Part, publication position, dependency, Method relation, or use order. A mnemonic segment may help recognition but does not define the pattern.
Whether a PatternID stays with a changed pattern is an authoring decision, not a grammar decision. For a DPF, use E.4.DPF. Use E.11.PFP to show current publication position separately. When the surrounding text does not already identify the framework, name the framework together with the PatternID. Add the edition when the reference must select the body published in one edition.
H-4 (Ordinals). Ordinals in section paths SHOULD track the canonical template numbering (1 = Problem frame, …, 13 = Footer marker) to maximise cross‑pattern comparability. During refactors or in previously authored patterns, ordinals MAY be local. In that case, the canonical section title at the start of <Title> is the semantic key. Readers and tools MUST NOT infer section semantics from the ordinal alone.
Architectural Rationale is the preferred title of the Rationale function. Rationale remains an accepted alias. Both identify one canonical content section, so a pattern carries exactly one of them.
When an existing heading is retitled, repair title-dependent links and direct consumers under E.8:4.1.2; retaining its ordinal alone does not preserve a Markdown return.
Keep addresses through change. Inserting, moving or retitling a subsection does not assign its address to a different contribution. Keep existing addresses when their contributions continue. Give a new contribution an unused address. Do not renumber unaffected subsections to restore display order or reuse a retired address for different content. For a genuine replacement, split or merge, decide which contribution continues and state which returns have changed or remain unresolved.
For example, when a new explanation is inserted between existing sections 4.1 and 4.2, an available 4.3 can name the insertion while 4.2 still names its former explanation.
Before publication, compare the earlier and revised targets, preserve the earlier carrier fragments through compatibility facilities permitted by the selected publication form and supported by the access route, and update maintained direct links. Test duplicate-title suffixes and alias collisions as well as visible FullIds.
A saved link must return to its intended contribution or expose that the return is unresolved; reaching a different heading silently is a failure. The author judges the contribution. A link checker can verify the declared correspondence.
Note: the Footer marker is exempt from ordinal encoding. It uses the reserved token :End (see H-9).
H-5 (Where kind and normativity are declared). Pattern kind (for example, Architectural or Definitional) MUST be declared in the Header block, not encoded into the heading text. Normativity (normative or informative) MUST also be declared in the Header block when it deviates from the default.
If a reminder is needed for readers, authors MAY add a short parenthetical note at the end of the heading, for example (informative) or (non‑normative), but headings MUST NOT use square‑bracket tags.
H-6 (Heading levels). Heading levels MUST preserve a fixed offset between structural layers (Part or Cluster (flat) → Pattern → Pattern sections):
- Part and Cluster headings MUST use
#(level 1) across the file. - A Pattern heading MUST use
##(level 2). - Inside a pattern, each nested section MUST add exactly one
#per level (e.g.,## A.2 - …,### A.2:2 - …,#### A.2:2.1 - …).
H-7 (Ellipsis discipline). Authors MUST NOT use three consecutive dots (...) as punctuation in headings or narrative prose. Authors MUST use the Unicode ellipsis … (U+2026) instead. For editorial elisions in quotations, authors SHOULD prefer […] to make the omission explicit and distinguish it from retrieval truncation.
Exception: literal three‑dot sequences that are part of an external language’s syntax MAY appear only inside code spans or fenced code blocks.
H-8 (Normative keywords). The key words MUST, MUST NOT, REQUIRED, SHALL, SHALL NOT, SHOULD, SHOULD NOT, RECOMMENDED, MAY, and OPTIONAL are to be interpreted as described in RFC 2119, as clarified by RFC 8174 (only when capitalised). Authors SHOULD avoid informal deontic phrasing (“need to”, “is required to”) in normative clauses.
Deontics vs admissibility. Use RFC keywords only for deontic obligations (requirements on authors, reviewers, implementers/tooling, or published pattern or companion texts). These concern things an agent can choose to do or omit. Do not use RFC keywords to state definitions, structural invariants, typing rules, or other admissibility conditions of the modeled world.
When you need an enforceable constraint that is mathematical rather than deontic, express it as a non‑deontic predicate using one of: Definition:, Invariant:, or Well‑formedness constraint: (optionally with formal quantifiers). Prefer mathematical terms like cardinality 1..1 (total), 0..1 (partial), or 0..n over deontic adjectives like “mandatory or optional” when the intent is cardinality, not duty.
Admissibility predicate discipline (recommended shape).
When expressing admissibility or validity constraints as predicates (Definition:, Invariant:, or Well‑formedness constraint:):
- Authors MUST NOT use RFC keywords inside the predicate block.
- Authors SHOULD give each predicate a stable identifier and short name (e.g.,
RA‑1 (Locality),RE‑3 (Method gate)), so that Conformance Checklist items can reference it without re‑authoring the rule. - Authors SHOULD write the constraint as a declarative predicate with a truth condition (optionally quantified), for example “every selected interval lies within the declared qualification window”, rather than as “X MUST …”.
- If the constraint needs to be checked as part of pattern conformance, authors SHOULD reference the predicate identifier from the Conformance Checklist, and state validator behaviour when relevant, rather than duplicating the predicate with RFC keywords.
H-9 (Footer marker sentinel). Footer marker SHALL be a single heading line whose FullId is the pattern ID followed by the reserved sentinel token :End (no ordinals, no title, no square‑bracket tags):
### <PatternId>:End
Apart from a retained published address under H-3, it is the only allowed heading inside a pattern whose section token is non‑numeric. It MUST be the final line of the pattern and MUST NOT carry any prose. Tooling and readers MUST treat it as a boundary sentinel, not as a semantic section.
H-10 (Publication-token classification and addressability). Before emitting an FPF-governed token as a reference, authors MUST classify it under exactly one of these seven E.8-local publication-token classes and use the matching form:
PatternRefuses one PatternID to name a pattern that continues across editions of the framework identified by the surrounding text. In the assembled publication being checked, it resolves to one complete H2 body, one matching:End, and a truthful ToC status for that PatternID. A reference intended to select the body published in one edition also names that framework edition. A structural checker may verify and report publication conformance but does not establish the pattern's identity, status, or authority.PlannedCatalogEntrynames an explicit future catalogue commitment. It has no current pattern semantics, governing force, prerequisite force, or addressable body. A useful prose mention MUST sayplannedorfuture, and a current semantic dependency MUST cite existing content that supplies the needed definition, constraint, test, method, or other rule, or state the current gap.SectionRefnames one exact heading path inside one current pattern, one framework publication unit declared underE.11.PFP, or one non-framework Reference declared underE.11.DSG. Authors and tooling MUST read the complete section identifier and its declared scope before examining any substring. For example,STR.Reference:1returns to the Strategy Architectural Reference. It does not declare a pattern namedSTR.Reference. Similarly,Engineering.Reference:2.1returns to a section of the Engineering Suite Reference without declaring a pattern or a framework namedEngineering.Reference.LocalDeclaredIdnames an exact declaration within one pattern, such as a conformance clause, component, interface row, or predicate. Its scope is local unless an explicit stable anchor or a separate promotion decision establishes wider use.LocalAliasnames a declared compatibility alias and resolves to its declared canonical local target.PatternFamilySelectorselects a navigable pattern family using canonical spelling<base>.*. It requires a current base pattern and at least one current matching member and MUST NOT stand in for one exact governing target.NonReferenceTokenclassifies a schematic example or ordinary local prose/code that neither occupies a reference-bearing position nor declares a local public ID. This class denotes no reference. Key-like typography or backticks alone do not change that class.
A portable reference MUST let the reader recover the intended document within its publication set and the intended contribution from the reference and the context that travels with it. Preserve the complete PatternID or SectionRef when the contribution has one. Add the publication set, the document's full title or its explicitly declared DocID only where that identification is not already unambiguous. Do not repeat a qualifier that the complete address and retained context already supply.
For example, once FPF Core is identified, C.39:4.1 is sufficient. Use FPF C.39:4.1 when the document qualifier is needed. Resolve the complete address under its declarations rather than inferring its scope from a fixed substring.
A DocID identifies a whole document. It is not a new PatternID, an edition identifier, a publication-unit identifier, a globally unique name, or an eighth publication-token class. An existing framework code may serve as its DocID only through an explicit declaration.
A whole-document reference needs only enough information to identify that document. A reference to an ordinary unnumbered heading adds its title without inventing a PatternID. Name a selected edition separately when its exact published body matters. When extracting a fragment, restore any identifying context that will no longer accompany it.
A URL, file path or service locator is an optional projection to an access carrier, not the portable reference's identity. To resolve a reference, recover the document and publication set from the complete reference and retained context, then match the complete local address or named unit in the available text.
A catalogue may associate those declarations with current carriers. A changed carrier does not change the reference. Preserve any more precise valid SectionRef already stated in a link's visible label even when the URL lands at its enclosing body.
If the document, target or requested edition cannot be recovered unambiguously, report that unresolved reference rather than silently substituting another document, section or edition. These requirements add no URI syntax, HTML anchor or service API.
Resolution and checking are declaration-first and context-sensitive. Authors and tooling MUST NOT split complete SectionRefs, strip a local-ID prefix, promote a local symbol by visual resemblance, or replace these classes with an ignore list. An unresolved token in a reference-bearing authoring form is an error. Ordinary code or local wording is not silently upgraded to a reference.
H-11 (Assembled Part boundaries and title agreement). In the assembled publication, every compact ToC Part label MUST be a bold separator with a blank line on both sides, not a duplicate structural Part heading. Its title and ASCII space-hyphen-space separator MUST agree exactly with the corresponding # Part <letter> - <title> body heading. A reserved body Part that has no compact ToC table, including current Part H, does not require an empty compact label or table.
Problem frame names the working situation. Generic Context and universal U.BoundedContext do not identify the actual value a claim needs.
For each use, go directly to the rule that supplies the needed value. Recover source-local meaning through F.0.1, use F.1 to select answer-changing sources, state ClaimScope through A.2.6, and use A.1.1 for an admitted bounded-model use.
Add F.17 only when a durable address or basis relation is needed, F.9 only for an obtaining Bridge between two exact local senses, and the applicable plane relation for a ReferencePlane claim. Otherwise leave the relation unasserted rather than inferring it from a shared word, source, or context.
Preserve Pattern Use Value Across Material Revisions
A revision is material when the change can alter what a working reader recognizes, does, obtains, or must stop doing, regardless of whether the change is labelled as cleanup, clarification, terminology repair, or ontology alignment. Treat the revision as material when it can change at least one of these values:
- the primary
EntityOfConcern, governed kind, direct relation, claim kind, or scope; - the recurring situation or practical question that lets a reader recognize the use;
- a Solution action, action condition, result kind, first useful result, stop, return, risk disclosure, or stronger-neighbor handoff;
- the definition, constraint, test, method, cited-pattern contribution, split, merge, relocation, or source/SoTA stance that changes what the reader may do;
- the asserted commonality, member set, membership rule, order, or governing premise of a list; or
- ordinary first-use affordability.
For this comparison, the earlier edition is the exact accepted pattern edition that this candidate is intended to replace for the declared use.
A formatting correction, spelling repair, citation repair, exact mechanical rendering, or wording change is not triggered only when the smallest comparison of the earlier edition and proposed text shows that all these values are preserved. A clean comparison needs no additional positive ledger, evidence table, or pattern section.
Physical line count, file size, section count, inventory rows, and the author's label for the change do not establish materiality.
Use one bounded material-revision loop over the actual prose. Before treating a materially revised pattern as authored:
- Recover the useful earlier-edition use at idea level: the recognizable situation and intended reader, first admissible action or judgement, first useful result, action-changing boundary or stop, and any domain claim, example, or relation needed to perform that move. Classify a changing or disappearing earlier-edition use only as retained, a valid outcome whose defective mechanism is repaired, an explicitly authorized retirement with a corrected action or boundary, or unsupported residue.
- Draft the candidate's positive practitioner path in domain-recognizable language before guards: governed subject, recurring problem, action the reader can take, first useful result, and next action-changing condition or stop.
- Compare the earlier edition and proposed text at comparable application effort. Preserve every useful earlier-edition move or deliberately replace it with an at-least-equally-usable action, result, or boundary. Admit a candidate-only use only from an exact accepted decision, source/SoTA stance, finding, or working need.
- Apply
F.19to each changed natural span. Remove exactness intensifiers, invented counterreadings, role or process wrappers, formal identities, and assurance apparatus that fail its contribution test, while preserving every kind, relation, use, and action-changing detail. Keep ordinary pattern-use wording ordinary; open a deeper FPF route only for an unresolved value. - Check that recognition, first action, and first useful result still precede optional modeling, evidence, conformance, and assurance work.
F.19is the common semantic pass over the changed span.E.10supplies cues and routes unresolved FPF wording to its governing rule, not a second normal-pass algorithm. - For every changed public or consumed interface—entry wording, input or result, field or position meaning, action order, stop, return, or reconsideration condition—repair each determinate stale ToC or README cue, example, relation, and direct consumer in the same authoring increment. Find consumers by the meaning they teach or use. A shared word, identifier, or nearby reference is not enough.
Earlier-edition and candidate-only uses remain different bases, and both may be present in one revision. Compare that exact earlier edition with the candidate edition. An earlier-edition use keeps its earlier-edition basis and one of the four classifications above. A candidate-only use keeps its exact accepted basis. Do not classify a candidate-only use as an earlier-edition use or invent history for it.
Treat a selected use as required when its loss changes action or boundary, and as optional when it demonstrates breadth only. Backward compatibility alone is not improvement, and a candidate-only promise is not improvement until the text supports its executable use.
Use desk replay by default and escalate to a cold reader, AI-agent, or observed-work check only when ambiguity or consequence justifies it. If later independent review needs a recoverable note, use the smallest existing authoring source. Do not create a card, score, universal schema, or one written row per idea.
Test first-use affordability by checking whether the positive Solution supports this short rendering:
This rendering explains the pattern. It does not claim that actual work is linear. Use an optional local mantra only when it improves recall, and show one ordinary traversal only when several rows materially improve explanation. Choose the smallest form that keeps the action, result, and boundary recoverable.
Explanatory rows may fade as competence or task demand permits, but an independently action-changing condition or boundary may not. An explanation retained for reuse can concern the actual subject, question or proposed continuations before CGUS qualification. Use C.2.1 when its episteme identity matters. A demonstrative slice about a qualified CGUS is a separate episteme under A.22.CGUS, whose use depends on that qualification.
Put a subject-side check immediately before the continuation it changes, and keep authoring, review, quality, and release checks outside the subject Solution.
Resolve authoring lists with F.19. When a list can change pattern use, apply the same connected F.19 reading used for prose. E.8 keeps only the authoring effect. Make clear which working question or operation the passage serves when the reader first needs it. A case can open the reasoning. If the reader must devise an answer, keep that choice open and show how the material bears on it. Declare a normative closed set as closed under its governing rule. Signal examples as non-exhaustive when a plausible reader could mistake them for a classification. Do not let a noun series or catalogue replace the Solution.
Do not add a second enumeration taxonomy or a per-member result form. E.10 may cue a suspicious head or series, F.19 decides its membership semantics and discourse load, and the applicable subject pattern settles any unresolved kind, relation, or normative set.
Decide Whether a Narrower Contribution Changes Practice
Use this when a broader available contribution and a proposed narrower contribution both appear to answer the same recognizable working situation. State the intended reader, use, and scope. Apply both contributions at comparable effort and find the first difference in what the reader notices or decides, does, needs or checks, obtains, or uses as a stop, return, or retry.
A narrower title, domain noun, paraphrase, or extra example is not enough by itself. If no action-changing difference remains, omit or merge the narrower text and point to what already answers the situation.
If the two contributions address different situations, state that boundary before deciding their relation.
Developing a Method often involves methodological synthesis: selecting, adapting and connecting available operations and their conditions. C.39:4 develops that general work, including explanation for later use and selective change, without requiring framework authorship.
A working account of the relevant objects and relations can develop with it; use F.0.2 for semantic synthesis across source ontologies and A.7.1 for a consequential ontological repair. E.4.CM applies the general way to a framework's connected Methods and their public explanation.
The authored pattern is an episteme about the Method and its use. Its content includes the reasons, conditions and relations needed to understand and adapt the way. The canonical template organizes that content rather than supplying missing operations.
For a proposed composite Method, compare the whole's organization and explanation as well as its constituent operations. Apply the same use comparison to an account of independent Methods: compare the explained result uses, shared conditions and returns without assuming one composite. Existing operations do not by themselves supply selection, joining, coordination, adaptation or return rules. C.39:4.3–4.6 develops those connections. E.4.CM adds their framework-authoring treatment. A supplier list or an expert's private successful reconstruction does not establish that the public explanation is sufficient.
An action-changing difference shows that the contribution is distinct. It does not show that the contribution is worth keeping. Retain or merge it only when the changed action, result, boundary, or saved source reconstruction is warranted and useful for the declared reader, use, and scope under the applicable domain, evidence, currentness, affordability, and architecture checks.
Use only the checks that can change this decision. Repair or reject a distinct contribution that is wrong, stale, unsafe, unsupported, incompatible, or needlessly burdensome. Keep an explicit gap when no acceptable contribution answers the situation.
Naming a dependency does not settle the comparison. Say which available result supplies the reusable part, what kind of result it is, which product and edition or current state supplies it, how the reader uses it, and which currentness or availability condition can change that use. State maintenance separately only when it changes the receiving use. Then preserve any remaining domain problem, filling, constraint, relation, evidence limit, return, or discovery need without copying the general rule.
When reuse or a gap closes the reader's question, state which of these applies:
- Use an available result. Name the result, what kind of result it is, the product and edition or current state that supplies it, the receiving use, and any currentness or availability condition that can change that use. The supplying product may be an FPF, DPF, LPF, or a separate non-framework product. If maintenance changes the use, state its separately established relation and evidence.
- Use a MethodDescription. Name the public description, the Method it describes, and how the reader uses the description to select or perform that Method. State availability, currentness, or a separately established maintenance fact only when it changes that use. Do not report the expected result as already obtained.
- Use a direct source as evidence. Name the source, the claim or decision it supports, the receiving use, its limits, and a usable locator. Source availability is not result production.
- State a named unavailable result. Name what is missing, the action or decision it blocks, the missing condition, and the observable condition for retry.
For example, "feed the animals" may be true for both a mouse and a tiger yet fail to tell the feeder what food to give. Grain and meat change the action, so keep or link the animal-specific guidance when that difference is warranted for the declared use.
By contrast, a pump-maintenance restatement of an available evidence-use contribution adds nothing if it changes only pump nouns and one example. Omit or merge the restatement, point to the maintained result that already answers the situation, and judge any promised maintenance-framework coverage separately.
A tiger-feeding proposal may instead require manager approval and a laboratory certificate before every ordinary feeding. That proposal changes the feeder's action, but if no safety rule, evidence limit, law, or observed failure warrants the burden for the declared use, reject it or repair it to the smallest warranted check. Distinctness alone does not preserve it.
A result maintained outside the receiving framework may answer the reader's use without becoming part of that framework. In a package-coverage account, count that external result only when the exact result and supplying product, receiving use, practical discovery route, and any material currentness or availability condition are explicit, and say that the result remains external. Otherwise keep the promised family as a gap or omission.
When the resulting stable pattern set materially changes a promised problem family, obtain a current E.4.DPF.DA D12DomainProblemFamilyCoverageAdequacy result for the resulting exact DPF or LPF edition. Reuse a matching current result when the exact edition, promised families, declared use, relied-on results, and relevant conditions did not change. Do not record proof that a revisit happened.
Carry the content functions across Method-description scales
Begin with the practical explanation developed for its intended use, using C.39:4.5–4.6 and the relevant professional Methods. Framework authoring applies the twelve functions below to make that content understandable, examinable and reusable as a pattern or connected framework account. The general development of a Method does not depend on choosing this format. A sufficient procedure, calculation or ordinary description can serve another receiving use.
Use the twelve substantive functions in the canonical template as authoring questions for a whole FPF, DPF, or LPF and each selected substantive profile. E.11.PFP:4.7 governs their public answers, inheritance, and placement in the framework's publication units. The same functions can describe a broad Method, a composition of Methods, or a narrower use. The declared subject and applicability determine the scale.
The Solution explains the organization and use of the described Methods: their contributions, relations, and the results that make a next move possible.
Architectural Rationale explains why that organization and those choices serve the declared use, which alternatives were considered, their trade-offs, and the conditions under which another choice becomes preferable. Preserve shared source explanations there when users need them to understand or adapt several patterns together.
The same explanation must make recoverable how constituent actions perform encompassing work where that connection matters to learning, performance or change. Show what larger work is already being done through the selected action, what constituents make that action possible, and which conditions of the whole constrain it. Preserve genuine result-before-use dependencies alongside the constituent–whole relations of the relevant Method and Work structures.
A representative case should let the reader change the action when a relevant condition of the whole changes. Use B.1.5.EW for recovery and B.1.5.RS for a constituent replacement. Do not require a separate level diagram or exhaustive decomposition when the connection is already understood.
The pattern heading, header block, section grammar, and footer apply to each individually declared pattern. The whole-framework account uses the publication form in E.11.PFP. Its content questions do not turn every publication unit or intermediate group into another pattern. Its Architectural Reference subsection headings identify the framework, publication unit and ordinal path through E.11.PFP:4.7.1, so an isolated excerpt remains locatable without being classified as a pattern body.
Keep generality, specialization, Method composition, reuse, bounded-use projection, and publication grouping explicit under E.8:4.2.2. There is no prescribed maximum depth or exclusive-parent rule.
Choose a title that helps the reader select the pattern
Choose the title from the working question the body answers. Read its Problem frame, Problem, action-changing Forces and Solution together. Recover the intended reader, governed object, useful result and nearest non-use boundary before trying shorter wording. The title should let that reader consider the pattern without already knowing an unfamiliar solution name.
Use a recognizable question, a result under distinguishing conditions, or a familiar method name. Preserve the object and condition that separate the pattern from its nearest neighbors. “Test a method” cannot distinguish obtaining trial observations, checking description coherence, judging fit in another situation and comparing practical worth. A single striking example should not narrow a more general construction, and a failure symptom should not hide constructive use before anything goes wrong.
Keep a choice open when making that choice is part of the Solution. A method that first decides whether further evidence is worth obtaining can finish with an existing answer. “Design experiments” would hide that first result. Name whether and how to experiment when both questions belong to the method. Conversely, retain a direct action title when that action is already the justified entry.
Keep an established technical term where the intended finding or discussion needs it. A quotient method can pair a phrase about treating objects as the same with quotient. A definitional pattern can retain its technical name and add a short recognition phrase. The technical designation and recognition phrase explain the same subject. Use the existing ToC keywords and query cue for additional familiar expressions. If the intended search sees only headings, a body-only term cannot support that search.
Compare the candidate with the present title and the closest plausible neighbors at the actual entry: title alone, complete ToC row or other displayed cue. Include a use the pattern should attract and an ordinary case it should leave elsewhere. Open the bodies to check that the proposed result and boundary survive. Retain a sufficient title, or improve only its existing cue when that resolves the difficulty. A question shape, complaint form, shorter length or uniform style is not a reason to rename.
Finish with the chosen title and its truthful discovery cue. Keep the PatternID unless the governing continuity decision changes identity. Repair title-dependent fragments and direct navigation consumers under §4.1.2. E.11 supplies comparison under the reader's actual exposure. A clearer title supports consideration. The Solution must still supply the method and any connected explanation needed to use it.
Stylistic Principles (S-0 … S-19)
Use the principles to make the explanation coherent and readable. Semantic discipline takes priority over engagement. Hooks, quotable lines, Plain restatements, and didactic images may improve recognition, but readers must be able to recover any ontological, evidence, causal, assurance, bridge, gate, work, decision, or admissibility claim kind or admissible-use boundary they carry through the governed Tech reading or named neighboring pattern.
Ordinary Plain prose without that claim kind or admissible-use boundary stays ordinary prose.
S-0 (Governing-claim flow) — explanation
Give readers the working question by the point at which they need it to interpret what follows. A task can open with evidence from which they will have to construct the claim. An instruction for using an established result can give that result directly. Add history, related patterns, examples, imagery, or a recall line only when it helps the intended reader understand or use that claim. A prerequisite may come first when the reader needs it to interpret the claim or act safely. Apply F.19 when atmosphere, coordination, or rhetorical scaffolding delays the governing message.
Recognition text and assurance text
Every canonical pattern SHALL stabilise one primary EntityOfConcern, relation record, or claim record early enough that a cold reader can tell what kind of thing the pattern governs. If ordinary forms vary (note, sheet, guided UI, rendering, review aid), the text must identify which are presentation forms of the same primary EntityOfConcern, relation, or claim and which name a different act, process, work-result record, or governing companion.
Recognition and assurance texts may refine that selected item differently, but they must not silently swap the central kind.
If a pattern uses a broad umbrella or head together with a narrower operative branch, the text must tell a first-time reader what the broad head names, which narrower branch applies, what primary EntityOfConcern, relation record, or claim record the branch governs, which exact action assertion and predicate apply, and what wider work or process remains outside the pattern. A qualifier alone does not restore that stack.
Recognition text and assurance text name two reading-order functions carried by existing sections or projections inside one pattern. Additional didactic scaffolding remains optional and is justified when the family is especially easy to misuse, easy to over-read, or hard to teach. The instruction needed for the pattern's declared use is part of its substantive contribution under §4.2.1.1, not optional scaffolding.
The recognition text is the first-reading text.
It is the part of the pattern that lets a cold working reader recognise the situation quickly enough to decide whether to keep reading.
It should start from a subject-domain or practice moment before internal taxonomy whenever the pattern is meant to guide work beyond internal canon maintenance.
In practice it usually appears in an early Use this when line or equivalent opening, plus the upper parts of Problem frame, Problem, Solution, Consequences, and nearby worked slices.
Its job is to make visible:
- what ordinary working situation this pattern is for;
- what goes wrong if the pattern is missed;
- what the pattern buys the reader in practice;
- when this is not the right pattern;
- what primary
EntityOfConcern, relation record, or claim record stays fixed; - and, when technical terms must appear early, a pairwise plain gloss for each early FPF-governed technical term.
The assurance text is the second-reading text. It carries the heavier FPF-governed material that makes the pattern reviewable and auditable:
- declaration blocks and typed fields when those are part of the pattern's declared conformance or boundary claim;
- representation ontology, EntityOfConcern discipline, or primary-EntityOfConcern discipline;
- any minimal modeling or mathematical lens that keeps the primary
EntityOfConcern, relation record, or claim record stable; - guidance or check material, invariants, admissibility, and stop or neighbouring-pattern conditions;
SoTA-Echoingwhen it carries explanatory work;- and the review hooks that let a broader or more consequential interpretation or use be checked.
The assurance text may sharpen, justify, and discipline the recognition text. It must not silently replace, strengthen, or universalize the claim that the recognition text made visible. If the recognition text says “this pattern helps with a bounded working situation”, the assurance text must not quietly turn that into an unbacked carrier claim, unbacked guarantee, or broader universality claim.
If a pattern claims universal or transdisciplinary status, that claim must already be visible in the recognition text.
It is not enough for universality to appear only later in a guidance or check sheet, declaration block, or SoTA-Echoing rationale.
Make the breadth of a claim recognizable through the conditions and relations that hold across its stated uses. Show different situations where they reveal a material difference. Vary the reader when the claim concerns who can use the explanation. For a claim across disciplines, explain what enables the same operation or inference in the receiving disciplines. One argument can cover all disciplines that meet its premises. Where that depends on a disciplinary premise, show what establishes it in the receiving practice. The separate three-domain minimum in A.8 applies to universal-core U-kind admission.
When a compact matrix helps readers recognise the method across disciplines, F.16 is the preferred template. Show what each situation illustrates and what warrants the method across the claimed range. The general argument can be given within a worked case. When examples share a constraint or supply different premises of one conclusion, show how those premises support it together.
If SoTA-Echoing carries an FPF-governed claim, the practical implication of those rows should be recoverable from the recognition text and case bank rather than remaining a late-only justification layer.
Additional didactic scaffolding helps readers with a particular interpretation or learning difficulty after the necessary instruction is present. Useful additions can include:
- grounding that shows the subject of a change, what changed and the grounds for that conclusion. Evidence for a change in an illustration, wording or access condition depends on the property at issue.
- at least one boundary-heavy example or anti-example when nearby or companion patterns are easy to confuse;
- reviewer guidance that says what to inspect first, names the neighboring FPF pattern that defines or constrains the failure mode, and gives the project-side FPF kind and reference by value for the claim or effect;
- local mini-definitions or glossary material for recurring terms that would otherwise be recovered only from project context.
Provide enough recognition, assurance and, when needed, additional didactic material for the reader to understand the pattern, apply it in ordinary cases and recognise when a heavier review profile is required.
Provide enough explanation for the reader's use without requiring a complete reading
Use this authoring move when the intended users differ in relevant preparation, when a complete account burdens a current application, or when a compact instruction leaves its selection, adaptation or combination unexplained.
The useful result is a sufficient connected reading for the present use, with the fuller content and its suppliers still available for the uses the pattern promises.
Distinguish preparation from coldness and purpose. Under F.19, a cold intended reader lacks the author's private context; that reader can be an expert. The plausible-reader test uses the knowledge and task the text reasonably presupposes.
Under C.2.8, identify the useful structure to recover and the preparation, operations, help, access and effort budget that matter. Knowing the subject, knowing this framework's distinctions and knowing how to change a particular Method are different capabilities. A model name or broad professional qualification does not establish them.
Keep the full contribution and the present reading distinct. Maintain the explanation needed for the declared uses in the pattern language and its public suppliers. A body may rely on a common worked application or an already explained constituent Method when the contribution and usable return are clear. The reader need not load every such explanation for every use.
A short problem–solution instruction can serve recall or a familiar application; its success does not establish that it supports learning, adaptation or combination. Canonical headings and additional words likewise do not establish sufficient content.
Construct the reading from the work the reader needs to accomplish:
-
Keep the receiving question fixed. Determine what the reader must recognize, infer, perform or change here. Use available evidence or explicit assumptions about relevant preparation.
If a missing prerequisite changes the next explanation, expose it and provide the needed support or return. Do not silently raise the reader's preparation or narrow the requested work to make an inadequate explanation appear sufficient.
-
Carry the connected instruction. Retain the operation with its necessary inputs, distinctions, conditions and use-changing reasons. Ordinary prior knowledge may supply a familiar notation or inference. Supply the relation or way of constructing it that the publication undertakes to teach. If the reader is to devise an arrangement, give access to the premises and means of making that choice while leaving its outcome to the reader.
When several Methods work together, retain what connects their results and how constituent actions enact the encompassing work; independent reminder phrases do not explain that connection.
-
Choose what is needed now and what can be reached later. Keep a meaningful inference together. An applicability condition that changes the current action belongs with that action even when its derivation is elsewhere.
Give a direct return to the fuller explanation and state the situation that calls for it: an unfamiliar operation, changed premise, disputed ground, or proposed adaptation, for example. A reader who cannot recognize such a condition needs enough explanation or assistance to recognize it before relying on the compact instruction. One sufficient reading path needs no additional version.
-
Compare the actual use and burden. Follow
C.2.8,E.12andF.19for the appropriate recovery question and evidence. Keep the task, preparation, assistance and access conditions comparable; distinguish an expert walkthrough, an actual reading and an assisted retry.Include an affected changed case when the promised use includes adaptation. Consider navigation, source retrieval and reconstruction as well as the amount of text. Preserve a sufficient existing account when another form gives no worthwhile improvement.
-
Repair the missing contribution or the reading arrangement. Supply an omitted relation, make a consequential condition visible, explain a prerequisite or improve access to a supplier.
If the required content cannot be used reliably within the present budget, change the support, work decomposition, reader allocation or budget. Leave the original unmet use explicit until it is supported. A lower-cost answer to a different question does not satisfy it.
Develop a missing instruction. Work through the promised use with a concrete input and the reader's stated preparation. At every consequential inference, choice or action, ask what the reader has at that point and what the next result requires.
Follow an unexplained requirement backward to its producer, then attempt the connection forward. Include a branch or a change of scale when the promised use depends on it. A jump from a source result to a conclusion, or from a constituent's success to the whole's success, is a candidate gap even when each sentence is clear.
Determine what is missing before rewriting. If the author performed an unreported interpretation, selection, conversion or coordination, recover how it was done and why its result was acceptable. Write the determining distinction and operation, show the intermediate result, and explain how the next operation uses it.
For a judgement, expose the evidence and criteria that distinguish materially different answers; do not invent a mechanical rule where professional judgement remains necessary. A worked example should reveal the reusable reason for the step, not merely the author's chosen answer.
If the operation is available elsewhere, supply a usable return to the explanation and make the required input and resulting contribution clear. Ordinary knowledge reasonably presupposed for this reader need not be retaught.
If the operation itself is unknown or its necessary evidence, means or capability is unavailable, prose repair cannot provide it: develop the operation through the relevant subject Method, use C.39.RO for a missing reusable construction, or keep the exact use-changing gap explicit.
The instruction “adapt the method” does not supply an adaptation being promised to the reader. If the complete explanation already exists but is lost in the reading arrangement, restore access and the conditions needed to choose it.
Recover the recurring problem–conditions–remedy relationship from the case. Separate what the reader must preserve from values or choices that change with another instance. Work through a consequential variation to expose that distinction, especially when the pattern promises adaptation or combination.
Repeatable use need not give an identical output: changing valid inputs can rightly change the result. Where the claim is reproduction of a particular research result, specify the intended comparison and agreement conditions rather than substituting successful use of the general Method.
Continue through the promised result and its material returns, not just the first repaired sentence. Keep independent uses distinct when they need different contributions. Stop elaborating when the intended reader can obtain the necessary operations and connections under the stated preparation and access, or when an explicit unresolved contribution limits the claim. More elementary detail is unnecessary when it changes no declared use.
Apply the connected F.19 reading to the resulting explanation and its actual public presentation; that wording judgement does not establish the Method's subject-matter effectiveness.
Worked authoring repair. The date-conversion case in §4.2.1.2 develops the missing source-to-order selection and then constructs the expression that makes it usable. The converter can perform its operation once the applicable source supplies the order; ambiguous characters alone supply no such interpretation. The case preserves both correct conversions and the changed-source stop while comparing the forms that carry this connection.
Divide a method account from pedagogical companions by the work each supplies. A pattern can state a Method adequately for competent readers while a pedagogical companion provides the fuller teaching needed by a less prepared audience. Completeness is relative to the declared use and the sources that supply it; it does not require every beginner explanation in the pattern body.
The pattern's contribution still includes its problem and frame, forces, operative distinctions and connections, conditions, grounds for important choices, and criteria and limits of the result. Keep or directly supply the worked and changed-condition reasoning that the declared use needs. A title list or imperative sequence is insufficient when a competent reader must invent why its steps connect.
A competent author of instruction should be able to reconstruct what must be learned and preserved from this method account and its named accessible suppliers. State relevant prerequisites rather than presupposing an unspecified “smart reader”. A reader may know the domain but not the framework, or know the framework but lack the subject skill. Competence in the subject Method also does not automatically supply competence in teaching it.
A companion may explain prerequisites, add analogies and more examples, arrange a progression of tasks, provide practice and feedback, and gradually remove assistance. It can also supply extensive derivations and worked applications to which the pattern returns where needed. Distinguish these additions from the subject Method's essential rules.
A new pedagogical Method is content in its own right, not merely a different publication form of the subject pattern. E.11.PFP:4.4 selects a support publication unit or a separately identified product when that boundary matters. Different audiences need not receive independent authoritative copies of the subject Method.
To choose the division, follow one intended use and one consequential variation. Identify the missing competence or explanation, then decide whether the repair supplies an unexpressed subject-method connection, a learning activity, or a better presentation of an existing explanation.
Place the missing subject-method connection with its Method owner or the public source that supplies it. Give audience-specific learning activities and their pedagogical grounds to the companion. Repair presentation where the content already exists.
Keep an action-changing condition visible at the point where a user needs it. The route to a fuller explanation must be usable within the declared support arrangement. A companion that has not been written or cannot be obtained leaves that audience's promised learning unsupported.
When the division is uncertain, ask a suitably prepared reader who lacks the author's private context to construct a small piece of instruction for a named audience from only the published method account and its declared suppliers.
The result should identify what the learner must be able to do, connect a worked case and practice task to those operations, and state how a consequential mistake or changed condition would be handled. Trace a consequential instruction back to its method grounds, an explicit pedagogical choice, or an acknowledged gap.
Inspect any necessary invention. Choosing a motivating example or a teaching sequence is normal pedagogical design. Inventing the method's missing selection criterion, inference, transformation or coordination shows that the method account or its supplying explanation is insufficient for that use. Repair that omission at its owner; do not conceal it in a privately generated course.
Conversely, failure to teach an unprepared learner from a compact competent-reader pattern alone does not establish such an omission when an adequate companion is explicitly part of the selected arrangement.
This construction probe establishes only what that reader could recover and design. It does not establish that the proposed teaching works, that a generated guide is correct, or that learners can perform independently. Test those claims with their actual audience, task and assistance conditions. E.4.PFAD:4.1.1 compares the cost and value of the resulting pattern, companion and short-use routes, including their change dependencies.
For human and AI readers. Different readers may need different amounts of explanation for the same operation; the same reader may need more when adapting it.
For an AI use, distinguish all available publication content from the material actually supplied or reachable during the invocation, including the tools needed to obtain it. A successful reconstruction qualifies that use under those conditions. It does not show that another model, another context budget or an unassisted human can do the same. A human-learning result does not by itself establish an AI context-management effect.
Authoring examples. A portfolio-comparison pattern can keep its general Method concise by naming the common application that supplies the alternatives, complete small calculation and changed-condition cases. The shared application must contain that work.
A reminder to reuse a previous evaluation instead needs the same-question and still-applicable-evidence conditions beside the instruction; leaving them only in a distant rationale permits an unsupported reuse. A reader altering a construction may need its invariant and why the composition preserves it, even when an experienced operator can execute the unchanged construction from a brief reminder. These are different reading needs, not reasons to publish a separate authoritative Method for each audience.
Use existing sections, examples and direct references to organize these paths. E.11 governs discovery; it does not replace the required explanation with an entry card. E.14 keeps practical use and assurance connected; it does not defer a condition needed before action. E.17.EFP governs faithfulness when selecting or expressing an account: a faithful partial expression can serve a bounded use, while changed claims or a narrower use need their own explicit treatment. No fixed number of reading levels, parallel versions, characters or model-specific editions is required.
Construct the expression that supports the reader's operation
Use this move when the needed explanation is available but the author still has to decide what the reader should read, compare, remember, manipulate or carry forward. It also applies when arranging incomplete material is how the reader will discover the question.
A representation can help produce a distinction that was not available before it was used. Its usefulness is not limited to storing a finished answer.
Begin with the reader's operation and actual material. Try it with a sufficient existing expression before designing another one. Retain the reader's preparation and available help from §4.2.1.1. If the operation itself is missing, develop or obtain it; rearranging its name cannot supply it.
A.6.3.RT.OE explains how to make an expression operative. Developing formation, interpretation or transformation rules is a separate notation-engineering task. The external Notational Engineering DPF (NOT) offers optional methods for that work.
Construct the contrast before choosing the format. Place together the pieces that the operation needs to distinguish or combine. They can still contain uncertain interpretations or an unanswered question. Attempt the comparison, grouping, substitution or inference.
Where it fails, determine whether a missing fact, a missing operation or the arrangement of the expression prevents the result. Change the arrangement when it hides or confounds an available distinction. Obtain a missing fact when the receiving operation requires it. Preserve an unresolved question when it is the useful result of this attempt.
The unit follows the operation. Two cases may need to remain adjacent to expose a difference; one thought may need to be independently movable without losing the reason it states; an instruction may need its prerequisite at the point of action. Start with these relations rather than a blank form inherited from another task. Stable labels and a small common structure can make repeated comparison possible. Retain those constraints while they help, and revise the one that suppresses a needed distinction.
Choose among expressions by what the reader does with them. Continuous explanation can keep a causal argument or an unfamiliar inference together. A mantra can help retain an already developed connection during work; it cannot teach a missing operation merely by naming it.
A table makes the same questions about several cases jointly inspectable. Its rows and columns need meaningful common interpretations, including the conditions under which values may be compared. A card makes a coherent contribution separately addressable and rearrangeable; explain what a connection between cards means.
A diagram can expose a relation whose incidence, order or spatial organization matters; the appearance of an arrow does not establish a dependency or cause. An executable expression can perform a specified operation, provided its inputs, interpretation, execution context and failure behavior are supplied.
These expressions may be combined. For example, a short explanation can establish a table's interpretation, the table can support comparison, and a reminder can later cue that same operation. Select only the contributions needed by the intended use.
Expose the interpretation where it is used. A column headed Date can leave its event, role and unit unresolved. A row can represent an observation, an attempted operation, a relation or a claim rather than one physical object.
State the distinctions that affect the operation, including what an empty position means. A missing observation is not zero; a not-applicable question differs from an unanswered one. Use the domain's kinds, quantities and roles and the relevant FPF distinctions where they settle such a difference.
Distinguish content that must be available from content the practitioner must enter again. A visible value may be derived from a qualified source or retrieved from another record. Show that return when it matters, and require manual entry only for the contribution the person supplies. An illustrative table in a pattern is not automatically a record schema for every performance.
Place, retain and change the support. Place an interpretation or prerequisite beside the first operation that depends on it. Place a reusable example where its intermediate result can be inspected. Keep assurance conditions available before consequential reliance even if their derivation is elsewhere.
If a form will travel without the surrounding prose, carry the interpretation and source return needed in that separated use. A direct link can be sufficient when the recipient has the access and can recognize when to follow it.
Compare the candidate with the best simpler expression by performing the same use at the same preparation. Retain a sufficient existing form when no worthwhile gain is shown. Remove a field or whole support when its contribution is unnecessary or supplied more effectively elsewhere, checking the use that previously depended on it.
Rework instead of adding a second copy when the existing expression contains the right material in an unhelpful arrangement. Add a new support when an operation remains unavailable or requires avoidable reconstruction. Revisit this choice when the question, source, recipient or separation from context changes.
An expert construction establishes what that construction makes available; evidence of reader success or lower effort requires an actual reading under stated conditions.
Worked choice: a date conversion and the form that teaches it. The explanation in §4.2.1.1 requires a source-to-order selection before conversion. Supplier North guarantees month/day/year; South guarantees day/month/year for the files at issue. The intended reader can run a converter given a date order. The author's task here is to choose an expression that makes the selection, its reason and its limit usable.
Begin with the tempting two-column arrangement:
The cell exposes an unresolved interpretation; selecting one output by appearance would invent information. The author asks what differs between the two legitimate conversions. The supplier guarantees distinguish the two interpretations. Rework the comparison:
The unchanged raw characters and different results remain adjacent. The table expresses the source-to-interpretation relation which the two-column prototype lost. The surrounding explanation shows how the applicable guarantee determines the date order and why guessing from the characters is invalid.
Source designates the supplier for this row; the interpretation is an input to conversion derived from that supplier's guarantee, not a newly observed property of the date. Neither the interpretation nor the output must be manually recopied by every user.
A separate card for each row would make this small comparison harder to inspect together without supplying a needed rearrangement. A dependency diagram can explain the same selection when its connections become the difficult part, but adds no needed distinction to these two rows. The table and its short explanation therefore serve this example.
For experienced users, “use the applicable source order; do not infer it from an ambiguous date” is a recall cue with a return to the explanation. It does not replace that explanation for someone learning the selection.
Now change the receiving use. A work session contains only North files under one applicable guarantee, and the converter receives that fixed order from the session's qualified configuration. A repeated source and order entry in every manually completed row supplies nothing further to this session's conversion. Remove those repeated inputs from the local working view; retain the scope statement and its return to the guarantee. The two-column raw/result view is sufficient there.
If a row is later exported alone or South material enters the session, that contextual supply no longer suffices: recover the row's source and applicable order before conversion. This is a conditional removal, not a claim that source identity never matters.
In a different change, South starts using mixed orders and supplies no row-level interpretation. Preserve North's results and the earlier qualified South results. Mark the ambiguous new South row as unresolved; neither a fuller form nor running the converter twice chooses its meaning.
Obtain trustworthy row-order information or correctly matched converted dates before relying on that result. The unknown blocks that conversion, not every other row.
Apply the construction to the pattern's own arrangement. Keep the selection operation and its use-changing conditions in Solution. Keep the small comparison with the instruction that interprets it; a distant collection of unexplained forms would make the reader reconstruct the missing connection.
The canonical sections and stable addresses organize publication and return. They need not become the author's order of discovery: rough notes can reveal the source-to-order problem before the first pattern body is composed. Once a pattern body is authored, §4.1 governs its canonical frame; the external notes are not a substitute for its unfinished content.
The raw/result prototype, the source-qualified comparison and the reduced single-source view expose why different forms support the same conversion in different conditions. They are worked constructions, not observations of improved learning or production performance.
Choose examples by the explanation they make usable
Recover the question from the commission before assessing the answer the author has in hand. If the reader needs to explain why a practice repeatedly produces an outcome, locating one occurrence can supply evidence. The explanation must relate the practice to the conditions under which that outcome would recur or change. Apply that distinction to the question entrusted to the author, the question the text actually teaches and the work a separately issued assignment leaves to its recipient. Determine the content owed from the subject, declared scope and receiving work. Include the relations, grounds, alternatives and conditions needed for that work; do not let the circumstances of a convenient example decide which of them to explain. This is a priority in deciding adequacy, not a compulsory order of discovery or presentation.
Choose the check that fits the reader's promised use. To isolate an inference from a constituent to a whole, compare the two at a fixed level of model abstraction where that comparison is meaningful. Could the statement about the constituent hold while the whole fails to do what the text promises? To isolate the reach of a rule across instances, compare like structural scopes and vary the permitted circumstances. When the question requires models at different levels of abstraction as well as several scales or kinds of holon, identify which distinctions the models of the parts preserve and which dependencies the conclusion about the whole requires. Separate models can allow each part to take either of two states while leaving open whether the parts may take the same state together. A conclusion about their joint behaviour needs that missing relation, or grounds that the answer holds for every permitted combination. C.39.SA:4 helps select the scopes, C.39.WA:4.2 deals with relations across composition scales, and C.37.MM:4.1.1 connects the models for that decision.
A case, sketch or attempted calculation can reveal a missing relation before the general account is finished. Carry that discovery back into the account and distinguish a reusable condition from an incidental feature of the case. Use C.39.GA to check and repair a missing general argument before relying on the example to teach it.
When the promised capability is to choose and connect ranges, recover the choice still open to the learner after they received the commission and any help available before they decided. The heading, diagram or table may already identify the holon kinds, the whole and constituents, the level of model abstraction or the link that settles the decision. Credit the reader for the choices their own work supports. Give them the commission and usable information with the promised choices open when testing independent construction. Explain positively how to make those choices and use their connections under C.37.MM:4.1.1. Vary the particular circumstances while keeping the author’s argument available, including a general derivation within a single illustration. Can the reader recover how to choose the ranges, connect them for the same decision and establish where that method works? Evidence of performance on that assignment can then be judged at the reach it actually covers.
When the reader is learning to model a domain, say what the reader must do with the domain's own types. For a transdisciplinary principle, explain which distinction applies to those types, how to recognize the relation and what inference it warrants. For a domain principle, supply the field's own rule and its conditions. If the intended result is a way of working, explain how to select or construct the next operation, what it produces, and which dependent choices to revisit when its grounds change. A single publication may teach all of these. On unfamiliar material, ask the reader to select, relate or revise the local types and use them to reach a result in the recipient's terms. A worked case makes that reasoning inspectable. C.39.WA:4.3 distinguishes the relations used in such inferences.
When an author uses a case to test an independent choice, identify the choice the prompt leaves open, the sources and help available at that point, and what the learner actually supplied. A problem statement in a brief can support a test of evaluating that formulation. To test formulating the problem, use work in which the learner still had to frame it. An earlier working record can supply that evidence. Apply the same test when the task is to construct a disciplinary model. Inspect which unit, identity criterion, correspondence or decision rule was already given. A worked answer can show how the learner applied it. For independent construction, establish which model choices the learner made and whether the resulting model serves the assigned use. When the learner must also justify those choices, examine the grounds they supplied. If a completed form could be copied, check the result itself and keep attribution open until there is evidence of the learner's work. E.23.CDI:4.2.1 guides the separate decision about how a holder can acquire and demonstrate a capability under its working conditions.
Keep the explained subject and its example distinguishable. When teaching a Method, identify the reusable way, the actual or stipulated performance that illustrates it, and the material on which that performance works. Show which choice or action in the example enacts which operation and why its result is usable next.
An explanation designer wants an analyst to see why equal weight per person and equal weight per group can give different overall means. The designer chooses two nonempty groups that partition the people and have different counts and different means. The conditions make the results differ. In this example the calculation supplies the mathematical result. The choice of conditions, and what the analyst can now distinguish, explain the designer’s Method.
Make the common claim, its scope and its grounds available in the explanation. It may be expressed alongside cases, through a derivation or through a named accessible supplier. State what may vary and which conditions make the relation or operation applicable.
If the text calls the example a stricter model, compare the two models for the same inquiry under B.5.MR.
When cases suggest a new general claim, develop and qualify that claim through the relevant subject Method. A constructed instance can demonstrate an operation, a counterexample can defeat a universal claim, and observed cases can contribute empirical evidence; the wider inference retains the grounds appropriate to that use.
Select an example for the contribution it makes: making an unfamiliar relation inspectable, exposing an intermediate operation, distinguishing plausible rules, showing a boundary, or connecting several Methods in use. Work the correspondence between the account and the instance.
A correct answer can fit more than one rule; choose values or a comparison that lets the reader distinguish the intended relation. Preserve adequate direct explanation when an additional example would contribute nothing to the declared use.
Archetypal Grounding remains a content function of a canonical pattern; it is not a requirement to insert an example into every paragraph, slide or section derived from that pattern.
Choose meaningful variety from the promised subject. Recover the differences in objects, relations, conditions, available grounds and receiving work that the account must handle. Select cases that make those differences intelligible, including ordinary applications as well as contrasts and limits.
A numerical derivation, a causal explanation with uncertain evidence and a recommendation under competing purposes can require different explanatory moves. Changing numbers within a derivation does not supply the other two.
Several materially different examples may be necessary to ground a general account even when each example is locally simple. Explain what remains common across them and what changes for a subject-specific reason.
For human learning, HCD.5:4.2.1 and HCD.6.1:4.2.2 connect explanations with a distribution of qualified encounters and focused contrasts. Use that contribution when a learner must recognize or use a family that one pair cannot expose. A demonstrated distinction can be available before a complete verbal rule; preserve its qualified examples and conditions instead of inventing a formula. The framework author's obligation is to make the claimed commonality and its basis usable, including the limits of what has been explained.
Decide whether to continue the same case. Continuity can make an earlier result, a changing state or the consequences of a choice available to the next operation. It can also avoid reintroducing circumstances the reader already understands. Retain that benefit when the case still shows the relations now needed.
At each consequential transition, return to the subject being explained: which operation or principle is this instance making usable? If a new question needs different objects, a competing method or a condition the case cannot represent faithfully, change the case, juxtapose a contrasting one or use a direct explanation.
Keep the general reasoning connected while examples change. Retain the varied cases needed for that account; shorten incidental story development when it displaces the next required connection.
Check the account at its promised scope. Compare what the explanation lets the intended reader recover with what the subject and receiving work require. The reader should be able to explain the governing relation, its grounds and conditions, and use it across the material variations included in the promise. Choose those variations from the subject, not only from what is easy to change inside the chosen setting.
Inspect the correspondence back to the Method or principle being taught as well as the correctness of the example's domain result. Supply a missing general operation, explanation or accessible supplier; add or revise cases when the required correspondence or variation remains hard to recover. Use F.19's evidence and coverage conditions for a reader-recovery conclusion.
For example, a continuing transaction can show why a later update must use the state produced by an earlier one: retaining the same transaction makes that dependency visible. In an account of resource use across Methods, show the different relations of a consumable specimen and reusable information. Explain consumption and reuse, then show how each changes the receiving operation.
The author's reusable move is to select and connect cases by the relations owed to the reader; the transaction, specimen and information are the material used to perform that move. A reader who already understands a short algebraic derivation may need no narrative case for that inference. These are choices about the contribution required, not a fixed allocation of examples.
Package-form and neighboring-pattern reference discipline
FPF package-form words and neighbouring-pattern references carry stable meanings. State the relation the sentence asserts, and use the governing subject pattern when that relation is not recoverable from ordinary wording.
For an ordinary neighbouring-pattern reference, state the concrete contribution and cite the PatternID. Distinguish a source for what this passage says from material the reader is directed to apply, using E.8:4.2.2.2. An identifier or locator only helps the reader find that content; it does not declare this purpose. Identify an exact claim-bearing episteme, ClaimGraph, edition, or relation assertion only when a named later use depends on that identity.
A local ...PatternLocator field may remain where a schema already uses it as a non-semantic convenience, but ordinary prose and entry cues do not require one. It never substitutes for the cited content's concrete contribution or, when the stronger identity branch is active, for the exact claim-bearing content.
Changing only a locator without changing what it resolves is a representation change; changing the defining content or exact assertion may reopen the semantic object whose receiving use depends on it.
Keep the following package and relation words distinct:
- pattern reference = a reference to pattern content whose contribution and purpose in the referring passage are stated; citing that content as a source and directing its application are different uses;
- specialization = an exact relation in which the child carries the required parent content plus an explicit child delta and use boundary;
- overlay = a cross-cutting reading or review projection over stated source content; it adds no authority or obtaining relation by name;
- profile = a declarative bounded-use or review projection from stated source content, not a replacement pattern or actor;
- family = a recurring class of cases under an explicit membership rule, not a hidden common owner;
- bundle = a packaged set of defaults, allowances, or coordinated members whose actual relations remain explicit;
- cluster = a navigation or reading-order grouping, with no semantic relation by grouping alone;
- suite = a coordinated set whose suite-level membership and coordination semantics are stated;
- pack = an editorial, source, review, or delivery grouping, not semantic authority;
- kit = a reusable coordinated publication or boundary-description package with exact kit-level membership and use;
- record = a case, report, assertion, representation, or review record under its own identity;
- umbrella = a provisional review head spanning possible subfamilies before an exact membership rule and the relevant claims and relations are settled.
These words are not interchangeable and do not stand in for a missing relation. Say specialization of … with delta …, profile projecting … for use …, overlay reading …, bundle containing … under membership rule …, or another exact formulation.
A source-defined position name may be reused when the cited content defines that position and the current assertion uses it in that sense; otherwise recover the meaning through E.10.ROLE and do not improvise near-synonyms for stylistic variety. The preceding receiving-use discriminator decides whether exact claim-bearing content must also be identified.
Precision-restoration placement discipline
When a pattern or companion text is drafted from E.10 or E.10.ARCH, distinguish three authoring objects:
semanticAreais the Part-F semantic unit for a wording-use restoration row: one Concept-Set row, one UTS row, or an explicitly bounded row-set. It is declared withsemanticAreaBaseConceptandsemanticAreaSenseFamily.ontologicalNeighborhoodis the applicability neighborhood around that namedsemanticArea: nearby primaryEntityOfConcernkinds, relation kinds, claim records, content that defines or constrains the current use, non-use boundaries, and remaining reader use that can carry the recovered meaning after the wording is repaired.pattern nestis the publication and specialization placement of a pattern under a declared family or membership relation.
These are not synonyms. A precision-restoration pattern is placed in the pattern nest whose primary EntityOfConcern, relation record, or claim record it repairs. Its semanticArea states the Part-F semantic unit it repairs, while its ontologicalNeighborhood may name several direct relations and pattern content that defines or constrains the asserted uses.
For example, quality-term repair lives in the C.16 characterization nest, even though its neighbouring relations can include relation construction, action invitation, evidence, assurance, source-use assignment, engineering quality bundles, pattern-quality evaluation, or mathematical-lens use.
Affected patterns should use a thin pointer when the first-stage wording repair belongs elsewhere. The pointer names the selected restoration pattern and the condition that triggers it; it does not copy the trigger registry, the full E.10.ARCH recovery algorithm, or a second local architecture for the same repair. The affected pattern then keeps its own subject matter: the characteristic, structure, view, episteme, relation, evidence, assurance, gate, work, decision, or adequacy question it already governs.
If a draft proposes a new precision-restoration pattern, the authoring claim must show the repeated wording failure, semanticAreaBaseConcept, semanticArea, semanticAreaSenseFamily, the recovered primary EntityOfConcern kind or relation/claim record, the intended pattern nest, the neighboring governing relations, and the admissible action left after repair.
A new pattern is not justified merely because a word appears often, because a local checklist wants a bucket, or because a campaign needs a tidy grouping.
Make the purpose of a reference clear
A reader should be able to tell whether a reference gives the source of an explanation already supplied or points to material they are being asked to use. Express that difference in ordinary prose:
- A citation identifies the basis of a claim, definition, quotation, comparison or explanation. Make clear what is taken from or justified by the source. When the quoted or explained material is sufficient for the present work, its citation adds no obligation to open the source before using it.
- An application reference asks or invites the reader to use the named material for a contribution to the current work. Say what to obtain or do there and how it bears on the present question. Keep any condition and the force of the advice clear: a required dependency, a conditional next move and one possible way are different. This is an explanation, not a set of fields to add to every sentence.
Prefer wording that already makes the purpose clear. “This distinction follows X” cites a basis; “use X to compare these arrangements” directs an application. A conventional citation attached to the claim, or an explicitly identified source column, can establish the first purpose without another label. Do not prefix an already clear instruction with another label.
Short Source: and Apply: cues are available when they help distinguish otherwise ambiguous entries in a compact list or table; they are not mandatory markup. Keep the address or ordinary Markdown hyperlink unchanged. No new link protocol, suffix, HTML attribute or address separator is needed.
The purpose belongs to this use of a reference, not to its destination. A pattern can be a source in one passage and material to apply in another; an external paper can also describe a Method to use.
When both purposes matter, explain both contributions rather than leaving the reader to infer an instruction from a citation. Do not hide an operation or use-changing condition that the reader still needs behind a source citation. If it must be obtained elsewhere, say so and explain the return to the present work.
A sentence, list introduction or table heading can establish the purpose for a clear group. A generic Relations, Solution, Rationale or See also heading does not decide it. Separate mixed uses where their difference changes what the reader should do.
Pure navigation, declared identifiers, address examples and addresses that merely name participants in an explicit relation need no source or application cue. In “X specializes Y”, Y names a participant; it is not automatically a citation or an instruction. But “X supplies comparison” at a step needing a comparison can leave the needed action unstated. Decide whether the comparison is already supplied or the reader must obtain it.
For example, these references to the same material do different work:
- This account distinguishes Work, capability, means and authority following C.32.MWA.PP:4.3. Authority to allocate a task does not by itself supply the capability to perform it.
- When those contributions must be allocated differently, use C.32.MWA.PP:4.3–4.8 to construct and compare complete arrangements. Bring the feasible alternatives and their combined demands back to the present choice.
Neither form settles truth, authority or sufficiency. Inspect a cited source when its meaning, assumptions or evidence need examination for the present reliance; do not make that an automatic extra reading. For an application, an adequate existing result or already understood Method can suffice without rereading or repeating the work. An application reference cannot make unavailable material available, excuse an unexplained connection or establish that the promised result has been obtained.
Read a proposed destination far enough to establish that it supplies the contribution the referring text promises. If it does not, change the direction or develop what is missing rather than merely changing its wording.
Judge the connected passage as prose for its intended reader: can they tell what is available here, what they are being invited to obtain elsewhere, and what happens next, without decoding labels? When an extract loses a heading or introduction that supplied this meaning, retain the necessary purpose and condition in the extract. Use E.11.PUA:4.1.1 when deciding how to follow such a reference.
Intended-reader discipline for pattern prose
The intended user is the agent that uses the pattern. Write for the work it supports, such as organising thought, investigating a question, changing a system, publishing a description or reviewing a result. One agent may apply the pattern while another performs the work it describes.
Check which system the claim concerns. Try the operation with a different performer. What capabilities, access or authority does it require? Use the action or role in the general instruction. A person, a technical system or a division can work as an engineer. Then hold the work result fixed and vary the system to which a decision or a capability is attributed. If the attribution changes what the evidence establishes, name that system and the conditions that matter.
A robot can measure a part against a tolerance. If the contract assigns acceptance to the customer’s inspection unit, that unit decides whether to accept it. The same accepted part could follow from a judgement supplied by a vendor or from the unit’s own interpretation. To judge the trainee’s ability, identify which interpretations the trainee made with the aids available in the intended work. A suitable earlier performance can suffice. Seek a further performance when the available evidence leaves that judgement open.
The pattern’s sections explain the described subject, the user's action and result, costs, and grounded boundaries that change use. Public architectural explanation belongs here when it helps the reader understand, select, combine, or adapt the Methods.
For example, explain why a profile specializes one step but reuses another, how a combination makes its next result possible, or when an alternative would be better. State the relations that apply in Solution and explain their reasons in Architectural Rationale. A whole language and its profiles use the same content functions through E.11.PFP:4.7.
E.8 reader and reviewer wording is framework-authoring wording. Project-side publication readers, explanation readers, comparative review units, and participants in named project-side review relations are governed by the patterns that name those units and relations, such as E.17, E.17.ID.CR, E.17.EFP, A.10, A.15.4, A.20, or A.21.
Keep the development history of the current pattern version in its DRR, companion, review, or release material. This includes current arguments for promoting a draft, authority-reference or naming freezes, merge and landing state, and review correspondence.
When a development decision contains a durable reason that changes practitioner use, publish that reason with the necessary explanation and source return; retain the dated decision and its evidence in the development record. Users need enough reasoning to understand the choice without reconstructing that record.
Human-facing fit beyond intended-reader correctness
Human-facing fit is also subject-domain fit. A recognition text that starts from internal taxonomy, pattern-placement convenience, or package-architecture wording before the problem-domain moment is still under-authored even if its later guidance or check text is correct.
When a broader umbrella name and a narrower operative branch are both used, the recognition text should also tell the reader how the branch fits under the umbrella, what it governs and which action and predicate apply, as described in §4.2.1 rather than leaving that reconstruction to a later declaration block or companion note.
P-2 epistemic precision check. When the E.10 criteria call for epistemic precision restoration in pattern prose, the first admissible action-guiding move must survive as remaining admissible reader use or be replaced by a neighboring FPF rule whose content now defines or constrains that claim application. E.2 P-2 and E.12 require this.
Intentional didactic metaphors and vivid Plain recognition lines are admissible when they are ordinary recognition aids or when their claim kind or admissible-use boundary maps back to Tech under E.10:6.2. A precision-corrected rewrite that leaves the recognition text inert is still under-authored.
For method guidance, explain in Solution what the performer should attend to or compare when it can change the next move: conditions needed for the operation, consequences expected from it, and whether the obtained result serves the receiving work. Give a reachable continuation when a condition is unmet, unknown or the result leaves the original concern unanswered.
B.5.PI develops an opening into inquiry; B.5.EA helps when the concern cannot yet be expressed, and an early cue can remain unresolved. Put these connections where they change use, including in first authoring.
Do not add a compulsory monitor, extra form or copied “watch for anomalies” paragraph to every pattern. A known adequate result can still end the work directly.
Use the recognition and assurance functions in §4.2.1. Put the working situation and first useful result in Problem frame and Solution; add candidate-pattern comparison cues when a question about finding the right pattern remains under E.11.
Entry-cue authoring rule. Begin with one ordinary question about the user's object and claim, before any PatternID, card, template label, or internal taxonomy. In the same compact cue, state what cited content contributes, cite the pattern id, and name the smallest result usable now plus its stop or return condition.
Add a tempting overread only when the F.19 plausible-reader test finds independent local ground and an action-changing effect. Name an exact episteme, ClaimGraph, or edition only when a later use depends on that identity. The cue guides reading; it does not by itself constitute a result or relation.
Compact candidate-pattern comparison belongs in E.11-distributed entry material. Longer first-entry comparisons belong with E.11's bounded-comparison explanation; E.11:5.7–5.14 supplies worked examples.
Keep a claimed breadth visible at first reading and preserve the same claim as the assurance text develops it, under §4.2.1.
Illustrative migration example (informative).
Old pre-template top:
Repaired Problem-frame recognition signature:
Design-time and run-time referents stay separated in pattern prose
Pattern prose keeps the referents needed by its declared subject and use recoverable. Authoring, review and publication are valid governed subjects; state their actions and rules directly. Distinguish design-time, run-time and normative-publication referents when that distinction changes the claim.
Apply §4.2.3 to place this edition's drafting, review and landing history in its development carrier. Retain any durable reason that changes the user's action, choice or boundary in the public explanation.
If a pattern or example claims autonomy, name the admitted U.System whose freedom of action is being evaluated and use the current E.16 pattern that defines or tests the claim. Add another relation only when it is current under its own governing pattern.
Admit dated Work under the A.13-first and independent A.15.1 rule in E.8:0.3, adding F.6 only for current assignment-bound attribution. Add autonomy apparatus or a vignette only when it helps the reader use that claim. Apply F.19 after recovery; if the corpus supplies no direct governor, return A.6.RCD missing-governor.
Archetypal Grounding
Turn a framework outline into a usable common account
An author has three short chapters about selecting a source, reusing a rule and refreshing a decision. The commissioned introduction must explain how readers can keep source-based decisions justified as their sources change.
The draft says: "Find a source. Record its title and date. Apply the rule. Review the decision whenever a new edition appears." Its sequence names the activities but leaves their connection unexplained.
The author follows the required decision back through the material. A source is useful here because a particular claim, under stated conditions, contributes to the decision. Its title and date enable recovery of that claim. A later edition matters through a change to that contribution. This dependence gives the common account its organizing ground. The author writes:
Begin with the question the decision must answer and recover the source claim that can contribute to it. Establish whether the claim's conditions hold for this use, then make its effect on the decision explicit. Keep enough source identity to recover that basis. When the source changes, compare the used claim, its conditions and limitations; reconsider the decisions whose grounds are affected. An already adequate source or unchanged contribution can continue to serve its existing use.
That paragraph changes the organization of the three chapters. Selection now returns a qualified contribution to a question. Reuse explains how that contribution enters the decision. Refresh follows the resulting dependence, returning to selection when a usable source is lost or to application when its conditions change.
Each chapter can still serve an independently arising question. Their common account explains how their results connect.
The author distributes the remaining content by function. The Problem explains why a title and date can coexist with an unsupported decision. The Solution supplies the operations and their returns.
The Rationale compares dependence-based refresh with redoing every source-based decision after any edition change, including the cost and the risk of overlooking an affected use. Grounding shows how the decision depends on a source claim; the Conformance Checklist asks whether the used claim, applicability and affected decision can be recovered. These functions guide what is written, rather than providing places for repeated assurances that the account is complete.
In the constructed case, a handbook's rule sets a maximum reporting interval of seven days, and a team has chosen weekly reporting on that basis.
One new edition changes the chapter title while preserving the rule and its conditions; the common account leads the author to retain the decision and update its recovery route. Another edition shortens the applicable maximum to five days; the author shows the reviewer returning the weekly decision for revision. The difference supplies the reason for selective refresh and explains which earlier contributions remain useful.
Suppose the commission expands to recommendations whose suitability also depends on the recipient's preferences. The current source account leaves that additional ground implicit.
The author adds the recipient's objective and preference to the common decision basis and connects their change to the relevant reconsideration. Merely adding a fourth chapter called "preferences" would leave the return unexplained. The existing source-recovery operations remain usable, while the expanded account makes another condition of reliance explicit.
Keep expression, organization and evidence in their roles
The date-conversion repair in section 4.2.1.2 shows a different authoring choice: how to preserve an operation, its conditions and a return to fuller explanation when changing the reading arrangement. Use that case for the form-and-access question and section 5.1 for construction of a common account and its constituent explanations. Their different contributions follow the authoring work the reader needs.
Small subject sentences can illustrate density without jargon: "The pump casing seals at this face" names the relevant action and location; "The claim states its predicates explicitly" names an epistemic subject and its property. Their role is to illustrate expression. The organizing explanation and performed authoring choices in section 5.1 carry the larger construction.
An individual pattern uses the canonical sections and required :End sentinel. A whole-framework account uses the publication arrangement in E.11.PFP:4.7 while answering the relevant content questions. For a claim about F (Formality), use C.2.3 and state the expression-rigor criterion it satisfies.
Bias-Annotation
Lenses: Gov, Arch, Onto/Epist, Prag, Did. Scope: Universal for the authoring conventions in this pattern. This guidance biases toward Did (readability, narrative flow) and Arch (template regularity) by design. Mitigate this bias by giving each reduced section the smallest content-bearing statement that justifies the reduction: a grounding, misuse, boundary, or reduced case. An omission placeholder cannot do that.
Conformance Checklist
CC style (canonical).
Conformance Checklist items are authoring checks: they test whether the pattern guidance has been applied and written correctly in a pattern or companion text that claims conformance. They do not replace Solution, do not make the pattern a control form, and do not state deontic obligations about the modeled world.
A CC clause of the form “X SHALL …” is to be read as “In a conforming pattern or companion text, X SHALL …”.
Passing a Conformance Checklist supports the selected conformance claims, not a claim that all anomalies were excluded or that the result answers every receiving use. A.15.11 explains how relevant checks can become an execution aid with an actual occasion, inspectable conditions and a continuation for unresolved items. Preserve the difference between that aid, the method's Solution and inquiry into a concern that the list does not cover.
Preferred wording for new or edited CC items: start with an explicit conformance subject (e.g., “Authors …”, “Reviewers …”, “A conforming implementation …”, “A validator …”). If a CC item is enforcing an admissibility predicate, it SHOULD cite the predicate’s identifier (from a Definition: / Invariant: / Well-formedness constraint: block) rather than restating the predicate as “X MUST …”.
For boundary/interface/protocol/declaration patterns, prefer A.6.B-scoped claim IDs (L/A/D/E) or cite an existing Claim Register (A.6.B:7) instead of restating mixed prose.
Common Anti-Patterns and How to Avoid Them
Consequences
Architectural Rationale
The common structure and style principles serve:
- P‑1 Cognitive Elegance – uniform, minimal surprises.
- P‑2 Didactic Primacy — narrative flow and grounding in the kinds and relations the reader’s work requires.
- Guard-Rails 1 & 2 — abstract definitions remain portable; the heading grammar keeps the exact syntax needed for its own conformance conditions, and expressions retain E.5.2 interpretation and mapping.
A unified template also improves retrieval: a chunk containing A.2:<n> - Bias‑Annotation remains self‑identifying even when parent headings are missing, and the required footer marker makes truncation detectable.
The ASCII space-hyphen-space separator in H-2 keeps heading entry inexpensive: authors can type it directly on ordinary keyboards, and readers can reuse the same characters in search and plain-text tools. Typographic dash variants require an extra input or conversion step while adding no information to the boundary between an identifier and its title. Where a prose dash is useful, -- is a keyboard-accessible option; the identifier/title separator remains the one shown in <FullId> - <Title>.
The Conformance Checklist groups the criteria for judging a pattern that claims conformance.
SoTA-Echoing (normative; typed comparison to contemporary best-known practice)
Canonical definition and contract. This is the FPF definition of SoTA: the best-known currently defensible answer to one named practice question. F.1 may prepare the question-relative source cut and E.21 may evaluate the resulting pattern, but neither redefines SoTA.
A SoTA-Echoing section earns its place by changing the pattern's Solution, boundary, case, check, relation, evidence requirement, stop, or reopen condition. It is not a bibliography, source-currentness register, or lineage shelf.
An original answer can be the selected line. A bestKnownLine may be an original semantic or methodological synthesis developed in FPF or a DPF. It need not occur whole in an external publication. State the unresolved problem, the limitation of the serious alternatives, the new distinction or constructive move, and the useful consequence it makes possible.
Support that consequence at the scope claimed, with its cost, limits and reopening condition. Originality permits consideration; comparison supports selection.
For example, an authored connection may preserve two independent uses when a changed requirement defeats only one. Work that difference through the case and compare the available alternatives.
A conceptual construction or conditional derivation can support that bounded result; empirical effectiveness or practitioner learning requires its own evidence. Neither an external publication with the same answer nor a publication date later than all rivals is required. One such gain establishes no ranking of the entire framework.
Source roles in plain wording. Classify each retained source by what it can do for the question:
- a best-known-line candidate supplies or critically synthesizes the strongest current answer being considered;
- a serious current rival supplies another answer that could change the selection;
- failure or counterexample evidence shows where an answer breaks or does not transfer;
- an official or popular comparator exposes a default worth comparing but gains no rank from authority or adoption;
- lineage only explains history without supporting the current selection; and
- identity/currentness only identifies a source, edition, date, or maintenance state without supporting its truth, adequacy, or rank.
Only the best-known line, serious rivals, failure evidence, and a necessary explicit comparator belong in SoTA-Echoing. These are comparison roles, not publisher or institution classes. An official standard, widely used practice, or university-endorsed line can be the best-known-line candidate when its substantive answer wins the comparison, but authority, freshness, prevalence, or praise contributes nothing to that win.
Lineage-only and identity/currentness-only material stays in source records, notes, or evidence carriers outside the pattern body. An official or popular default stays as comparator only when its precise defect is needed to explain the selected answer and changes a governed pattern locus.
Positive comparison contract. Every positive SoTA use states, in readable prose or one compact table:
practiceQuestion— the exact working question;bestKnownLine— the selected answer, not merely its newest source;seriousAlternativeOrDefault— the rival or default that could have changed the answer;defectOvercome— the action-changing defect, limit, or trade-off that selection repairs;patternMutation— the exact Solution, boundary, case, check, relation, evidence, stop, or reopen locus changed;sourceRolesAndLimits— the exact source edition or stable locator, why it has this comparison role, and what it does not establish; source identity supports replay, not rank; andreopenCondition— the smallest new evidence, rival, failure, or use change that would require comparison again.
For each material source use, name the claim, distinction or method being considered and mark its disposition adopt, adapt, or reject. Distinguish those uses from any newly constructed move. The origin of the selected answer and the disposition of its sources are different questions: an original answer can retain a source's premise, adapt its method and reject its proposed generalization.
Explain which defect of the incumbent, popular, or official answer is repaired and why the selected line is no worse at comparable application effort on the values that matter and better on at least one, or state the trade-off deliberately accepted. More sources, a later date, a wider deployment, institutional praise, or a longer review cannot replace that comparison.
Source gaps and the lightest sufficient evidence. If an adequate best-known comparison cannot be established, say which rival, counterexample, or source role is missing and return that source gap. Do not fill the section with a current standard or recent paper. Use F.1 for the smallest question-relative cut and its SoTA-specific role branch. Use F.0.2 only when the conclusion needs cross-source synthesis. Use a broader G.2 pack only when repeated refresh or a wider claim justifies that cost.
Evidence and relation discipline. Reuse an existing G.2 pack's exact ClaimSheet, corpus-ledger, Bridge rows, and source roles instead of forking a second narrative. Inherit non-conflicting comparison content from an accepted DRR and its source materials while keeping the DRR as the decision and placement record.
For an obtaining semantic Bridge, identify the two exact F.17 local senses, the F.9 relation, and a separate bounded-use claim; otherwise leave that relation unasserted. Keep numeric comparison under its applicable ComparatorSet or CG-Spec without hidden scalarization.
Writing guidance. Lead each row with the practice question and practical choice. Name the selected line and serious alternative, state the defect and pattern change, then give source roles, limits, and reopen condition. Complete sentences are preferred to tag lists. External terminology or tooling stays out unless the comparison itself needs it.
SoTA alignment for this pattern (E.8 self-echo)
Relations
E.8.ML constructs a problem-first language for connected Methods: independently useful bodies, their result dependencies, alternatives and returns, and the explanation of their combined use. Its direct-description exit avoids making a language where one adequate account is sufficient. E.11.PFP continues to govern public arrangement.
-
Coordinates with:
E.9.DAwhen an authored pattern body is drafted from a concreteDRRand the blocker is whether theDRRselected, distributed, carried source use, carried accepted decisions, or supplied a first drafting action sufficiently for that authoring use.E.8still governs the pattern body;E.9.DAis not a mandatory authoring section, review card, or substitute for writing the Solution. -
Constrained by: Guard‑Rails E.5.1–E.5.4 (lexical firewall, notation independence, etc.)
-
Coordinates with:
E.21when one authored FPF pattern version is evaluated as a scoped pattern-quality claim.E.8governs authoring shape, recognition text, action guidance, worked cases, SoTA grounding, and conformance material;E.21governs the pattern-quality evaluation, required coordinate values,PatternQualityStatus, and stop condition. Do not importE.21as a mandatory authoring section or full review card. -
Coordinates with:
E.23when an authored FPF pattern body is being improved through repeated passes.E.8still governs the authored pattern body;E.23governs the repeated quality-improvement method; the object-under-improvement evaluation such asE.21orE.9.DAsupplies value meanings and stop meanings. -
Coordinates with:
E.13when an authored pattern claims practical payoff or uses a visible quality value, metric, checklist result, review result, or release posture as if it were the intended value.E.8keeps the payoff in user-facing prose;E.13repairs proxy-to-value substitution. -
Coordinates with:
E.4.DPFfor choosing a DPF reference code, PatternID plan, continuity across editions, and reader return after split, merge, replacement, or retirement; andE.11.PFPfor current Part, position, public order, and citation display.E.8owns only the common identifier grammar and reference wording; identifier form and checker success decide none of those authoring or publication questions. -
Coordinates with:
E.11.PUR, which supplies the recommended-pattern-use decision for a current concern, andE.10.MOVE, which disambiguates whether move-like wording names pattern-use recommendation, direct work, plan, gate, transformation, publication, source, architecture, call-planning, or language-state material. These references state concrete contributions; an exact assertion, claim-bearing episteme, orClaimGraphis added only when the named receiving use depends on that identity. -
Constrains: All patterns.
E.9defines the DRR decision kernel and decision-inspection content blocks.
E.8:End
Construct a Pattern Language for Connected Methods
Type: Method pattern Status: Stable
Problem frame
Use this when. The Method-related content has been recovered and allocated, but a linear description leaves recurring users without a way to start from their own problem, reach a neighboring decision or return to a source when they need to. You must decide which patterns and relations will make that content usable as a language.
The working reader is a Method engineer or domain-framework author. Start with a user's question, such as “May I reuse this review after the source changed?”, and name the first useful result. Then identify the smallest body or cooperating set that can supply it.
The first useful result is a brief account of how to organize the language. State the problem that brings the user to the language, each needed contribution and what it concerns, the relation between contributions that matters for this use, and where to stop or return.
One direct MethodDescription can be the correct answer. Use A.3.2.DE without a language when it already serves the use. Use E.11.PFP alone when only public arrangement or navigation is missing.
Problem
Source order, a procedure diagram or a repository taxonomy can become the pattern architecture without a separate decision. Every chapter is then treated as a Method part, every pattern as a MethodDescription, and every link as the next step in Work.
The opposite failure is a bag of useful patterns with no explanation of when they contribute together. A user must reconstruct prerequisites, alternatives, conflicts and source returns from private authoring notes. A whole-framework introduction can hide the same problem when it gives only labels and no practical answer.
Forces
Solution
E.8 supplies a pattern's substantive functions; E.4.CM and E.11.PFP govern the connected account and public arrangement. The construction here assigns Method problems and results to independently useful bodies and explains the relations a reader needs to combine them.
The construction can finish with an adequate direct description instead of a new language.
Build the language around the questions it must answer and which results each answer uses. Keep Method semantics, pattern relations and publication arrangement distinct.
Select the smallest useful language
-
Recover the receiving uses. Start with a commissioned use. Who needs to understand, decide, perform or devise what, what would warrant that result, and when is it sufficient? Include another promised use when it needs different grounds or a different contribution, or leaves a choice to someone else. Two uses can lead to the same action.
-
Recover the allocated content. Use F.0.2.MA's source allocations or an already sufficient equivalent result. Preserve current suppliers, unresolved candidates and direct-source remainder. Do not infer a new Method from a source heading.
-
Identify what each body describes or governs. Use A.3.2.DE to describe how an admitted Method is performed or to improve an account of a proposed way while retaining its unresolved identification conditions.
Keep neighboring selection, evidence, permission, architecture, representation and source-return questions under their own subjects. A body can cite those results without absorbing them.
-
Compare the direct alternative. Try a direct description or existing pattern entry for the same use. Retain a new body or language relation only when it adds a warranted action, result, boundary or saved source reconstruction at acceptable burden.
-
Assign the contributions to bodies. Give each body its own recognizable problem, useful move, grounding under E.7 and stop. Explain the shared conditions and the reasons for the selected combination in the profile's account of the whole. For each inherited answer, point readers to where the contributing body gives that answer.
-
Name the material relations. State in ordinary language which result another use consumes, what condition enables that use, what is an alternative, what conflicts and where a failed or missing result returns. Apply the defining FPF relation when a stronger relation claim is needed.
-
Provide direct reader entries. A reader should reach the relevant full body from a problem question without reading the complete production account. The explanation of the whole remains available to readers choosing, combining or adapting the language.
-
Challenge the proposed arrangement. Use F.0.2.RC to attempt reconstruction and a changed-source return. Use C.37.CF only when a material claim about content or representation advantage remains. Keep a direct-description or smaller-language answer when it is sufficient.
The list is an authoring aid. It does not prescribe the order of every user's Work.
Keep the described wholes separate
A MethodDescription has one admitted Method as its subject. A source-grounded language may publish such a description across several bodies when their bounded claims jointly describe that Method.
State that subject and the claims included; do not infer membership from a coherent book or a shared title.
A profile account concerns the selected pattern/Method arrangement for its declared uses. It can include several MethodDescriptions and neighboring results.
E.4.CM:4.8 develops the source-grounded production construction as a specialization of the existing framework-contribution Method, with C.39 governing general Method development outside publication. The account of that production work remains distinct from independent situated-use claims and from general claims about profile relations. Sharing a carrier does not merge those epistemes.
Method parthood is another claim. Independently identify the proposed whole and smaller Methods, then use B.1.5 for their whole-forming relations, joins, bounds and failure conditions.
In the absence of that result, the proposed Method composition remains unestablished. Do not make the pattern inventory stand in for the missing composition.
Make each relation say something useful
Use the relation on which the commissioned action, explanation or construction depends. The table gives examples of common questions.
A formal CGUS is appropriate only when that structure is needed. A.22.CGUS supplies the constituents, selected obtaining relations, applied constraints and named use frame, as well as potential branching and case-judgement requirements.
Plain alternatives and conditions do not automatically admit that structure. C.39.MC supplies ordinary situated continuation and its optional formal branch.
A changed fact can change a continuation judgement under the same frame. A changed question, admissible action or stop can require reidentification.
The user or capable system assesses the new situation. During ongoing Work, A.15.7 supplies the next-action decision.
When the promised use includes performing or learning a Method, show its relevant constituent–whole connections as well as result dependencies. At a selected moment, which constituent action enacts which encompassing work? What changes in that action when a condition of the whole changes? B.1.5.EW supplies this recovery.
Show the capability, support and combined resource conditions that the language relies on another contribution to supply. When a missing intermediate performance can affect continuation, show how the larger work depends on it. A direct argument or a worked case can establish that dependence. Trace it as far as the receiving use requires.
B.1.5.RS supplies the distinct comparison when a constituent is replaced across encompassing uses.
Explain the profile at its own scope
Use all twelve substantive E.8 functions for the connected account: working situation, problem, forces, selected solution, grounding, bias, practical checks, recurring failure, consequences, Architectural Rationale, source comparison and relations.
E.11.PFP governs where readers reach those answers. An inherited answer can be cited precisely; shared whole conditions and profile-specific consequences still need an explanation.
If a language is claimed to expose more structure to its reader, use C.37.CF with C.2.8 to test that claim. Select the dependencies, alternatives or stops the reader needs, and distinguish new content from a different expression of the same claims.
Identify the actual preparation, help and source returns. More patterns or links do not establish more extractable structure, and a larger amount does not establish that the language is worth its reading burden. Keep the direct description or smaller language when it suffices.
Identify the relation being traced. Method composition relates constituent Methods to the encompassing Methods they help constitute. If the inquiry also compares the abstraction of models, establish which distinctions each model preserves for that same inquiry under B.5.MR. A specialization narrows conditions while retaining the relevant general claims. A bounded-use projection selects claims for a reader. Reuse can connect several profiles. A Method-composition result establishes which Methods are parts of the whole. A publication group provides navigation.
There is no fixed maximum number of scales, and overlap does not establish a mathematical lattice; use C.29 only for that qualified representation.
Archetypal Grounding
From a review paragraph to three independent questions
This is a constructed case. An engineer has an admitted Method for checking a Method-description claim against its source and qualification basis. A use-bounded A.3.2.DE description of that Method already explains the checking action.
The new language must also help recurring users find out how different source contributions were allocated and decide what to revisit after a source change.
The proposed arrangement gives users three entries.
These are independently useful entries. Their colocation does not make these pattern subjects parts of one review Method. The A.3.2.DE description retains its one-Method subject; the profile explanation concerns the broader arrangement.
For a user who already has the needed allocation and only one contradictory claim, direct A.3.1.CC use is enough. A user investigating an unexplained coverage loss starts with F.0.2.RC.
The language is useful because the first result and return are visible at either entry, not because all three rows are traversed.
A source-change return
The allocation relied on an argument now unavailable for a required new recheck. The earlier answer is preserved, but the dependent claim about current use has no recoverable basis. The reader returns only that claim and its source relation. A separately supported allocation and review remain usable.
If the user's question changes from reviewing the description to authorizing publication, it is another receiving use. The language points to the applicable publication result; a completed review is not that authorization.
Bias-Annotation
A procedure-shaped source encourages a fixed workflow; a graph tool encourages graph-shaped explanations. Start with the questions users bring and the results each answer depends on before choosing either form.
The worked arrangement demonstrates language and identity distinctions. It is not evidence that every Method engineer benefits from the same grouping or that the resulting profile is an enacted composite Method.
Conformance Checklist
- The language begins from recognizable uses and first results.
- Each body has a recoverable subject, practical contribution and ordinary stop.
- Every MethodDescription claim identifies its one admitted Method.
- Each material relation states what it changes in use and retains its own governing basis.
- Direct descriptions and sufficient existing entries remain available.
- The whole/profile account explains shared conditions and reasons without replacing the full bodies.
- Reconstruction and changed-condition use can return a localized failure or a wider gap.
Common Anti-Patterns and How to Avoid Them
Consequences
The language supports problem-first reuse and makes its connections inspectable. Users can select a small contribution, follow only a necessary dependency and revisit affected claims after change.
The cost is maintaining the cross-pattern explanation and relations. A direct description is preferable when that extra structure does not change recurring action or save meaningful reconstruction.
Architectural Rationale
A source-to-language architecture lies between source allocation and situated use. It must show how independently useful descriptions and neighboring results contribute together while preserving their different subjects.
A.3.2.DE alone does not decide that whole arrangement; E.11.PFP alone does not supply its Method-specific content.
The selected structure can be useful without an admitted composite Method or a formal CGUS. Progressive exactness lets ordinary users act from a clear result and stop while preserving a path to the stronger claims when a later use needs them.
SoTA-Echoing
Ralyté, Deneckère and Rolland's SME Map (2003) supports on-the-fly construction. Continuous SME composition and enactment is an implemented alternative. Adopt their support for adaptive use.
Current E.8 and E.11.PFP supply the form of an individual body and of the account of the whole. A.3.1, A.3.2, B.1.5, A.22.CGUS and C.39.MC supply the separate identities and use results. This pattern adds the Method-specific language arrangement after source allocation.
Against the cheaper direct description in the worked case, the language adds independently reachable allocation and reconstruction questions. It earns that cost only when those questions recur or change action.
No general advantage over prose, a method base or a fixed suitable route follows from the architecture. Reopen when a direct entry becomes sufficient, a relation loses its basis, or changed use makes the grouping burdensome.
Relations
- F.0.2.MA supplies semantic allocation. A.3.1.QM, B.1.5.CM and A.3.2.DE supply Method qualification and description.
- E.8 and E.11.PFP govern body and whole-account form. C.2.1 governs retained account identity.
- Use B.1.5 to establish whether and how the constituent Methods form the proposed whole. A.22 and A.22.CGUS govern selected and unfolding structures.
- C.37.CF supplies a bounded description comparison when content or representation claims matter, using C.2.8 when structural recovery is the selected question.
- F.0.2.RC tests reconstruction and affected refresh. C.39.MC supports situated use without requiring production.
- C.37 guides representation selection for one action. C.37.MP relates those selections across actions in a Method representation profile. A.3.1.MV guides variant development and identifies what to maintain when a change leaves the Method itself intact.
E.8.ML:End
Author an FPF Pattern from an Accepted Evaluation CharacteristicSpace Specification
Type: Authoring method pattern Status: Stable Normativity: Normative
Problem frame
Use this pattern when an accepted EvaluationCharacteristicSpaceSpec constructed or repaired under A.19.ECS has been selected for durable FPF publication, and an author must turn it into a practitioner-facing pattern. The author must explain how to perform the evaluation defined by that specification so its values remain usable, reviewable and bounded.
A.19.ECS guides construction or repair of the specification: what is evaluated, for which use, how values are justified and what the result permits. Section 4.3 below identifies the complete accepted content to publish. E.8 supplies the ordinary FPF authoring form. E.8.ECSPF tells the author how to carry the accepted specification into that form.
The specification, its CharacteristicSpace, the authored pattern content, a later evaluation of an object, and the result of that evaluation remain different things.
Not this pattern when. Use A.19.ECS when the characteristic-space specification is missing or inadequate. Use E.8 when the pattern is not an evaluation-characteristic-space pattern.
Use E.21, E.9.DA, E.2.DA, F.18, C.25, or a project-local evaluation when one already supplies the value meanings for the evaluated object and use. Use E.22 to frame one quality evaluation and E.23 to run repeated improvement.
Use a local rubric, table, or project rule instead of an FPF pattern when the evaluation is not intended for durable FPF reuse.
First useful move. Start from the accepted A.19.ECS specification. Before presenting coordinate tables or conformance rows, name the evaluated object kind, declared use, working reader, qualification window, and first action-guiding evaluation use in the pattern's recognition text.
FPF-publication boundary. If the evaluation is local, temporary, or project-specific, do not publish an FPF pattern. Keep the A.19.ECS specification in the local publication form. For each neighbouring FPF pattern the evaluation uses, cite the pattern and name its contribution.
What goes wrong if missed. The pattern, the accepted specification, the evaluation, and its result collapse into one supposed object. The pattern then becomes a score sheet, review form, checklist, or taxonomy. The coordinate table appears before the working situation.
Readers can see values but cannot tell when to use them, what to do after an evaluation result, which objects are outside the declared evaluated-object kind, or which neighbouring pattern supplies the needed evidence, assurance, gate, work, decision, naming, measurement, or improvement guidance.
What this buys. E.8.ECSPF lets an author publish evaluation guidance as a pattern: practitioner-readable first, exact enough for review, and bounded enough for a later evaluator to use with the framing guidance in E.22 or the repeated-improvement guidance in E.23.
Primary EntityOfConcern in plain terms. The primary EntityOfConcern is the authored FPF pattern content and its publication form for one accepted evaluation characteristic-space specification.
Primary working reader. The first reader is an FPF author or reviewer turning an accepted evaluation characteristic-space specification into a reusable FPF pattern for later practitioners, managers, and stewards.
Problem
An author can use A.19.ECS to produce a good evaluation characteristic-space specification and still lack guidance on publishing it as an FPF pattern.
The author can use E.8 to produce a good generic FPF pattern and still lack guidance on where to place a coordinate set, object-kind-fit rule, evidence basis, result-row shape, calibration points, status set and stop condition when they are the pattern's main content.
Recurring failures:
- Publication-form/content collapse. The accepted specification, its
CharacteristicSpace, the authored pattern, a later evaluation, and the evaluation result are treated as one object. - Table-first pattern. Coordinate rows arrive before evaluated object kind, use, first move, FPF-publication boundary, and object-kind boundary.
- Checklist substitution. Conformance rows replace the
Solutioninstead of checking a readable evaluation method. - Underpublished values. Coordinate names are present, but reader or qualification limits, value meanings, missingness, polarity, protected trade-offs, comparison rule, status meanings, neighbouring exits, or stop and reopen conditions are missing.
- Wrong-kind examples. Worked cases show only passing examples, so the pattern cannot teach below-floor and outside-declared-object-kind boundary outcomes.
- Neighbour theft. Claims about evidence, assurance, gates, work, decisions, naming, measurement, OEE or NQD, or mathematical lenses are carried as if this evaluation-characteristic-space pattern defined or justified them.
- Pattern-quality confusion. The author uses
E.21to judge whether the FPF pattern version is good, but forgets that the new pattern must still carry the accepted evaluation characteristic-space specification for one evaluated object kind by value. - Quality-carrier leakage.
E.21values, corpus projection, README/ToC/E.11 alignment, retrieval, cold-reader evidence, monolith parity, landing evidence, or developer/reviewer/executor correspondence for the publication form are written into the evaluation pattern as if they were the evaluated object's method.
Forces
Solution
When an accepted A.19.ECS specification is selected for durable FPF publication, use E.8 to write a pattern that teaches the specified evaluation, with these additional placement rules:
-
Keep the objects separate. The accepted specification says what the evaluation requires. The publication form arranges the pattern. The authored content teaches a later practitioner how to evaluate an object. That later evaluation produces a result. The specification, its
CharacteristicSpace, the evaluation, the evaluated object and the result each remain distinct from the pattern. -
Put recognition before coordinates. The opening text names evaluated object kind, declared use, working reader, qualification window, first evaluation use, FPF-publication boundary, what goes wrong, and what the pattern buys before any dense table.
-
Carry the complete accepted specification by value. Put every required value, and every optional value whose trigger holds, where a practitioner needs it. Do not discharge this move by citing
A.19.ECS, copying field names, or pointing to an author-only record. TheSolutionand its nearby practitioner-use sections carry the selected values from the accepted specification. -
Use worked slices as the discriminating-case test. Archetypal Grounding and worked cases include a passing evaluated object, a below-floor evaluated object, and an outside-declared-object-kind boundary case.
-
Keep ordinal coordinates separate and protect against proxy improvement. Do not create an undeclared total, average, or “overall score” from ordinal coordinates. Whenever a visible value improves, ask whether any intended value or protected trade-off became worse. If the published guidance would reward that loss, stop the comparison and reopen the specification.
If a bounded use needs scalarization, name the method, its use, any information loss, and its applicability and stop or return conditions; do not present that scalar as “the evaluation”.
-
Keep checklist rows secondary. Conformance checks verify that the evaluation is recoverable and usable. They do not become the user's method.
-
State the concrete contribution used for each outside claim. When
Relationsor a grounded local boundary makes a claim about, for example, evidence, assurance, work, naming, measurement, or improvement, cite the applicablePatternIDand say in ordinary terms what its content contributes here. It may supply an evidence-use boundary, an assurance calculus, a gate decision rule, a measurement test, repair guidance, or something else; these are examples, not a closed vocabulary.The
PatternIDis enough for ordinary use. Identify an assertion, episteme edition, orClaimGraphonly when interpretation, migration, conflict, publication, or reuse depends on that identity. Treat guidance as aU.Method, a qualifyingU.MethodDescription, or a Method use only after its own admission test passes and the current claim needs that identity.Use
F.19for ordinary wording repair. When a repair can change an FPF-governed meaning, confirm that the evaluated object and its kind, relation or claim kind, live ontic slot, relation position, use relation, admissible use, and scope remain recoverable before and after the repair, as applicable to the changed claim. -
Keep pattern-quality evaluation separate. When a pattern-quality conclusion is required, a reviewer uses
E.21to evaluate that pattern version. A later evaluator uses the guidance published in the pattern to evaluate the declared object kind.The
E.21result, corpus-projection evidence, README/ToC/E.11 alignment, retrieval or cold-reader evidence, monolith parity, landing evidence, and developer/reviewer/executor correspondence stay in the quality, review, projection, or release carriers unless the pattern's ownEntityOfConcernand user-facing action are that evaluation or projection work.
The authoring flow and the quality-improvement flow are different. First an author carries an accepted specification into a pattern. Later a practitioner may use that pattern's guidance to evaluate an object and record a result. E.22 and E.23 provide guidance for framing or repeating that work.
A reviewer's later E.21 evaluation of this pattern is evidence about the authored pattern, not part of the object evaluation that the pattern teaches. That evidence may cause edits to recognition text, coordinates, cases, or boundaries, but it remains outside the pattern unless rewritten as user-facing evaluation guidance.
Canonical placement table
Local names and kind settlement
A later practitioner uses the authored guidance to evaluate an object and obtain a result. Establish the identity of a MethodDescription, Method, assignment, or dated Work only when it matters to the receiving claim. The specification, its CharacteristicSpace, the authored pattern, that evaluation, and its result remain distinct.
By-value carry-through
Carry the accepted specification through the pattern in practitioner order. “By value” means that the reader can find the selected value and use it; a field name, an A.19.ECS citation, or an author-only attachment is not enough.
The fields may be expressed in plain language, tables, or worked cases, preserving the accepted ResultRowShape and any serialization required by the receiving use.
Keep them close to the practitioner action they qualify. Do not hide required values in conformance rows, source notes, or review evidence.
Archetypal Grounding
Guidance based on an evaluation CharacteristicSpace becomes reusable in FPF only when a practitioner can recognize the evaluated object and use before reading the coordinate table. The publication form must teach practitioners how to carry out the evaluation. A list of values alone is insufficient. The following slice shows the author's move from an accepted specification to practitioner-facing content.
Accepted specification. An author has an accepted EvaluationCharacteristicSpaceSpec for one version of a field-service handover instruction.
Corresponding recognition lines in the authored pattern.
Use this pattern when you must decide whether a field-service handover instruction is ready for a supervised first use by a maintenance lead who did not write it. Use it only for the named equipment configuration and safety-procedure edition. Give that reader the current instruction. Ask them to identify the first move and the condition for stopping or escalation. In the supervised trial observe whether they can carry out the move when its conditions permit. Record where they obtained the stop condition. A spare-parts catalogue is outside this evaluation.
These lines carry the selected object kind, use, reader, qualification window, first move, and wrong-kind boundary. The citation “see A.19.ECS” alone would not convey them.
Minimal Solution and result form. The pattern then tells the practitioner to use the current instruction version, observe the cold-reader trial, judge both coordinates from their stated value meanings, and record both coordinate results. For example:
The result is repair because one coordinate is below the declared ready value. The revised instruction puts the stop condition before the first move and states when to stop or escalate. In a cold supervised trial, a lead who did not write the instruction states the first move from it, identifies why the condition permits proceeding, and carries out that move. These observations support both 2 values and ready for supervised use for this configuration and edition. A spare-parts catalogue returns to evaluation selection before these rows are opened.
A simple A.10 citation is enough to locate the evidence-use discipline for this ordinary case; identify the assertion or ClaimGraph only if later interpretation or reuse depends on its identity.
Near miss, proxy improvement. A shorter instruction makes the first move easier to find. In the stipulated cold trial the lead also states and performs it from the instruction, establishing FirstMoveRecoverability at 2. The editor removed the only statement of the stop condition and its reference, so HazardBoundaryVisibility is 0. Finding the shorter sentence alone would leave performance untested.
The author must not add or average those ordinal values and call the rewrite better. The protected safety trade-off has been lost, so the pattern returns repair and the accepted specification must be reopened if its current status rule would reward that rewrite.
Pattern-quality evaluation. E.21 is an evaluation for one FPF pattern version. Its publication form must still open with the working question "is this pattern good enough for the declared use?" before showing coordinates such as first-action recoverability, boundary fit, and SoTA binding.
Local rubric. A project team defines a temporary rubric for choosing a meeting room. The A.19.ECS specification may be adequate locally, but no durable FPF pattern is needed because the evaluated object kind and use do not recur across FPF practice.
Object-kind boundary. A nuclear-plant evaluation can judge nuclear plants and declared comparable power-generation alternatives. A plant inspection report can supply evidence about the plant but is outside that evaluated-object kind: before the evaluation is opened, select a suitable evaluation; after a forced invocation, record an object-kind-fit defect/value rather than treating it as a weak nuclear plant or skipping declared coordinates.
The pattern publication form must show that boundary before readers try to use the coordinate table.
Bias-Annotation
Evaluation-characteristic-space patterns are vulnerable to domain-example bias: the first examples can silently choose the evaluated object kind, use, and value family for later readers. A conforming publication form names known skew in examples, sources, reader family, domain tradition, measurement preference, benchmark preference, or FPF-internal reuse.
When the evaluation claims broad use, establish why the accepted value meanings and their grounds apply across that use. Include additional situations where a material difference affects evaluation. Retain the accepted discriminating cases specified in §4.3.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
A conforming E.8.ECSPF publication form makes evaluation guidance findable, teachable, and reusable inside FPF. It lets a practitioner frame an evaluation with E.22 or repeat improvement with E.23 without re-inventing values.
It also makes the cost visible: a reusable evaluation pattern must publish more than a local rubric, because it must prevent wrong-kind use, hidden value drift, neighbour theft, and proxy-for-value substitution.
The pattern's output is bounded evaluation guidance and the form of its result. For product certification, release approval, an evidence claim, or an improvement decision, use the applicable pattern and establish the required result or decision.
Rationale
The split between A.19.ECS and E.8.ECSPF preserves the distinction between an evaluation characteristic-space specification, the pattern that teaches its use, a later evaluation, and the resulting record. A.19.ECS says what the specification must contain. E.8.ECSPF says how to carry that accepted content into an FPF pattern when durable publication is selected.
This prevents two symmetric mistakes: stuffing FPF pattern-format requirements into a general characteristic-space construction method, and publishing guidance whose accepted coordinate set is not recoverable by value.
SoTA-Echoing
Source-use convention and qualification. The current-source decisions below are qualified through 2026-08-15 for the identified editions and this publication-form question. Each source is used only for the content named in its row. Reopen the smallest affected row when a new edition, successor, or materially better competitor changes that adopted content, its scope, or its currentness; a bibliographic change alone does not reopen the pattern.
Relations
E.8.ECSPF:End
Design-Rationale Record (DRR) for FPF Content Decisions
Type: Governance and authoring pattern Status: Stable Normativity: Normative
Use this when
- A proposed normative change needs a statement of what FPF should say, why that answer is preferred, and which neighboring patterns or selected non-pattern FPF kind-reference pairs it affects.
- Several such loci must change together. An external record must keep the complete content decision available while authors distribute enduring text across Core.
- Authors would otherwise have to settle the same consequential question separately in several patterns or selected non-pattern FPF kind-reference pairs.
- Authors need to review a deprecation, narrowing, or cross-pattern amendment without reconstructing its rationale from patches, conversations, or scattered notes.
Not this pattern when. Normative Core rules belong in their selected loci. Keep campaign and process briefs outside the DRR. If a purely editorial Delta-0 or Delta-1 change needs a decision record, use the lightweight variant in CC-DRR.5 when it fits. Use E.9.DA when a DRR already exists and you need to judge its answer, obligations at selected loci, source use, lexical closure, and drafting actionability for a declared downstream authoring use.
What goes wrong if missed
- Later readers cannot recover why Core changed, which alternatives were rejected, or which exclusions were intentional.
- Coordinated amendments drift apart when the selected answer survives only in patches, handoffs, or a reviewer's memory.
- Later repairs can fit local wording while losing the constraints supplied by Pillars, taxonomy lenses, the impact graph, practical use, or pattern placement.
What this buys
- An external record states the content of a bounded FPF change before authors rewrite Core.
- Four components keep the Problem frame, Decision, Rationale, and Consequences available for later review and replay.
- For a distributed change, the record holds the selected answer while authors draft the enduring Core text at the selected loci.
First useful move. State the working FPF problem, selected answer, practical change, selected loci, first substantive drafting action, and nearest boundary in plain, precise language before drafting or landing Core text.
Cheap stop. If local wording repair, application of an accepted pattern, or editorial cleanup preserves FPF semantics, obligations, boundaries, names, admissible uses, and normative force, it needs no full DRR.
Use F.19 for kind-preserving plain rewriting and E.10 for an unresolved FPF wording-use question. Use E.17.AUD.LHR for an overloaded local lexical head inside one publication unit and C.2.P for an episteme, publication, or source-use distinction still unresolved after E.10. Use F.18 only for minting a durable reusable name and E.8 for an authoring-form correction. Use E.9 when a bounded content decision needs its rationale stated.
Kind-or-boilerplate diagnostic. Before proposing wording from a DRR as pattern prose, use F.19 to separate boilerplate from substantive content. If the remaining wording still obscures its use, name, relation, claim, admissible use, selected locus, user action, or flow position, name the contribution applied: E.10, E.10.ARCH, F.18, or the pattern that defines the object or relation. Keep process, architecture, review, and reference boilerplate in the material it belongs to.
Before using proposed wording in a pattern, establish from the selected answer which object, relation, claim, slot, use, admissibility condition, or scope would change, or that none would change. Use F.19 and the pattern that defines or constrains the distinction. Introduce Method, Work, application, or ClaimGraph details into ordinary wording only when that distinction changes the decision or a later reliance.
Primary EntityOfConcern in plain terms. The external decision-rationale record for one bounded FPF content decision or coordinated change set. The author must be able to recover the problem, selected answer, reasons, consequences, practical change, selected loci, and boundary without inventing them. Add exact C.2.1 identity only when the decision or a named later reliance needs it.
Primary working reader. An FPF author, reviewer, or steward uses the DRR to evaluate, challenge, or land a bounded content decision. Downstream pattern readers can use the guidance once it has been incorporated into Core.
Problem frame
Under Pillar P‑10 Open‑Ended Evolution, FPF's normative rules must adapt as new calculi and insights arrive. Changes whose answers and reasons cannot be recovered erode those rules. A formally detailed record can still defer the decision or distribute a harmful rule. A Design‑Rationale Record (DRR) preserves the selected answer and its reasons for one bounded change before authors distribute its normative content.
Problem
Either a direct edit to Core or a DRR that documents the apparatus but leaves the answer unusable can expose the following risks.
- Lost decision grounds – future authors cannot recover why the answer was selected, which alternative or boundary mattered, or when to reopen it.
- Decision deferral – a polished record leaves the answer, affected loci, or first drafting action for each later author to invent again.
- Proxy-led fanout – an abstract rule passes a schema, checklist, or unrelated replay while making an actual predecessor/proposed host harder to understand or use.
- Conceptual and authority drift – incremental edits blur the framework's foundations, or the temporary rationale record becomes a shadow Core law, process brief, or authorization source.
Forces
Solution — state the decision before distributing it
Recover the decision the receiving practice commissioned. Who will use it, which alternatives can they choose, and what changes for them if those alternatives differ? Where the choice concerns several wholes, identify the relevant parts of each and the connections on which the result depends. Where it concerns alternative models, identify what their differences allow the practitioner to decide about the same subject. A decision about how to describe an architecture can contribute to a decision about that architecture. State that contribution and what the architect still needs to settle.
Present the Decision first in plain language. Before introducing the DRR schema, a Method or Work account, or a catalogue of alternatives, let the reader recover the following:
- the working FPF problem and why it matters now;
- the selected answer stated positively;
- what changes in practitioner or authoring use;
- the selected loci and the positive obligation each one carries;
- the first substantive drafting action; and
- the nearest boundary, blocker, or reopen condition.
If the answer is unsettled, develop it through a working expression under §4.2. The order above is for readers of the resulting decision; inquiry can start with an open question.
Once supported, the short decision account is the primary source for drafting. Add exact Method, Work, application, episteme identity, source use, assessment, or authority only when the decision or a named later reliance depends on that distinction.
A nontrivial DRR must make the four components below recoverable. They form the minimum decision kernel. CC-DRR.5 gives the lightweight editorial variant.
In this pattern, a bounded coordinated change set brings together mutually dependent content-decision questions within one decision boundary. Several patterns or selected non-pattern FPF kind-reference pairs carry the enduring FPF expression of that decision.
The selected answer is what FPF should say, which selected loci carry it, what practical action changes, what stays outside, and which source-use, evidence, validation, or loss/recoverability conditions remain live.
A selected non-pattern FPF kind-reference pair is a content-distribution instruction, not a new kind. It names an admitted FPF kind and one exact reference by value—for example a U.View, source map, source-use note, evidence-path record, review-finding record, or architecture-decision record. The selected patterns and such non-pattern items are the selected loci of the decision.
A temporary convergence record holds the selected answer while authors update several selected loci. Enduring normative rules belong in Core, and process state belongs outside the DRR.
Keep a rejected alternative only when it explains the selected answer, a live boundary, or a reopen condition. Pillars and taxonomy lenses may inform the decision, but the DRR records their load-bearing effects rather than rehearsing every lens or preserving the history of discussion.
Before adopting a language, ontology, or authoring rule for wider use, apply it to at least one real host and its proposed replacement, including their dependencies. At comparable effort, compare the recognizable entry, required inputs, first action, practitioner vocabulary, formality and assurance burden, first useful result, stop or return, preserved useful ideas, and direct consumers.
Use that comparison for the uses and conditions it reaches. For a rule proposed across a wider range, show which feature of the rule preserves the required work as the objects, their organization or the reader's conditions vary. A general argument can occur inside the host comparison. A successful use with the original conditions supplies a result for those conditions.
The proposed wording must pass the E.8 first screen and the F.19 kind-preserving plain-rewrite test. A schema, invented fact pack, unrelated lane test, checklist, or promised review cannot replace this comparison. If the pilot makes the host harder to use without a compensating semantic gain, repair or reject the rule before applying it elsewhere.
A DRR records the selected answer. When exact identity or reliance makes the distinction material, separately identify the decision work and method, selected-answer result, C.2.1 DRR episteme, source-use relations, assessment and result, any acceptance or authority, and later realization work. An exact DRR episteme then follows C.2.1 identity by <ClaimGraph, EntityOfConcern, effective ReferenceScheme>; ordinary use does not require materializing that whole account.
Minimum decision-inspection content blocks
A nontrivial DRR must also make the following parts of the decision available for inspection. Authors may present them within the four kernel components or in a dedicated Decision grounds used or decision-inspection block. They are part of an adequate decision before it reaches a later review.
These blocks supply the content of the decision. Four headings cannot tell a later author the grounds, first-minute working situation, naming choice, distribution, sufficiency or necessity of a pattern or non-pattern FPF kind-reference pair, treatment of overlap, impact, or remaining uncertainty. Make those answers recoverable in the DRR without creating separate process paperwork.
For a coordinated change, settle the common answer in the DRR before drafting. Each author can then express that answer at the selected locus without having to decide it again.
When a decision connects several selected loci or mutually dependent edits to patterns and non-pattern FPF kind-reference pairs, its DRR MAY include more than the minimum kernel. It can state obligations at each locus, the choice between a new and an existing pattern, how impacts or non-goals map across the selected loci, coverage or agreement across them, and whether overlapping changes converge. A provisional account of the decision rule can hold the full bounded answer until its enduring content is distributed in Core.
These additions leave the DRR's kind unchanged. It conforms only while it records what FPF should say and why, what is excluded, how the selected loci are affected, and what improves in practice or authoring. A record with richer coordination content MUST NOT become a campaign plan, process script, baton carrier, packet checklist, staging log, or other development-process brief.
If an existing pattern or selected non-pattern kind-reference pair might suffice, or a new locus might be needed, the DRR must settle that choice. A tentative list leaves the choice to the next author.
Classify each pattern or non-pattern kind-reference pair that the accepted grounds or the DRR bring into the distribution question. State its obligation if selected. Otherwise identify it as an unchanged boundary neighbour, inherited unchanged, or outside the current decision with a named pattern, kind-reference pair, or decision record. Saying it may need local hardening leaves that choice open.
If the grounds reveal a potentially reusable non-pattern carrier or a related source-use, evidence, assurance, validation, or architecture-decision mechanism, the decision must classify it as generalized now, kept local with a reason, rejected, or outside with a named locus or decision record. The DRR records the choice; mentioning an artifact does not decide its use.
For source loss, simplification, redaction, summarization, or other declared loss, the DRR must state which distinctions survive and which are lost. Identify each admissible branch reading and its selected non-pattern FPF kind-reference pair, the uses without an admissible carrier or evidence path, the recoverability class, and the stop or reopen rule when the use exceeds the declared loss or narrower scope.
Before another author can use a nontrivial DRR to draft the selected changes, it must settle the material choices about its EntityOfConcern, selected loci, outside boundary, reusable content, and loss or recoverability regime. Each is selected, rejected, inherited unchanged, or placed outside with a named locus or decision record. While any of these choices is missing, the work is still establishing the decision grounds; it has not produced an accepted DRR.
The DRR episteme lives outside normative Core. A separately governed acceptance, authority, or realization decision may rely on it, but the word accepted, a record status, review mark, or publication does not make its claims true or authorize change.
Once realization is separately authorized, authors apply the selected answer to its loci in dated authoring work. The changed Core content, authoring work, result claims, checks, witnesses, publications, and any landing or release record remain distinct; apply the relevant pattern to each claim. The DRR remains external provenance and temporary convergence support; after realization, the enduring semantics must be available in Core.
Authors using a separately accepted selected answer may elaborate examples, SoTA-Echoing, recognition sections, local wording, and neighboring fit inside its declared stability boundary. A change to the selected answer, selected loci, outside boundary, reusable-content disposition, or loss/recoverability regime requires a successor decision result and DRR episteme rather than a silent edit to downstream prose.
Improvement work may apply E.23 to a DRR episteme. That work, its method applications, quality-result claims, and witnesses are separate from the DRR and do not turn the record into a pattern draft. When a successor decision relies on SoTA, its result must show what changed in the selected answer, locus obligation, boundary, example, validation obligation, or reopen condition; otherwise the source use remains rationale-only or lineage-only.
When a campaign changes route-shaped, unfolding-shaped, first-entry, DPF or multi-pattern path material, make the connection from its actual starting situation to the selected solution and next use recoverable. The following optional form can carry those contents; include only fields that describe the current decision:
After host drafting, identify any enduring selected content that remains available only in the DRR. Move it into the selected pattern body, unfolding block, README seed, E.11 expansion or concrete relation locus so that the public reader can obtain and use the solution.
To preserve P-2 Didactic Primacy without duplicating meta-text, realization work using a separately accepted selected answer should distill stable Rationale, Consequences, SoTA-Echoing, Grounding, and other valid convergence content into the selected informative pattern loci under E.8. The DRR episteme remains external provenance; it is not itself landed or transformed into Core.
A substantive DRR is one claim-bearing episteme about one bounded current content-decision question/change set. It may carry selected obligations only in its Decision or Consequences, but no route, gate, handoff, packet, monolith, mutable status, or future-campaign state. Any undecided remainder is explicitly outside the decision with a named pattern, kind-reference pair, or successor decision record.
Process-source method admission into FPF
When an author considers a stable method described in a process source, the author decides the FPF-admission disposition by value. The DRR records that decision and any source-use relation that matters; neither the source passage nor the record becomes a second canon.
Record the following when deciding whether to admit the method:
- the content of the passage or accepted decision ground being considered from the process source;
- the reusable FPF method recovered from that passage;
- the current FPF pattern, section, or accepted
DRRthat already carries the method, if any; - the contribution still missing from FPF;
- the FPF pattern selected to carry that contribution;
- process-control material excluded from FPF pattern prose, such as task dispatch, seam state, helper behavior, Git recovery, packet transport, review transport, chat cadence, and mutable release state;
- how the passage or decision-ground item was used: quoted by value, narrowed in scope, instantiated as a case, used as decision grounds or drafting guidance, retained only as an example, or retired;
- any meaning loss or addition created by that source-use result: changed scope, relation, evidence path, admissible use, non-admissible use, reader use, or recoverability condition;
- the first improved FPF use that the admitted method gives to an author, reviewer, or downstream FPF user;
- the disposition: selected now, inherited and sufficient, rejected now, or outside this decision. For an outside item, name the evaluation pattern or accepted
DRRthat carries it, or state the content of the accepted decision ground that does so.
Reusable process-source methods may contribute beyond semio wording or pattern-authoring language. Admit the contribution when it is separable from local process mechanics, improves the selected FPF use, and has adequate grounds for the claims being made. Apply E.19 to pattern admission and E.21 when a pattern-quality conclusion is required. After the method lands in FPF, process documents should cite the selected FPF pattern instead of keeping a parallel long-form rule.
Develop an unsettled decision through a working expression
If the answer is unknown, start with the difficulty and the material available. Use §4's order to present the answer once selected. Filling the Decision block in advance would require the author to guess what the inquiry is meant to establish.
Use a provisional expression to bring a consequential relation into view: juxtapose two cases, arrange competing explanations, trace a dependency, or compare an intended use with the result currently available. E.8:4.2.1.2 and A.6.3.RT.OE supply the expression construction. Identify what the arrangement lets you infer and what remains unsupported. A question mark can make an unresolved interpretation visible; it supplies no premise for choosing one answer.
When the evidence reveals several questions inside the original commission, distinguish them and show how their answers bear on the recipient's decision. Return any open choice to the person entitled to make it. For example, agreeing where a decision will be recorded settles its publication only. If the commission also assigns a choice between architectures, compare their consequences for the work they must support. Retain a working table while its distinctions help make or explain that choice.
Before another author relies on the selected answer, resolve each premise that can change its selected obligation, or bound the result so that the remaining uncertainty does not defeat that obligation. If neither is possible, state where the decision must stop. Use the four kernel components and the applicable inspection content in §4.1 to inspect that closure. Material already recoverable in the account need not be copied into another answer column. A source pointer without the needed substantive contribution is not recoverable decision content.
Keep this difference visible in the writing. “The source order for this row is unknown” can be a valid investigation result. “Convert this row using day/month/year” would require the missing premise. Conversely, a settled decision to stop such rows and request their interpretation can be supported even while their dates remain unknown. What remains open must be compared with the decision being made.
The record carries the resulting problem, selection, reasons and consequences, not every discarded worksheet or the chronology of deciding. Retain a working expression when a later use needs its reasoning or recoverability; otherwise carry its useful result and source return into the decision.
Archetypal Grounding (System / Episteme)
Decide what support a conversion pattern should carry
An author is revising the date-conversion explanation in E.8:4.2.1.1. The explanation already supplies the essential rule: use the applicable supplier guarantee to choose the date order before running the converter. North guarantees month/day/year, South day/month/year for the files at issue. The same text, 03/04/2026, accordingly yields March 4 or April 3. The intended practitioner can run a converter with a supplied order. The author can follow that reasoning but has yet to decide whether the public method should prescribe a table, a per-row card or prose alone.
A first working note says: “Choose a standard date card; include source, raw value, order and converted value.” It proposes a form before establishing whether all four entries are independent user inputs or what their arrangement must let the reader do. Turning those words into a completed Decision would hide the open question.
The author constructs the raw/result expression and its unresolved cell shown in E.8:4.2.1.2. It keeps both possible results visible. Trying to select an output from the raw text alone fails. The supplier difference leads to the source-qualified comparison. This yields a more useful decision question: which source-to-interpretation connection must the pattern explain, and which expression makes that connection usable without requiring unnecessary repeated entry? The comparison shows which information must be available and which the user must enter. Removing repeated entry still requires access to the source information.
The author now compares three alternatives for practitioners with the same preparation:
The comparison does not demonstrate that every reader is faster with a table. It establishes different supported operations and context dependencies. The author selects the third alternative for the explanatory example, retains the first as a sufficient compact use, and rejects the universal manual-card requirement. The selected pattern account has to teach the conditions of reducing the view and restoring the missing context.
A short decision account can now carry the useful result:
- Problem frame: the existing conversion method is available, but authors lack the worked choice between explanatory comparison and a mandatory record for each conversion.
- Decision: develop the expression-construction explanation in E.8:4.2.1.2, with the source-qualified comparison and a conditional reduced view. Preserve §4.2.1.1's source-to-order selection and stop when the date order is unknown. Do not prescribe a card for every row or change the converter's subject method.
- Rationale: the different outputs for identical raw text require an interpretation supplied by the applicable source. A juxtaposed comparison teaches this relation. Re-entering a value already supplied by the session is a separate burden with no additional contribution to that local operation. Detached reuse changes that judgement.
- Consequences: keep the guarantee's scope and source return beside the operation; restore row-level context for mixed or separated use. First draft the explanation and the expressions being compared at the named locus. Examine whether the intended reader can recover and apply the selection under the stated preparation. Do not claim observed learning, speed or universal usability from this construction.
These paragraphs supply the bounded content decision; they are not answers copied into every inspection row. Its exact grounds are the existing E.8 explanation, the stipulated supplier guarantees and the performed comparison. The related conversion operation remains unchanged. No new date kind, source-independent interpretation or general record schema is selected. A changed supplier guarantee reopens the affected interpretation; an actual reader failure would reopen the explanation or its support.
Now South changes to mixed orders without row-level information. The author can still settle the content decision to preserve the earlier results, represent the new unknown and stop that row's conversion pending a trustworthy interpretation. The missing date order does not prevent explaining that boundary. It does prevent selecting a particular converted date. If the proposed public method instead promised to convert every row without obtaining such information, the author would have to change that promise or stop the dependent decision.
E.9's four components and inspection content remain stable aids for making the final selection recoverable. The provisional card is not preserved merely because it was the first form used. Its useful source-to-interpretation distinction survives in the selected account, while unnecessary per-row manual entries are removed. Here E.9 applies to the choice of support for an author: the working expressions helped form the question and select an answer; the record preserves that supported answer for the next author.
Bias-Annotation
This bias annotation applies to every FPF semantic change governed by E.9. It does not turn project-management state, helper state, or review logistics into DRR content.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
FPF evolves through explicit, reviewable decisions rather than silent edits. The DRR is the minimum structured argument and, when several loci must move together, a temporary convergence record. This keeps P-10 Open-Ended Evolution compatible with P-1 Cognitive Elegance and P-2 Didactic Primacy. Exact method, work, result, episteme, assessment, and authority distinctions remain available when a current claim depends on them; they do not precede or replace the readable decision.
Every nontrivial conforming DRR must make Problem frame, Decision, Rationale, and Consequences recoverable. It may also carry the coordination content that prevents shadow documents or new decisions being invented during distribution to Core patterns and selected non-pattern FPF kind-reference pairs.
The minimum includes what authors and reviewers would otherwise have to reconstruct: exact grounds, use-value, the first-minute working situation, scenario, alternatives and current dispositions, naming or ontology obligations, content distribution, whether an existing pattern suffices or a new one is needed, overlap, stability of the answer, the graph of impacts and boundaries, practical payoff, and uncertainties that could change the decision.
Pointer-based DRRs (CC‑DRR.1a) prevent duplicated prose, and distribution into Core patterns and selected non-pattern FPF kind-reference pairs (CC‑DRR.4) keeps the specification itself learnable without turning the DRR into a permanent shadow canon. Process-law ordering, gate, and handoff records stay outside because they are not part of the content answer that FPF is selecting.
SoTA-Echoing
E.9 draws on established decision-record and design-rationale practices. A cited standard or template does not establish current SoTA for every FPF decision. The DRR selects current problem-owning evidence under E.8 when that evidence is load-bearing. Its distinctive contribution is a decision-rationale record for one bounded FPF content decision, with enough by-value rationale to distribute durable content without making the record a shadow specification.
MADR's template connects context, considered options, a justified decision and consequences; its confirmation section is optional. E.9 retains this concise rationale function. When one FPF decision changes several loci, it additionally assigns their content obligations and the conditions under which authors can use the selected answer. A longer record is useful only for those remaining dependencies.
ISO/IEC/IEEE 42010:2022 distinguishes an entity's architecture from its description and leaves architecting processes and recording media outside its scope. E.9 likewise keeps its decision record distinct from the selected architecture and the work that realizes it; the standard does not supply the FPF content decision or its admission result.
Van Heesch, Avgeriou and Hilliard distinguish viewpoints on decision detail, relationships among decisions, chronology and stakeholder involvement. A reader reconstructing why two decisions constrain each other needs their relationship. A reader arranging a review may need who was involved. Choose the content of the record from the decision its recipient needs to make or understand.
To improve the choice of content when a decision has substantial semantic consequences, the deciding work settles the answer, alternatives, losses, boundary, and loci. The DRR episteme lets another author replay that selection before drafting. Any durable rule, example, or obligation useful after realization belongs in the selected FPF pattern or non-pattern kind-reference pair, not in the DRR as permanent shadow canon.
When an identified source shapes the answer—for example a prior decision, standard, plan, or review packet—the DRR records how it is used and which payload is selected or left behind, including any material loss, destination locus, stop or return, reopen condition, and non-use boundary admitted by F.19's grounded-contribution test. Name the exact source episteme, publication, and source-use relation when the decision or a named later reliance depends on those identities. A citation supplies traceability; every stronger receiving claim needs its direct result or relation.
Relations
-
Instantiates: P‑10 Open‑Ended Evolution, P‑2 Didactic Primacy
-
Pattern publication form governed by: E.8.
-
Interacts with: E.5.4 for the lens check.
-
Complemented by:
E.9.DAwhen a particular DRR must be checked for a declared downstream authoring use. An ordinary bounded review judges the decision and returns precise findings or repaired text; a complete coordinate result and its exact assessment identities are added only when expressly requested or needed by a named later reliance. E.9.DA is not a second DRR form, review gate, acceptance status, or mandatory editorial step. E.12 separately governs didactic clarity, rationale and review of cognitive workload. -
Coordinates with:
E.23for repeated improvement work on a DRR; C.2.1 for DRR and evaluation-result episteme identity; C.2.P/A.10/G.6 for exact source use and provenance; A.15.1 for performed decision, assessment, and realization Work; A.6.1 when a claim depends on an independently identified application of a declared mechanism operation; F.10/G.11 for status and currentness; E.24.PUB for publication; and C.29 for a mathematical-lens use when such a lens is used. None of these neighboring records or results changes the E.9 selected answer by implication.
E.9:End
Evaluate a DRR for Its Declared Authoring Use (Decision-Adequacy CharacteristicSpace)
Status: Stable
Problem frame
Use E.9.DA to judge whether a particular version of a DRR can support a declared FPF authoring use. The use can be pattern drafting, host amendment, distribution across selected loci, carrying an accepted decision or source use through subsequent authoring, or deciding a scope boundary, split, or architecture hold. Specify the exact C.2.1 episteme identity only when the judgement or a named later reliance needs it. E.9.DA supplies the questions for this object and the reusable meanings of its coordinates.
For other questions, use E.21 to judge an authored pattern version, E.19 to review an FPF pattern edition or bounded subset for admission or refresh, and F.19 to repair wording, with E.10 for cues and unresolved meanings. Use C.16, A.17, A.18, and A.19 for measurement law.
First read the DRR in its declared use and state the working problem, selected answer, practical change, first drafting action, and boundary. If the DRR selects a broad authoring rule, inspect the effect on the actual host by comparing its predecessor and proposed versions before preparing any optional assessment or result materials.
A formally valid DRR can leave the author unable to draft. It may summarize sources without deciding, name neighbors without assigning obligations, hide rejected alternatives, leave trigger words unresolved, or omit the first drafting action.
What is being evaluated? One DRR version for one declared FPF authoring use and qualification window. Identify assessment work, a reusable coordinate result, witnesses, records, status use, assurance, acceptance, and later repair separately when the case involves them.
Problem
E.9 defines how to decide and record FPF content, the ordinary minimum DRR form, and the exact decision-work, result, and C.2.1 identities to establish when a claim or named reliance needs them. To judge adequacy, examine whether the DRR under review settles enough for the declared downstream use. A reviewer can approve the headings, source volume or prose while the author still has to invent a decision.
Recurring failures:
- The decision question is broad or implicit.
- The selected answer is a summary rather than a decision.
- Alternatives, rejected options, and outside-decision items are not closed.
- Receiving loci are named but not assigned content obligations or non-obligations.
- The selected FPF content architecture is explicit but wrong.
- Source use is copied without saying what changed in the accepted decision.
- Architecture descriptions, views, graphs, packets, or notes are treated as the FPF decision.
- Administrative state becomes adequacy evidence.
- Ordinal adequacy values become repair targets, so the
DRRgains source rows, locus tables, boundary catalogues, or review proof without making either the selected answer or the first drafting action more decisive.
Forces
Solution
Read the commissioning problem beside the proposed DRR. Recover the choice, who needs it and what that person is entitled to decide. Follow the record to its conclusion, including what it settles about the relevant wholes, parts or model alternatives. Would a clear record of a different decision also pass the stated test? Restore the question or the missing reasoning where the answers diverge. Then establish whether an author can draft from the selected answer, consequences, loci, first action and boundary.
The following result suffices for an ordinary bounded review.
- the exact DRR and declared authoring use;
- substantive findings, repaired DRR text, or the unchanged checked DRR when the review is clean;
- the actual predecessor/proposed host evidence when the DRR selects a broad language, ontology, or authoring rule;
- the first drafting action or first repair; and
- the stop or reopen condition.
Plain prose can carry this result. Reserve an assessment-work record, application object, aggregate result episteme, precision-profile record, witness package, or evidence-use package for a use that needs it.
Use the complete coordinate result when a complete reusable evaluation was explicitly requested or when a named later reliance needs stable coordinate values. Materialize the exact characteristic-space configuration, semantic evaluation Method, A.6.1 application, result episteme, witnesses, or evidence-use relations only when that receiving use depends on their identities.
A reusable coordinate result can exist without a semantic Method, A.6.1 application, or dated Work. Decide separately whether the receiving claim uses each identity. A claim may use a Method, or an exact application with its actual bindings, without asserting Work.
If the claim asserts dated U.Work, establish the Method enacted, every precise performer's A.13 core, and independent A.15.1 Work admission. Add an A.6.1 application only when the claim uses an independently declared operation and its obtaining application. Establish any relation between application and Work separately. Use F.6 only when the result needs precise assignment-bound attribution.
Before assigning coordinates, make one bounded search for an important question the DRR omitted. Inspect the governed problem, the problem-owning practice, current sources, the strongest live alternative, failure and recovery cases, and direct consumers. If an omitted question would change the answer, architecture, source use, consumer obligation, first drafting action, or stop, return it as a substantive finding before completing the coordinate judgement.
In either branch, distinguish the checked DRR, evaluation specification, action or Work, application, result, record, evidence use, status use, authority, and later repair. A compact result can omit identities its receiving use does not need. A claim that asserts Work must still establish the facts required for that Work.
For a broad language or ontology rule, use the predecessor and proposed versions of the actual host to judge DraftingActionability, LexicalAndNamingClosure, and precision restoration. A promise in the DRR, complete table, test in another lane, or invented fact pack cannot establish how a practitioner can use that host. Require both formal precision and plain comprehensibility. For a claim about other uses, establish why differences in the objects, the reader's work or the alternatives leave the claimed conclusion warranted.
Local names and kind settlement
The names below are for a complete reusable result. Ordinary bounded review can proceed without them. In a reusable result, each name denotes the existing FPF object or reference described in its row. The names confer no authority and require no record on their own.
These names identify local evaluation positions and references. Release or review status, project evidence, gate results, assurance, work, publication, and pattern-quality values are separate from those positions and references. Establish each on its own grounds.
Optional exact evaluation application, Work, result, and record
An optional DRRDecisionAdequacyRecordRef can collect references to the configuration, result episteme, witnesses, evidence-use relations, publication, and currentness. Include the semantic Method and assessment application when used, and Work admission when Work is asserted. Establish the application and Work through their respective conditions. Coordinate claims, evidence use, assurance, F.10 status use, acceptance, and downstream authorization each require their own basis.
If reliance depends on a viewpoint or grounding claim, give the basis for that claim. The evaluator's identity, the record's packaging, and a source label cannot establish it on their own.
E.22 may frame whether the evaluation is floor-only, exceptional-improvement, trade-off, open-question, absorption, or proposal-producing and may raise a floor before evidence is judged. Keep the required authoring use fixed after evidence appears. E.23 governs later repeated improvement work on the checked DRR after result claims or findings exist.
Ordinal coordinate scale and effective floor
When a complete reusable coordinate evaluation is produced, each ordinal value is a content-evaluation claim about the exact checked DRR under the declared use and window. In an ordinary bounded review, the same scale may guide judgement without materializing a measure, Work record, result episteme, assurance, acceptance, or reward.
Every required E.9.DA coordinate has a default floor of 3 sufficientlyExpressedForDeclaredUse. The review request, an accepted decision, or an E.22 question frame fixes the required authoring use. Use the E.9.DA default unless a source declares a higher floor for all coordinates or for named coordinates before the evidence is judged. Record the source of each raised floor.
Once the evaluator has seen the result, they may not lower a floor, narrow the use or selected loci, or shorten the qualification window to make the DRR admissible. A diagnostic evaluation with a lower target may report values. It cannot yield admissibleForDeclaredAuthoringUse.
Required decision-adequacy coordinates
Before assigning values, run the bounded omitted-question search from §4. Ask whether the DRR omitted an important question that would change its answer, content architecture, source use, consumer obligation, first drafting action, or stop. Inspect only enough of the governed problem, problem-owning practice, current sources, strongest live alternative, failure and recovery cases, and direct consumers to answer that question.
If the search finds an important omitted question, state it as a substantive finding before closing the coordinates. Reassess each affected decision, architecture, source-use, actionability and boundary coordinate under its value rule, keeping the required use fixed. When the search finds no such question, an ordinary clean review needs no separate ledger. A reusable result records the checked basis and clean disposition its named reliance needs.
For a framework-architecture decision, recover the practice field and the full use the initiating problem and sources require before evaluating the proposed owner. Inspect the decisive E.4.PFAD comparison, including any current framework boundaries that the new material calls into question.
A correct list of suppliers, a suitable carrier, closed alternative rows, or successful use of a newly written application can each support a narrower conclusion. None by itself establishes that the previous architecture already supplies the whole. Preserve each result at its demonstrated scope. Leave the architecture conclusion open when the required field or redistribution comparison is absent.
Distinguish coordinates by the repair question each answers. One DRR section may support several coordinates. State which property it supports for each. When two coordinates always fail and require repair together, use A.19.ECS to repair the characteristic space in the DRR or the evaluation pattern.
Result-row discipline and calibration
Use the question-to-ground and result-acceptance operation in E.22:4.3 within this evaluation.
A complete reusable result must let the reader recover every coordinate, its value or unestablished state, the applicable adjacent-value rationale, and the evidence locus. Choose a form that suits the receiving use and preserves any required serialization. The table shows one sufficient form.
For values 1..4, explain why the lower adjacent value would understate the evidence and the higher adjacent value would overstate it. For 0, explain why 1 would overstate the evidence and what would raise the value or reopen it. For 5, explain why 4 would understate the evidence and what would lower the value or reopen it.
A reusable coordinate result is incomplete if any required coordinate, its value or unestablished state, the applicable rationale, or its evidence locus is missing. An ordinary bounded review is sufficient when it clearly identifies the checked DRR, the required use and its source, the effective floor, a substantive finding, repaired text or a clean unchanged result, the first action, and the stop or reopen condition.
When required evidence is missing or unchecked, identify the gap. Leave the affected value unestablished until the needed check is complete. Assign a lower value only if checked content establishes the inadequacy in that coordinate's rule. For example, the checked DRR may lack a required decision. If an omitted question changes the answer, reassess each dependent coordinate under its own rule. The question adds no coordinate, and strength on another coordinate cannot compensate.
Common calibration points:
Local result status and stop condition
Each status below applies to the DRR under review, required use, effective floor, and qualification window. An ordinary bounded review may state it. Create a C.2.1 aggregate result episteme only when a named later reliance needs that reusable object. A local status neither decides a review nor establishes a gate result, permission, assurance level, or work state.
Derive status without compensation across coordinates. An unresolved architecture blocker yields holdForArchitectureDecision regardless of the other values. A required split yields splitDecisionRequired, and an easier or different use yields newFrameRequired. Otherwise, any required coordinate below its effective floor yields repairBeforeDrafting. State the required-use source, effective floor, and floor source in either result form so another evaluator can reproduce the conclusion.
When evidence needed for a coordinate is missing or unchecked, that value remains unestablished. Complete the check before drawing a conclusion that depends on that value. An unestablished required value cannot support admissibleForDeclaredAuthoringUse. An independently established architecture blocker, required split, or below-floor finding remains usable.
A result carrying admissibleForDeclaredAuthoringUse states the first drafting action, stop or return, and reopen condition. A non-ready result states the first repair, split boundary, or architecture question. Add a non-admissible reading only when an independent local ground makes it plausible and action-changing; any later gate or authorization uses a separately defined receiving relation.
Compact result form
Keep an ordinary bounded result short. One possible form follows.
This form suffices when no named later use needs a reusable coordinate result. A clean focused review may point to the unchanged DRR without a separate clean ledger for the bounded omitted-question search. After an inspect–repair–verify pass, point to the repaired text and its focused verification.
Extend the bounded result when a complete reusable coordinate evaluation is explicitly required or a named later reliance needs exact result identity. Keep the bounded result's content in the extended form.
Only the reliance-bearing branch requires every coordinate and the exact identities its receiving use needs. Method, application, and Work remain independently conditional. A claim of dated U.Work requires the enacted Method, every precise performer's A.13 core, and independent A.15.1 admission. Include an A.6.1 application only if the receiving claim uses that separately established application. Add F.6 only for precise assignment-bound attribution. Downstream status use, assurance, E.19 admission, authority, and drafting permission require their own direct grounds.
Structured finding row when reuse needs it
Use this row when the finding must be reusable in a structured form. In an ordinary bounded review, you may put the same precise diagnosis and repair direction in the handoff or repaired text. Labels such as weak DRR, needs more evidence, or architecture unclear leave the repair unspecified in either form. Record each independently repairable defect once and identify all affected coordinates or statuses, preserving their distinct values and rationales.
When E.22, E.23, absorption, or exceptional-improvement framing requests improvement, below-floor claims support finding rows and subsequent repair. Use a proposal row for an above-floor change only when it offers a substantive, non-dominated improvement within the declared authoring use. It might settle the answer more decisively, change the adopted source payload, assign a selected-locus obligation, decide an architecture split or merge, close a rejected alternative, specify a first drafting action or regression case, or remove or relocate apparatus that would otherwise enter the pattern prose.
A value below 5 is not in itself a defect. Stopping at 4 requires loci showing why a further change to decision content is dominated, unavailable, or outside scope.
Worked slices
Ordinary case — an overbroad wording DRR. DRR-ROLE-TRIGGER-03@2 selects this answer for an E.10 amendment and its named consumers: “replace every bare role with system role.” The review request fixes the use as authoring input for that amendment. The E.9.DA default floor is 3.
Before assigning coordinates, the evaluator checks the wording problem, system-role practice, the strongest meaning-recovery alternative, misuse and recovery cases, and the predecessor text of the affected A.6.5 consumer. A.6.5:1 and A.6.5:4.1 expose the omitted question. Does each occurrence denote a local system-role kind, or does it denote a relation-participant meaning, a declaration-local slot, an assignment, or an ordinary non-terminological use? The answer changes the proposed rule.
Return repairBeforeDrafting. The blanket replacement collapses distinct objects and uses into one term. First repair the DRR so it selects recovery of the intended meaning. Write system role for the local kind. Where the text denotes an assignment, participant meaning, SlotKind, or ordinary use, name that meaning. Reopen when the revised DRR distinguishes those uses, names the consumers that depend on the change, and the actual-host replay preserves precision and readability. The compact result suffices here.
Reliance-bearing extension. Suppose a named later comparison needs one stable result episteme, E9DA-Result-ROLE-03@1, but no Method, application, or Work identity. The complete result would still cover every coordinate; the following one-row excerpt shows the level of detail and why this value falls below the floor:
The effective floor is 3, so this row contributes to repairBeforeDrafting; another high coordinate cannot compensate. The result's bounded search basis names the wording problem, actual host, strongest meaning-recovery alternative, and the omitted question above. This comparison needs no Method, application, or Work reference. Include witness and evidence-use references only if it relies on them.
Adequate multi-locus DRR. The DRR selects the precision-restoration move, assigns positive responsibilities to selected loci, states the first drafting action, carries source payload into examples and conformance, and closes only the alternatives needed to explain the answer. The reviewer can judge it from that content and the actual-host replay. A reliance-bearing evaluation may additionally publish coordinate values and an aggregate result, but the ordinary judgement does not wait for that apparatus.
Architecture-impact DRR. A checked DRR cites diagrams, graphs, dashboards, or architecture notes. Review the exact DRR against the relevant E.9.DA questions: did it settle the architecture or structure claim, structural-view relation, preserved and lost structure, missing-structure return condition or source-use relation, selected loci, and publication boundary? A description locates material; it is neither the architecture, the decision, nor the assessment result. Open exact Method, application, Work, and result identities only when the reliance uses them; do not open one because another is present.
Bias annotation
This pattern favors settling decisions before drafting. A missing decision costs more to repair once several pattern authors depend on it.
A small editorial decision can use E.9 directly. In an ordinary DRR review, judge the decision, perform the bounded omitted-question search, and identify the first action. This review needs no separate assessment record. When a named later reliance needs an exact reusable result, Method, application, and Work remain independently conditional. If Work is asserted, preserve its evaluator System, enacted Method, and independently established admission basis. An application, attribution, result, evidence use, or receiving status enters through its own claim and grounds. E.21 covers pattern quality, E.23 repeated improvement, and F.19 wording repair, with E.10 for cues and unresolved meanings.
Conformance checklist
Common anti-patterns and repairs
Consequences
Rationale
Settle a missing FPF decision in the DRR before distribution so several later authors do not have to make the same choice. For an ordinary review, judge the decision in the DRR using the bounded search for an omitted question and any actual-host replay the decision requires.
A separately requested reusable evaluation or named later reliance can require a complete coordinate result and exact evaluation or result identities. These two review forms let authors keep decision records as concise as the decisions allow and supply the evaluation detail that reliance needs. In either form, adequacy depends on the content of the decision.
SoTA-Echoing and source use
The current-best source set spans empirical decision-document use, rationale-completeness risk, multi-dimensional evaluation, and proxy optimization. ISO/IEC/IEEE 42010 is used as a narrow current-standard reference; the feedback, MCDA, Pareto, Goodhart, and Campbell traditions remain lineage where the current rows or neighbouring FPF patterns carry the present answer. A new publication date, official status, citation count, or popular template does not by itself reopen any row.
Relations
E.9.DA:End
Unified Lexical Rules for FPF
Type: Part E lexical-governance pattern Status: Stable Normativity: Definitional pattern; normative for text that claims FPF-governed wording use.
Reading path. F.19 owns the common whole-span precise-language repair; E.10:0.2 supplies compact cues and routes only an unresolved FPF wording use. E.10.D1, E.10.LRN, E.10.DEV, E.10.ROLE, E.10.MOVE, E.10.ARCH, the precision-restoration patterns, and F.18 retain their subject work. Sections 5–9 below are optional register, naming, morphology, and overloaded-head references, not a second general language method.
Builds on: A.7 Strict Distinction: Repair Conflations of FPF Objects (Clarity Lattice); E.5 Guard-Rails (DevOps Lexical Firewall; Notational Independence; Unidirectional Dependency); F.5 Name U-kinds, Local System-Role Kinds and SystemRoleKindDescriptions.
Coordinates with. E.10.D1 for action-changing uses of context; E.10.ROLE, A.2, and A.2.1 for system-role kinds and assignments; A.15 and F.6 for Method, Work, and performed-Work attribution; A.10 for evidence use; B.1 and B.3 for Γ‑algebras and assurance; and F.17 with F.9 for source-local meaning and Bridges.
Use this when
Use E.10 when a word, head, or short phrase in FPF-governed text still hides its kind, register, morphology, reusable name, or exact FPF relation after the surrounding sentence has been read in ordinary language.
What goes wrong if missed. An author replaces one broad word with another or imports a technical-looking classification without recovering the object, relation, or use that made the wording matter.
What this buys. E.10 supplies cheap cues and the shortest exact route to an existing FPF rule. It does not turn ordinary prose into a lexical record or make every candidate word open an ontology exercise.
First useful move. Read the complete natural span with F.19. Identify the work the reader needs to do with these words. When ordinary wording conveys the intended claim and leaves that work possible, keep or repair it and stop. Use the cue surface below only to locate an unresolved FPF wording use; then open the smallest applicable lexical or subject pattern.
Not this pattern when. Use F.19 for general plain-language repair, missing operands, predicate compatibility, grounded contrasts, referents, coordination, list load, and information order. Use the subject pattern directly when the object and the rule or operation needed from it are already clear. For non-FPF source prose, use C.2.P source-expression unpacking mode and borrow an E.10 cue only when preparing a possible FPF use.
A cue is neither a ban nor a verdict. Its absence is not semantic clearance, and its presence is not an instruction to formalize the sentence.
One-screen cue and route
- Read the complete sentence, row or passage with the request it answers. What must the reader identify, explain, decide or do? What in the text is an example, a fixed condition or a choice still entrusted to the reader?
- Apply the connected
F.19reading. If ordinary meaning settles the issue, write the repaired text and reread the changed sentence with only its meaning-dependent neighbors. - If an FPF wording question remains, name that one question: head or kind, register, morphology, durable name, direct relation and participant, reusable declaration, claim or report, representation, source or publication use, or another exact governed object.
- Use the applicable subject pattern directly when the object and claim are already recoverable. Use the detailed
E.10rows only for the selected lexical question. OpenE.10.ARCHwhen the distinction among a world-side fact, reusable declaration, claim or report, and representation is unresolved. - Try the replacement in the reader’s work. For a claim about composition, what whole does the word pick out and which relation to its parts matters? For a comparison of models, which alternatives must the reader distinguish for the same question? Use C.39.SA:4 if the range still needs choosing. Use F.19:4.3.1.1 to return a result about another subject, scale or model to the original task. State a conclusion common to the supported readings when it settles the request, and identify any difference that still matters.
“Why did this review pass the submission?” may ask about the decision in one meeting. “Why do our reviews keep passing these submissions?” calls for the way the review practice produces that result. The minutes may answer the first question. They can also supply grounds for the second if they establish the operative rule, how it is used and how far it applies. Choose the subject from the question before sharpening the word review.
A complete reading can preserve clear conversational wording or an exact locally defined meaning. Where something remains unresolved, distinguish a sentence repair under an existing rule, a needed distinction or relation for which the defining rule is absent, and facts still missing to decide the meaning or claim. Unavailable facts do not establish an ontology gap. An applicable pattern supplies its stated rule; its presence does not prove that every encountered case is covered.
The ordinary result is retained or repaired text, or a specific blocker. E.10 requires no per-correction card, field set, concordance row, or classification receipt. A using environment may manage attention over a large text, but that mechanism is outside this language pattern.
Common prose can often remain ordinary after the relation is stated: Test T is evidence for claim C; Index I helps readers find section S; Column C bears roof R's load. These sentences reach different subject patterns and do not need a common SupportRelation.
Local patterns may cite a useful recognition row. They do not copy the trigger inventory or create a local lexical registry unless a stable local vocabulary is the governed product. Self-application remains bounded: use E.21 for pattern-quality evaluation and the exact subject pattern for any non-lexical claim.
Scope split
E.10 applies to lexical conformance in FPF pattern text, extracted pattern hosts, FPF-Spec monolith text, FPF governing documents, accepted DRR text, and any project, product, research, engineering, or review text that deliberately uses FPF terms, pattern references, FPF relation names, FPF kind claims, FPF admissibility claims, or claims FPF conformance.
For ordinary source text, intake notes, seminar transcripts, external reviews, project documents, source publications, tool outputs, or other text that does not claim FPF-governed use, use C.2.P source-expression unpacking mode. That use may borrow tests or rules from E.10, A.6.P, A.6.6, F.18, or another relevant pattern, but it does not judge the source text as failed FPF wording.
Compact cue and routing surface
Keep these cues available while applying F.19. They help find candidate spans; they do not replace the whole-span judgement.
Semantic-area cues: classify FPF-governed trigger wording before acceptance by semantic area, not by a local forbidden-word list. Typical classes include admissibility/deontic terms, evidence and review-check terms, action-invitation terms, characteristic/scale and stratification source labels, state-family terms, lifecycle/process terms, pattern-application wording, publication-form terms, and local equivalents.
Package-form and relation cues: primary carrier, specialization, profile, overlay, family, bundle, cluster, suite, pack, kit, record, umbrella, and local equivalents.
Item count is only a cue. Two coordinated members may already form a needless catalogue; a long legal set, signature, inventory, or checklist may be required. Matching kinds do not earn a series by themselves: the receiving use must require readers to distinguish or retain the members together.
exact is not a general precision marker. Keep it for literal or bounded source identity, or when it distinguishes a live alternative that changes truth, action, publication, reuse, or reliance. An ordinary PatternID citation needs no formal expansion merely because it is exact.
Routing boundary
If the applicable pattern and the current object, direct relation and participants, declaration, claim-bearing episteme, representation and correspondence, or source use are already recoverable, use that pattern directly and state its concrete contribution—for example, a definition, constraint, test, method, or lookup.
The result is the repaired wording, the concrete result of the selected pattern, or the blocker. Do not require a rejected neighboring interpretation, a field schema, or a complete catalogue of possible routes when the current sentence needs only one.
Optional trigger index
The compact surface above is the normal entry. The detailed rules in E.10:0.2c and the L-rules in section 9 are optional references after the F.19 reading leaves an FPF wording question unresolved and a cue identifies the relevant rule.
Frequent search terms include broad heads (support, basis, context, role, method, process, result, record), state and authority-looking words (current, ready, valid, approved, required), source and representation words (source, publication, carrier, field, map, dashboard), and grouping or coverage language (and, plus, slashes, examples, variants, forms). They are examples, not a closed lexicon. Search can increase recall; only the natural-span reading can decide whether a defect exists.
Use a detailed row only when its distinction changes the repair. Otherwise write the ordinary sentence and return to the domain task.
requirement and required recovery
Treat requirement and required as cues, not as a shared engineering kind or durable suffix. Read the complete span through F.19; when an ordinary sentence can state the condition or action directly, repair it and stop.
When the wording still carries an FPF-governed claim, recover the bearer, the exact claim or relation kind, the governing pattern when its identity changes use, and the practical consequence. Write that construction directly and reread the changed sentence. The result is the repaired text or a blocker, not a separate record.
Name admission precedes slot verification. CandidatePatternUseBasisRelation@Context is admissible because the head exposes the basis relation and its two sides; CandidatePatternUseBinding is not repaired by kind-correct fields because Binding hides that subject relation. Likewise, BoundaryConditionKindSlot names a slot whose values classify boundary conditions; BoundaryRoleSlot would falsely suggest a role value. Apply F.18 only when a durable name is being minted.
External holon and graph-expression source wording
Cue. External holon-class or Holon Graph Architecture (HGA) graph-expression wording such as AgentHolon, OrganisationHolon, DataHolon, ProcessHolon, Portal, Projection, event envelope, provenance, target holon, projection envelope, projected content, envelope, payload, RDF graph, node, edge, traversal, or boundary-governed payload whose FPF object is hidden.
Recover the claim before importing the source label. Use A.1 for admitted system or holon claims; C.2.1, E.17, architecture-description, publication, source-relation, or evidence patterns for data, document, projected content, description, publication, view, or evidence claims; A.10, source-relation, evidence-relation, dated-work, or publication patterns for event and provenance claims; and A.3.4.P, method, work-plan, or Work patterns for process-like wording.
Portal, access, traversal, service-access, protocol, and agreement-like words do not select peer routers. While their object remains hidden, use A.6.RSIR; after recovery, use E.17 or the pattern for the publication claim, the pattern for the representation and correspondence, with C.29 added only for mathematical-lens use, A.6.0 plus A.6.5 for a reusable signature and its slots, A.6.M for a module-interface claim, the pattern for the policy claim, and A.10 or the pattern for the evidence claim.
Use A.6.P:4.11a only when a relied-on service or access phrase hides its concrete subject or direct relation; use A.6.C only when recovered service-term, SLA, protocol, or agreement-like wording bundles promise, utterance or publication, governance, Work or consequence, or evidence claims; use general A.6.P only for another under-specified direct relation.
For graph, RDF, node, edge, or traversal expression claims, use the pattern for the recovered subject, representation, source or publication relation; A.22, C.30.ASV, C.30.AD, and E.17 retain their questions, and C.29 applies when mathematical correspondence or returned inference is claimed; use A.6.B only for L, A, D, or E statement classification inside a boundary package.
W3C Community Group Holon Graph Architecture (HGA) vocabulary is retained as a source-finding cue or comparison term only after the recovered FPF object is named and differences from FPF are explicit. Do not mint source-class U-kinds such as U.AgentHolon, U.DataHolon, U.ProcessHolon, U.Portal, U.Projection, U.Envelope, or U.Payload; do not turn semantic-web class names or graph-expression vocabulary into FPF ontology.
Lexical Trigger Rewrite Rules
EntityOfConcern, primary entity of concern, and local topic wording
Do not replace every topic-like or object-like phrase with EntityOfConcern.
Classify the sentence first.
Use the selected row only when the EntityOfConcern distinction changes the sentence. State the recovered participant, publication-unit use, review target, or ordinary topic directly; preserve any claim-bearing episteme, declaration, representation, source wording, and reader use that remains current. The ordinary result is the repaired sentence or a blocker, followed by the local F.19 reread.
publication-unit wording that implies authoring or interpretation work
When a phrase makes the bounded unit sound like authoring work or interpretation work, split the sentence by kind under repair.
Apply E.17.0:4.3–4.5 when claiming that an episteme is a U.View: the same episteme must conform to an identified viewpoint edition. Publication-face membership alone supplies no such conformance.
Do not make a permanent technical modifier by joining authoring, interpretation, and unit-boundary concerns. That mix hides whether the sentence is about a publication unit, authoring work, reader inspection, or a carried claim.
content
Do not use content as a catch-all head.
Split it into:
- claim-bearing episteme content;
- publication-unit text;
- publication form;
- generic publication face;
- declared MVPK face;
- carrier data;
- payload of a record kind named by the pattern that defines that record;
- pattern section;
- source-basis excerpt;
- review target.
Plain explanatory prose may use content only when the sentence does not carry ontology, authority, or admissibility.
publication
Every FPF-governed publication sentence names the publication construction being used:
- act or occurrence of publishing, or publishing work;
- selected
U.Epistemeplus the exactEpistemePublicationRelationoccurrence when availability to a declared audience for a bounded use is current; - publication form;
- generic publication face;
- declared MVPK face;
PublicationUnit;- carrier or rendering;
- document with named source-basis, evidence-basis, architecture-basis, or review-basis relation or use;
- external-standard publication;
- project record publication.
If the sentence says a publication "supports", "authorizes", "proves", "permits", or "makes admissible" something, make the basis explicit: state the relation when a relation claim is being made, state the admissible-use boundary when a boundary-use claim is being made, and name the exact project-side kind and reference when project-side records, evidence or provenance relations, gate decisions, constraint or adjudication decisions, assurance records, work, action invitations, speech acts, commitments, methods, or carriers are being used. Record those values in relationClaimSlice, admissibleUse, and projectSideFPFRef, respectively, only when the receiving claim needs the named fields.
surface, view, face
source, target
These are relation words, not final kinds.
Recover the relation or use that makes something a source; common cases include a selected source U.Episteme, exact EpistemePublicationRelation occurrence or reference when availability is material, publication form or carrier when either is the source object, U.View over a source U.Episteme, document with named source-basis, evidence-basis, architecture-basis, or review-basis relation or use, independently established evidence-use relation recovered through A.10, authority-reference relation, named FPF pattern cited as source, file carrier, exact source edition, source-local meaning under an effective scheme, source frame only when its defining pattern and endpoint kind are present, the actual source-side participant of an obtaining named relation, the source-side A.6.5 SlotSpec only when a reusable relation declaration is current, or project-side FPF kind and reference named by value.
Recover the relation or use that makes something a target; common cases include an EntityOfConcern, target U.Episteme, review target, FPF rule being applied, project target, work target, target publication form, project-side FPF kind and reference named by value, target frame, target-side situation or model-use boundary only when its defining rule is current, the actual target-side participant of an obtaining named relation, or the target-side A.6.5 SlotSpec only when a reusable relation declaration is current.
Generic object and target are not final recovered kinds. Keep them only when the sentence explicitly declares a named field or participant designation inside a current episteme, such as ObjectKindUnderImprovement, ObjectVersionUnderImprovement, or ObjectVersionUnderQualityEvaluation; names a review target; or states one direct-relation participant meaning whose actual-participant kind is supplied by value nearby. If a reusable relation declaration is current, name its A.6.5 SlotSpec separately.
When the governed kind is known, write that kind by value: FPF pattern version, DRR, FPF corpus slice, publication form, PublicationUnit, file carrier, system carrier, exact changed referent, exact entity or value bound as the result of one A.6.1 operation application, candidate proposal, evidence or provenance relation, gate decision, work plan, method description, object-under-improvement evaluation, or another named FPF kind.
Do not recover an FPF pattern, publication form, PublicationUnit, pattern body, or view as a carrier merely from its label. Under E.17 and E.24.PUB, identify the physical or digital carrier independently under its domain identity rule and establish that its actual state or realization yields the particular expression. The reusable publication form supplies rules that expression follows; it is not the filled expression or its carrier.
Ordinary carrier wording elsewhere names its actual bearer relation to a system, medium, file, rendering, front end or transport object. If the text means the FPF pattern publication form, write FPF pattern publication form; if it means the file, rendered, front-end or transport side, name that exact carrier or relation.
Common repair examples:
Do not publish "source and target" if the selected relation needs the actual FPF kind.
input, raw material, source data, source material, artifact, output, result, outcome, deliverable, work product
These source words depend on the relation in the sentence; they do not identify a final kind or one result family. First name the exact entity and the exact object relative to which the word is being used. For epistemic source data or source material, close the exact source expression, episteme or publication, and source-to-use relation under C.2.P first. For physical raw material, name the relevant constituent, affected-referent, resource-use, supply, transfer, or transformation relation and use the pattern that defines it.
When the remaining current claim is relative to a method, plan, dated work, transformation, evaluation, delivery, transfer, or receiving use, apply A.6.P.WMR. Recover claim subject, modality and exact temporal extent, polarity, and recovery/support state independently. For closure, give the precise result and select exactly one of four families: a direct subject-relation claim, positive or governed negative; an A.6.1 operation-application binding; a local A.15.PROD or A.6.RCD claim; or a non-assertability result with the reason and continuation defined in A.6.P.WMR:4.3.
Failed tests, unavailable facts and known inapplicability retain their governing rule. A missing subject rule and missing constructor semantics keep their different repair questions; a future pattern or declaration need is named only when identifiable.
Classification, a generic result relation, a method-description field, planned filling, a designation that merely type-checks against an A.6.5 SlotSpec, or a polarity inference is not closure.
Before opening that branch, test whether the phrase already names an independently identified U.Episteme; U.View; publication form; publication face, including a declared MVPK face; PublicationUnit; carrier, front-end, or rendering relation; project-side FPF kind and reference named by value; evidence carrier or evidence relation; document under a named source-basis, evidence-basis, architecture-basis, or review-basis relation or use; review target; C.11 ChoiceResult; measurement-result episteme; evaluation result; diagnostic finding; decision; or another project object whose record kind and defining rule are named by value.
Retain ordinary input, output, result, outcome, or deliverable only while the exact defining relation rule remains recoverable. If no governor closes the selected WMR claim, return the bounded blocker. If the missing item is instead a non-WMR kind, retain an architecture-first candidate disposition under the pattern that defines that candidate. Do not invent either one inside pattern prose or replace it with a universal kind or relation.
For work product, name the thing and the claim being made about it: a corrected recording, a report's content, an assembled device, or another entity. A.15.PROD distinguishes participation in production, the first satisfaction of an entity's identity rule, and completion under a criterion. Apply only the question raised. A document can describe a wanted change without being that change; a publication occurrence, delivery and acceptance each need their own basis. A.15.2:4.2 keeps an expected output or delivery target in the plan from asserting its existence or completion.
For a recurring result of Work or result of a project, reuse the established meaning and obtain the facts for the new participants. A.6.P.WMR:4.7 explains that reuse, including production across entity kinds, preservation of a condition, joint Work and later use; :5.7 develops the authoring case and a missing-fact return. Different participant kinds do not by themselves require different result meanings, and the common word does not make those different relations one. A.15.6 supplies the actual-project and whole/part grounds when that claim is current.
goal, purpose, objective, and the concern of a method
Read these words with the action or claim they qualify. Whose aim is being stated, what is it about, and what change or preserved condition matters for the decision? A team's chosen goal can explain its choice of a Method. The same Method can serve several purposes. Keep a clear ordinary use without requiring a formal goal object.
For a Method, A.3.1:4.1 and :4.4 distinguish what future enactments concern from what successful enactment is meant to achieve or preserve. Observing a recording's timing, correcting it and judging its fitness for movement analysis concern related material but answer different questions. State the reusable action, intended effect and relevant limits. A Method concern is not automatically an actual participant, an object physically changed, a declaration-local argument or result slot, or a field to be filled in a WorkPlan. Use the exact relation or declaration only when that further claim is needed.
For example, a team wants movement analysis with timing error at most 3 ms. Its proposed preparation Method is concerned with correcting the recording's relative timing while preserving the recorded movements. The plan names the recording to prepare and the intended delivery. These sentences say what is wanted and how the team proposes to obtain it. Whether a particular recording meets the limit, was used in analysis, or improved that analysis requires the corresponding facts and rule; none follows from calling all of them “the result”.
The purpose of a guide likewise states the use it is intended to help its readers perform. Its explanation, the published file, a reader's use and any resulting improvement remain distinct claims when the decision depends on them. Use §0.2c.7 for an unresolved result or work-product claim, §0.2c.13a for an unresolved contribution claim, and the subject pattern directly once the meaning is clear.
record
Use record only when an FPF pattern or project practice defines the record kind and relation. The nearby wording says which FPF kind the record instantiates or records, for example:
- an A.10 descriptive evidence-provenance account for a named claim, citing independently established relations;
- A.21
GateDecisionResultorDecisionLogRef; A.20constraint or adjudication decision record;C.11ChoiceResultor decision record;A.15U.WorkPlan, oneA.15.1dated Work occurrence admitted underU.Work, or a separately identified claim-bearing episteme about that occurrence; use a record-kind name only when its exact kind and defining record rule are recoverable;A.2.8 U.Commitment, exactA.2.8.PERpermission result, orA.2.9 SpeechActpublication;- a separately identified assignment-assertion or occurrence-description episteme that designates one exact
RA : U.SystemRoleAssignment, or a status-register entry under its named status pattern; neither record is the world-side assignment occurrence; E.19review run record or another named review record whose review target and review relation are explicit;- process run record in process documents.
Do not let record mean "any file that remembers something", "the missing source", or "the thing to create when support is absent". If a named support relation cannot be asserted because a required actual participant or value is absent, name that missing participant or value.
If a reusable declaration is incomplete, name its missing SlotSpec or other missing declaration content and repair that declaration. If a receiving assertion or relation-occurrence-description episteme lacks a participant designation, name the missing designation under that episteme. If a U.WorkPlan lacks a planned participant designation or planned value, name that missing plan content under the WorkPlan.
Create a prospective repair request, future decision request, prospective work-plan entry, or explicit missing-source-relation note as applicable; none backdates support, establishes actual participation, or makes the direct relation obtain.
model, diagram, screen, dashboard, table, note, memo, summary, explanation
These words cue recovery; alone they do not identify the FPF kind.
For unresolved model or explanation wording, recover the subject at issue and what the work needs to obtain or change. An explanation of a predictor may concern its behavior, its internal computation or the phenomenon it predicts. Constructing an account, expressing an available account and helping another agent acquire an ability can require different methods; one request may need more than one of these actions. Use the method that supplies the required result once the question is clear. If the object or relation remains unresolved, use E.10.ARCH:3.
When model hides the question, state what answer is sought from it. For example, a model of stated axioms is a structure satisfying those axioms; modeling an observed process may seek relations from which to derive predictions or compare interventions. The same mathematical construction can serve several modeling uses. Retain the subject's vocabulary and notation when they already make that use clear.
For an unclear request to "run" or "execute" a model, recover how the model is interpreted for this question, how the answer is obtained, and who or what performs that operation. The operation may be reasoning with a diagram, solving constraints, running a simulation or experimenting with a physical analogue. A diagram's reader can supply premises and an inference method that are absent from the drawing; recover the missing premise or inference operation when it changes the answer or another agent must reproduce the reasoning. B.5.FM helps construct the model, A.6.3.RT.OE makes an expression usable in the intended operation, and C.29.2 develops a computational formulation when computation is required. Use C.2.P.DR when the wording attributes an unsupported action to the representation.
A performance claim states what the agent does or can do. A claim that it uses a particular internal representation concerns how that performance is obtained. Use C.2.1 when the content and identity of an episteme are at issue; use the applicable subject method to establish a claimed physical or computational representation. Use E.17.EFP when ambiguity about source faithfulness changes how an expressed account may be used, and C.2.8 when recoverability by a recipient is the question.
When the wording concerns representation, publication or reliance, classify the occurrence as applicable:
- episteme or episteme publication;
U.View;- publication form;
- generic publication face;
- declared MVPK face;
PublicationUnit;- carrier, front-end, or rendering;
- project-side FPF kind and reference named by value;
- explanation and source-finding relation under
E.17.EFP; - an A.10 descriptive account of evidence, currentness and provenance, with each relation independently established;
- gate-bearing claim or effect under
A.20orA.21; - assurance and engineering-justification record under
B.3; - work- or reliance-guiding appearance whose missing prerequisite is recovered under
A.15.4.
Use the ordinary word once the surrounding text makes the intended object and use clear. Add an FPF kind name when that distinction changes the claim or the next action.
reader, reviewer, author, operator
Do not use people-position words as hidden kind names.
Use:
working readerorintended practitionerfor ordinary usability;engineer-managerwhen the FPF use case is the engineer-manager applying the pattern in work;revieweronly for a participant in a named review relation; use review process, review gate, or review target for the process, gate, or object;authoronly for authoring or editing work;- operator only after the claim is recovered: it may name an admitted system classified under an exact local
OperatorSystemRole, a process or mathematical operator, an organizational or interface position, an ordinary title, or another direct relation; the word alone selects none;
If a text says "reader-facing" or "review-facing", it also names what is facing that person: generic publication face, declared MVPK face, packet, document with named source-basis, evidence-basis, architecture-basis, or review-basis relation or use, PublicationUnit, carrier, or UI or front-end.
owner, home, host, locus
Keep owner as ordinary architectural, organizational, policy, source-maintenance, or stewardship wording when the sentence already makes the precise ownership or responsibility relation and its participants clear. Otherwise name that direct relation. Owner does not mean a system-role kind, U.SystemRoleAssignment, authority, or Work by default, and it is not a substitute for pattern, DRR, selected U.Episteme, exact EpistemePublicationRelation occurrence, publication form, publication unit, file carrier, or project record.
Split the claim into the smallest applicable value:
- exact architectural, organizational, policy, source-maintenance, stewardship, ownership, or responsibility relation and participants;
- relation to an FPF pattern or an authority-reference relation;
- named source set used as authority for the claim;
- file carrying FPF pattern text, file carrier, or publication unit;
- evidence record or evidence source;
- FPF pattern or project target;
- support root; or
- ordinary or quoted non-use when owner is only title-like wording; otherwise use the result selected through
E.10.ROLE—for example, a local system-role kind and classification, a namedU.SystemRoleAssignmentoccurrence, or another relation.
route, branch, handoff, path, trajectory, move, flow
Recover the movement, control, and temporal relation set before using these words:
E.10.MOVEfor project-move, first-move, working-move, next-move, pattern-use, work-entry-readiness, architecture-candidate-use, call-planning next-action, or other move-like wording whose direct FPF target is hidden;A.16local move;A.16.0trajectory account;A.19,C.2.2aposition in characteristic space or state space;B.2.5for a two-sided supervisor-subholon feedback relation; use the direct subject pattern for any other control or control-layer relation andC.30.LCAfor a control-structure view;- process handoff;
- selector relation or selection mechanism;
- work transfer;
E.18graph path orPathSliceexpression;A.6.3,A.6.4episteme morphism or retargeting.
When handoff instead names an entity, package, result, delivery, transfer, acceptance, or receiving-use boundary, apply A.6.P.WMR to that exact relation-bearing claim. Use E.10.MOVE for a process-baton or project-move case only when that movement is current; a handoff record or package remains an episteme or another entity, not the transfer.
If no movement, control, and temporal relation is being made, keep the word ordinary and non-authorizing.
use, supported use, action, effect
These words are cues when the sentence hides who or what uses what, for which claim or action, and what changes as a result. State that ordinary sentence first. Pattern application can usually be written as Use P to ...; publication use, reliance, evidence, Work, gate, decision, and other governed uses go directly to their subject patterns once the object and relation are clear.
For a bounded or supported-use claim, name the admissible action and its basis. State an excluded adjacent or stronger use only when it satisfies the grounded-guard test in F.19:4. Do not turn a document into a generic capability or replace the sentence with an inventory of every nearby FPF kind.
Use C.2.P when a source or publication use remains hidden, A.6.P when a direct predicate or participant remains unclear, and E.10.ARCH when fact, declaration, report, and representation are still confused. Name relationClaimSlice or projectSideFPFRef only when the receiving claim consumes that identity.
contribution, contributes, what it adds
Use these words when the sentence makes clear what participates or is used, in which action or claim, and what is asserted about it. When that meaning is missing, restore the concrete statement before relying on the phrase. Contribution can remain an ordinary summary; the word supplies no common object kind or generic contributesTo relation.
A supplied recording, its use by an analyst, a claimed causal effect and a comparison of two configurations need different statements. Use A.6.P when a predicate or participant remains hidden, C.2.P for an unresolved source/content/use distinction, and C.28 for a causal claim. C.11.CRC does not replace a comparison of source content with no realizable configuration change.
Stop with a clear ordinary sentence when that answers the question. A.6.REL and A.6.RSIR apply only when relation occurrence or declaration distinctions matter; E.24 applies to an admission question. Replacing contribution with input, result or support while leaving the same uncertainty is no repair.
sign, concept, denotat, and school-semiotic labels
Do not import the school-semiotic triad as architecture ontology.
When a source or review text says sign, signifier, signified, concept, denotat, representamen, interpretant, or sign vehicle, apply the composite recovery order before the term appears in FPF-facing prose.
Possible recoveries include:
U.Epistemeor episteme species named by value;- selected
EntityOfConcern, grounding, reference-plane relation; U.View;- publication form, generic publication face, declared MVPK face, or
PublicationUnit; - carrier, front-end, or rendering;
- cue, displayed wording, mark, status display, credential display, provenance mark, signature evidence;
- evidence record, gate record, work-state record, commitment record, separate assertion or occurrence-description episteme about one exact
U.SystemRoleAssignment, or another project-side FPF kind and reference named by value; - FPF pattern, pattern section, accepted
DRR, FPF publication, or FPF view when the object is on the FPF side.
Use concept only where current FPF already has the relevant concept-set, UTS, local-meaning, or Part F machinery available.
Otherwise recover the claim-bearing episteme; the obtaining direct relation and actual participants; the current A.6.5 declaration, participant designation, or representation and explicit correspondence under their applicable pattern when one of those is present; or the record kind and defining rule named by value.
A representation need not use a mathematical lens. Recover its represented subject and correspondence under the direct pattern or domain rule. Use C.29 when choosing, transferring or relying on a mathematical representation leaves a separate correspondence or inference question; use A.6.3.RT for a claimed representation transition, E.24.PUB for expression and publication, and C.37 only when one receiving action needs a selection or co-use account.
pattern, generic FPF-side object wording, locus, row, target
Pattern is not a free synonym for regularity.
If the intended object is an FPF pattern, write FPF pattern or name the pattern and the definition, constraint, test or method being used.
If the subject is not an FPF pattern, name the concrete object or relation and the use asserted about it. Use §0.2a for an unresolved kind, source, publication, Work, authority or representation question; then return the recovered value to this sentence.
Avoid generic FPF-side object wording, generic named-target wording, locus, row, and host when they hide kind.
Use them only when the kind is literally a table row, document with named source-basis relation or use, file carrying FPF pattern text, or review target and the sentence does not need a narrower FPF kind.
For FPF-facing wording that carries a claim being made, direct relation, admissible use, or remaining reader use, these are candidate recoveries, not a group kind: FPF pattern, pattern section, accepted DRR, FPF publication, FPF view, record kind with its defining rule, obtaining direct relation and actual participants, receiver-needed relation occurrence, claim-bearing episteme, reusable A.6.5 declaration, or representation under its direct pattern and explicit correspondence. Choose one by sentence function and keep the different objects separate.
Union-field unpacking under A.6.P
Do not write authority-bearing FPF pattern, authority-bearing FPF row, FPF row named by value, selected FPF pattern, record, or relation, governing FPF relation, or required project record or action as final fields.
When one of these union-fields appears, make the A.6.P choice explicit:
- if the sentence is making a relation claim, recover the
RelationKindand actual participants, with the qualifiers, scope, time, viewpoint and admissibility target required by the direct relation or receiving use, then state that claim with those participants and its supported polarity or unresolved status; distinguish one relation occurrence only for a named receiving use; add a reusableRelationSignatureand A.6.5SlotSpecvalues only when declaration is current; and keep any claim-bearing row or field as an assertion episteme, participant designation, or representation under its direct pattern and explicit correspondence as its own assertion or representation rather than as the relation itself; - if the sentence is not making one relation claim, unpack the wording and claim under repair into an FPF-side kind, reference, or relation named by value and one project-side FPF kind with its reference, or state that no project-side FPF kind is triggered;
- if the same unpacking recurs across cases with one stable recovery shape, record a light A.6.P specialization candidate rather than minting a vocabulary-wide replacement field.
Apply this unpacking whenever a publication, display, cue, explanation, dashboard tile, schema, signature, badge, or generated output is being read as evidence, gate passage, work, deontic permission, work authorization, approval speech act, commitment, release authorization, safety assurance, evidence sufficiency, or engineering justification.
Do not fill one authoring union-field position with whichever nearby FPF kind is easiest to name. A project publication, claim-bearing episteme, or record with a named kind and defining rule is a description-side object; one A.15.1 dated Work occurrence admitted under U.Work is a world-side individual, while A.6.A action invitation, A.2.9 SpeechActRef, A.2.8 U.Commitment, U.Method, and U.MethodDescription belong to other kinds or relations.
Coordination and FPF-governed lists
Apply F.19:4.3, step 6, to recover the series' governing proposition, scope and membership semantics before choosing its form. When the series is unnecessary, preserve that content in the direct claim or relation; use cases for their contribution to its explanation.
When a series is needed and its FPF meaning remains unresolved, distinguish the current use:
- a closed set, with its classified kind, membership rule, and closure;
- illustrative examples, with their connection to the proposition or kind and a non-exhaustive cue when completeness or applicability is plausibly ambiguous;
- explicit alternatives or a sequence;
- several obtaining direct relations, each with its actual participants;
- a reusable relation declaration, with its
RelationSignatureand A.6.5SlotSpecvalues; or - a representation under its applicable pattern, with its elements, represented objects, and correspondences; use C.29 when the claim concerns a mathematical lens.
If it declares reusable relation shapes, name each RelationSignature and A.6.5 SlotSpec value and do not infer that any relation obtains. If it is a tuple or other representation, name the representation elements, represented objects, and explicit correspondences under its applicable rule, using C.29 when mathematical-lens use is current; if the same material is also a reusable relation declaration, name its RelationSignature and A.6.5 SlotSpec values separately.
If none fits, leave the ontology or architecture question blocking instead of making the list a new kind. Even when one case fits, keep only members whose distinction changes the receiving use.
strong, stronger, weak, weaker, support
Use strength wording only when the sentence makes the comparison dimension clear. Examples of more precise wording:
stronger claim-> wider claim scope, higher evidence-basis threshold, gate or admission threshold, claim requiring world-contact evidence or authority relation, authority claim, or named evidence-support class;weaker claim-> narrower claim scope, lower evidence-support class, bounded admissible act, work, or claim,source-loss modeunderA.6.3.CSCwhen a source-to-rendering loss is being claimed, coarsened rendering, or explicit abstain or reopen condition;
A named scale, evidence class, threshold, or CharacteristicSpace supplies the comparison when that is the dimension used.
Treat support as a cue to write the concrete subject, predicate, object, and use. If that sentence states a clear direct relation, use the pattern that defines or constrains it. If the predicate or a participant remains unclear, use A.6.P; if both are clear and no current pattern supplies the rule, use A.6.RCD. Use A.6.6 when the claim concerns basedness.
When the phrase carries a bounded-use claim, state the admissible action. Add a stronger or adjacent non-use only when it satisfies the grounded-guard test in F.19:4. A support-headed durable name reaches F.18 only after the governed object and use are settled; otherwise replace the head locally or leave the meaning blocked.
Applying patterns versus procedural calls
FPF patterns provide reusable guidance for recognizable problem situations. In ordinary prose, apply pattern P is acceptable metonymy for a person or system using the method, rule, test, constraint, or lookup described by P; the pattern does not perform the project action.
Ordinary application. Name the recognizable problem, state which rule, test or method the reader uses from the pattern, and state the resulting action or judgement. An ordinary PatternID citation is enough when the reader only needs to find that explanation. Do not require an ontology, conformance claim or section, exact assertion, ClaimGraph, or formal application record merely to use the guidance.
Identity-sensitive application. Open this branch only when a named live alternative or receiving use changes truth, action, stop, interpretation, migration, publication, reuse, or reliance. Name that dependency first, then add only the identity it needs: for example, the exact pattern edition, ontology, conformance claim or section, governed object, claim-bearing episteme, obtaining relation and participants, current declaration, or representation and correspondence.
Use apply pattern, use the pattern guidance, the pattern applies to this problem situation, or the case falls under this pattern for the ordinary FPF-side use. These expressions do not assert that the pattern acts.
For ordinary project-side activity, say who does what and which result or judgement follows. When the current claim or a named receiving use depends on a formal kind or relation, recover that value under its direct pattern and preserve its membership or obtaining conditions. Apply §0.2a only while that selection remains unresolved.
Keep route, path, branch, handoff, trajectory, move, or flow as ordinary navigation wording when no FPF movement, control, or temporal claim depends on it. When such a claim is current, name the relevant movement, control, and temporal relations and use the pattern that defines them.
Framework content and project-side subjects
Semioarchitecture distinguishes framework content and its use from the subjects and uses in a project:
- FPF-side content and uses:
FPFas episteme, FPF patterns, pattern sections,DRRs, FPF publications, FPF views, support documents and documents with named source-basis, evidence-basis, architecture-basis, or review-basis relations or uses, and review targets; - project-side subjects and uses: the engineer-manager's project epistemes, publications, views, records, carriers, cues, evidence records,
A.20constraint or adjudication decision records,A.21gate decisions, optional A.21DecisionLogrecords,B.3assurance or engineering-justification records, commitments, oneA.15.1dated Work occurrence admitted underU.Workplus any separate episteme about it,C.11ChoiceResultvalues,C.11decision records, andA.6.Aaction invitations.
Do not blur them with source, artifact, object, material, target, pattern, or broad semiosis.
If both sides are being used, split the sentence to make the relation explicit when a relation claim is being made, the admissible-use boundary when a boundary-use claim is being made, and the project-side FPF kind and reference named by value when that side is being used. Record the corresponding values in relationClaimSlice, admissibleUse, and projectSideFPFRef, respectively, only when the receiving claim needs those named fields.
For an unused side, record absence only when a named receiving use must distinguish it from missing information.
decision, action, work, method, plan
Do not let action cover every project-side event. An action nominal such as testing, assembly, maintenance, evaluation, or inspection is a morphology cue, not a kind. Placement in function- or flow-structure prose identifies no U.Function: apply A.6.F when the function-like use remains claim-bearing and its FPF object or relation is hidden; otherwise name the already recovered method, method description, required-transformation or required-effect claim, actual U.Transformation, TransformationFlowStructure locus, functional-view record, plan content, exact Work occurrence, or other value under the pattern that defines it. A WBS element, activity, or Work Package remains plan- or assignment-episteme content about intended work; none of these uses identifies an actual Work occurrence admitted under U.Work.
Split decision-making and decision records under C.11; local system-role-kind classification under A.2; U.SystemRoleAssignment occurrences under A.2.1; Method under A.3; planned work and WorkPlan under A.15.2; dated Work occurrences under A.15.1; actual launch or performed values under obtaining direct relations or A.6.1 bindings; separate performed-work, finalization, result, telemetry, and gate records under the patterns that define them; action invitation under A.6.A; communicative acts under A.2.9; commitments under A.2.8; and strong grants under A.2.8.PER. A method-description field, planned filling, compatible type, ticket, or nearby result record establishes no actual participant relation.
A reusable name for performed work goes to F.18 only after A.13 and A.15.1 independently admit the Work occurrence from its actual performer basis, time, Method, and containing System. If the reusable account also needs precise assignment-bound attribution, establish F.6 afterward through the same obtaining assignment. Keep the affected referent, application bindings, resource use, and other direct facts separately recoverable. Add a continuity policy only when occurrence identity matters. Keep production, measurement, evaluation, delivery, acceptance, and downstream-effect claims under their direct patterns.
P2W in E.18.1 carries accepted problem-side content into a question needed by the receiving work, then preserves that content in the result or stop obtained using the pattern that answers the question. E.18:2 describes a first-principles specialization that can connect principles, theories and signatures with method choice, work planning, work execution, separate measurement or evaluation, and cycle return. The parts used depend on the receiving question; E.18.1:4.0b governs when an explicit transformation-flow structure is needed.
method and its contextual synonyms
Preferred explanatory name. Use method as the main Plain word for a reusable way of doing something. It makes the question “Which method are you using?” easier to distinguish from a question about the work currently being performed. Use U.Method when the technical kind matters, under A.3.1. This naming preference applies to explanations; it preserves clear professional terminology and does not change the ontology.
Way, practice, pattern or template of work, workflow, culture, style, technology, strategy, a kind of labour, activity or engineering, and sometimes function can refer to a method in a sentence. They are contextual synonyms with useful nuances, not global identities. For example, style may foreground how a way is performed, and practice may foreground its established professional use. Recover the claim through A.3.1 before deciding whether another kind is meant.
Retain practice/practise in professional expressions and in ordinary accounts of practising to acquire capability. A concrete Work occurrence, community, discipline, transmitted tradition, tool or description keeps its kind. Neither a synonym list nor a preferred word turns it into a Method. Conversely, writing practice, culture and method in one account supplies no basis for inventing three distinct entities.
Try the word method in place of the professional name. What reusable way would the reader identify? What else would they lose? Way of working can name the method directly. Practice may also say that people in a particular field or workplace regularly use it. Technology can draw attention to the tools on which it depends. Style can identify how an individual or community performs it. State the extra claim when the reader needs it. Evidence that a team selected a development method leaves its capability, resource assignments and actual performance to be established separately.
Some engineering communities call a larger method a method, one of its constituents a practice, and a characteristic manner of execution a style. These names alone establish neither a part–whole relation nor a level of model abstraction. For a constituent, show what it contributes to the encompassing method under B.1.5. For a narrower choice of method, identify what it retains and what it changes under A.3.1:4.8. Keep the professional expression when it already permits the intended decision.
Apply this preference when authoring or repairing the explanation, with F.19's whole-span reading. Ordinary synonyms do not each require a registered Plain Twin. E.10.ARCH supplies a deeper route only when the kind or relation remains unresolved.
Whole-corpus trigger use
When a whole-corpus cleanup is selected, use this pattern's trigger guide over claim-bearing FPF text and project text that deliberately uses FPF-governed terms, pattern references, relation names, or conformance claims.
Do not perform a global string replacement. Use search only to locate candidates; read each natural span through F.19, preserve accepted FPF names unless a separate naming decision changes them, and let the selected review or campaign carry corpus coverage.
case, scenario, example, pilot, anti-case
Use F.19 to decide whether the text needs several cases, whether each changes recognition or action, and whether an illustrative series could plausibly be mistaken for a complete classification. Usually give the proposition first. When starting from a case, show the reader what decision or distinction it illustrates in time to interpret it, then bring its contribution back to the question. Mark an illustrative series as open where needed. An argument within one case may cover the full range. If the reader must choose a case or construct the explanation, show what guides that choice.
If the word also carries an exact FPF use, name that use directly—for example a project situation, worked example, pilot, negative control, evidence case, comparison case, or source example—and apply its subject pattern. Do not reproduce this menu in the repaired prose.
A case may illustrate or test a pattern. It becomes evidence, a decision basis, a source basis, or another governed object only through the relation and rule that establish that use, not through the label case.
basis, context, scope, frame
These words can hide different subject questions; they do not name one common kind.
- For basis, name the source or direct relation used by the sentence—for example, a decision, evidence, comparison, threshold, grounding, or admissibility relation.
- For context, apply
E.10.D1and recover the subject-defined content that changes the action—for example, a scheme, scope, model-use structure, situation, design-time or run-time referent, architecture, environment, domain subject, or local-practice use. - For scope, use
A.2.6when an actual claim scope and its slice-membership facts are current; otherwise use the subject pattern for the stated extent or boundary. - For frame, say what the word refers to—for example, a viewpoint, reference frame, comparison frame, state frame, or ordinary narrative framing.
If a basis changes what may be done, state the admissible use. State a relation claim or project-side FPF reference only when the sentence makes that claim. If context hides the EntityOfConcern, recover that entity and the subject relation before any Bridge, parity, or identity claim.
translation and multilingual heads
A bilingual alias is not a Bridge by itself and does not create equivalence, substitution, UTS admission, or a cross-local naming relation.
When translated wording has FPF-governed use, recover the FPF kind named by value, local head, publication construction, source relation, and admissible use before accepting the translation. Operative claims in a translated explanation retain the source-to-use grounds needed for reliance. Use E.17.EFP when an explanation-class or source-rendering ambiguity changes that use, and A.10 when its evidence/provenance account needs recovery or replay. A translated PublicationUnit may preserve form while shifting its primary subject or carried publication move; apply E.17.AUD or E.17.AUD.OOTD when that shift is being claimed. Local translated heads may use E.17.AUD.LHR or C.2.P without full F.18 unless durable cross-local naming, a UTS row, a Core-facing term, or a reusable FPF head is intended.
state, status, posture, readiness
E.10 only recognizes the wording problem; it does not duplicate the recovery procedure. If readiness or ready still hides which governed value is meant, use E.10.MOVE. It exits to A.19.SPR only for a hidden bearer and state frame, to A.2.5 for an assignment-state condition, to A.15.5 for work-entry readiness, to A.21 for a distinct gate decision, or to another direct pattern for the recovered claim.
For other state-family wording, use the direct pattern when the exact object and claim are already clear. Use A.19.SPR only while the object, state frame, or value remains hidden. Close with an ordinary statement of the recovered object and claim under that pattern; introduce a predicate only when the pattern defines or needs one. Otherwise keep ordinary prose, quote-only wording, a reduced-use cue, or a blocker.
live, current, active, and status or article overwrap
live, current, active, open, pending, and similar status-like modifiers are trigger wording when they attach to pattern, record, object, field, operation, route, locus, move, text, claim, question, use, or relation without saying which exact bearer and state or currentness value, temporal qualifier of an obtaining direct relation or assertion, source or use relation, or claim function the modifier adds.
First recover whether the modifier expresses an FPF value:
- If
readinessorreadystill hides which governed value is meant, useE.10.MOVE; useA.19.SPRonly if that recovery leaves a hidden bearer and state frame. For source currentness, another state or status, publication-use disposition, quality result, admission state, campaign state, or process state, useC.2.P,E.9.DA,E.21,E.19, the release or process carrier,A.19.SPRwhile its object or state frame remains hidden, or the direct pattern for the recovered value. - If it means a claim, question, use, or relation is currently asserted, relied on, or action-bearing in the described situation, keep the modifier only when the sentence also names the claim or claim-bearing episteme, obtaining relation or source/use relation, admissible use, and—when needed—the pattern that defines it, or says why ordinary prose is enough.
- If it only points to "the thing under discussion", treat it as phrase-level apparatus and apply
F.19: writethe pattern,pattern of concern, record kind named by value, affected field, operation claim, relation claim, or other object named by value instead oflive X. - If it is development, review, projection, landing, or current-campaign state about an FPF pattern version, keep it in the process, quality, projection, release, or campaign carrier rather than in the pattern unless that state is the pattern's primary
EntityOfConcern.
Deleting live or replacing it with another status word is useful only when the changed sentence now states the ordinary bearer and claim. Otherwise recover the governed state, currentness, temporal, source-use, or relation value, or leave the meaning blocked; then rerun the local F.19 reading.
claim, evidence, witness, ground, proof
Claim is not a synonym for sentence or prose.
Evidence is not a synonym for source, proof, approval, or confidence.
For claim, recover:
- claim-bearing episteme;
- claim node, claim content;
- EntityOfConcern or claim referent;
- viewpoint and the conventions needed to read the selected expression;
- admissibility target when the claim is used.
For evidence-like words, recover:
- evidence record or evidence-provenance relation;
- witness or source pin;
- grounding relation;
- validation result;
- assurance argument component;
- provenance mark only as provenance, not as evidence by itself.
If evidence is being read as engineering justification, gate passage, deontic permission, work authorization, safety assurance, evidence sufficiency, release authorization, or release confidence, apply the FPF pattern for that stronger claim or use the project-side FPF kind and reference named by value instead of strengthening the evidence word.
authority, permission, approval, commitment, obligation
Use this entry when the recovered claim concerns a duty, permission, approval, or authority relation. The word alone does not establish that claim. For commit, committed, or commitment used for a decision, model choice, or version-control operation, recover that operation under its own rule. Use E.10.MOVE only while a route or work-entry transition remains unclear.
Recover the values required by that claim:
- the beneficiary System or reference required by the selected predicate; if source wording says role, apply
E.10.ROLEand cite a local system-role kind and classification or an obtainingU.SystemRoleAssignmentoccurrence only when that permission or authority predicate uses it; - speech act or issuing act;
- individual duty under
A.2.8for obligation, recommendation-as-duty, or prohibition: its actual bearer, duty content, modality, scope and time, constitutive rule, and required instituting facts; keep any assertion or record about the duty separate from that relation; - exact
A.2.8.PERstrong grant, weak non-prohibition/non-violation finding, exercise relation, or permission-conflict finding; - policy claim and policy/currentness frame;
- authority relation;
- entry predicate or gate record or decision record when that is the claim;
- authority-changing decision;
- wording such as
delegated permission: recover theA.2.9granting or delegating speech-act occurrence and, only when the current policy validly institutes one, the resultingA.2.8.PER GrantedPermissionRelation@Context. Keep the grantor System, any grantorU.SystemRoleAssignmentoccurrence used by the predicate, the beneficiary System or reference, policy and currentness basis, scope and window, and any separate on-behalf-of or Work relation distinct. The cue mints neitherDelegatedPermissionRelationnor another generic delegation or authorization kind; if the pattern that defines the permission claim cannot be recovered, block operative use of the wording rather than naming a relation without a governor; - contestability, revocation, scope, window, and expiry condition.
Labels, badges, signatures, dashboards, certificates, comments, reviewer praise, and generated explanations may cue authority-looking cases. They do not carry authority unless the authority act, authority record, authority-reference relation, and evidence or provenance relation selected by the direct authority pattern are named.
profile, harness, catalog, registry, index, map
These usually point to a review profile, review harness, registry record, catalog publication, navigation index, map, publication form, companion publication, publication-companion relation, or relation between one companion publication and the publication unit or project record it helps readers inspect or use. Choose that kind named by value before writing; do not leave support record as the recovered head unless the named FPF pattern defines that record kind.
Treat one as an FPF pattern body, accepted campaign DRR, named current architecture document, or relation to one of them only when the named FPF pattern, accepted DRR, or architecture document and the obtaining direct relation with its actual participants are given by value; keep any row, index entry, or map element as its own claim-bearing episteme or representation under its direct pattern.
Split:
- review profile;
- review harness;
- source map;
- navigation index;
- registry record;
- catalog publication;
- benchmark harness;
- entry aid or discoverability aid;
- FPF pattern body.
If the named companion publication, review profile, review harness, registry record, index, or map mainly helps readers find, compare, test, or review something, keep it as a companion, navigation, or testing aid until a named FPF pattern or accepted DRR records the recurring action-guidance gain by value.
entry, front door, corridor, route
These terms often mix navigation, recognition, movement, and authority.
Split:
- entry publication or navigation aid;
- first-use recognition text;
- navigation-bearing publication;
- movement, control, and temporal relation;
- process sequence;
- corridor overview;
- FPF pattern applicable to the problem under repair; if source or local wording merely groups patterns, name the cluster phrase or relation phrase as literal wording and name the patterns involved by value; if an actual relation between patterns is being claimed, name the exact direct relation, its actual pattern participants, and the pattern that defines it.
An entry can make the right pattern easier to find. It does not prove the pattern is sufficient, complete, or ready for gate use.
same, parity, identity, equivalence, mirror
Similarity is not identity. Before accepting same, parity, or equivalence wording, name which relation is being claimed:
- mirror file in parity with a governing source;
- same EntityOfConcern;
- same claim content;
- semantic equivalence;
- bridge relation;
- version identity;
- file or carrier equality;
- source-publication identity;
- no-loss transform.
If the relation is about mirror parity, verify against the governing source or state that the check is not performed.
If the relation is semantic, use A.6.3, A.6.4, F.9, or the selected bridge pattern or equivalence pattern rather than relying on matching labels.
file, path, host, packet, bundle, package
These are carrier, transport, or package-form words.
Split:
- file or carrier;
- mirror file;
- file carrying FPF pattern text;
- document with named source-basis, evidence-basis, architecture-basis, or review-basis relation or use;
- review-facing target packet;
- review packet with its exact question, scope, and source set;
- release package;
- pattern package, pattern family, or pattern group under an accepted decision;
- governing source section.
A packet or bundle can carry a review target by value.
It is not automatically the authority-reference status, the target pattern, the accepted review result, or the FPF authoritySourceRef target.
quality, characteristic, metric, indicator, score
Name what is evaluated and how.
Split:
U.Characteristic;- characteristic space;
- Q-bundle;
E.21 PatternQualityQBundle;- scale;
- indicator;
- observed value;
- benchmark result;
- review finding;
- decision threshold;
- qualitative judgment with no scale.
metric can hide whether the sentence means a scale, value, or indicator.
If the text says a quality improved, name what changed: characteristic, scale, observed value, threshold, decision consequence, or admissible act, work, or claim.
If "quality improved" refers to an FPF pattern version, name whether the change affects an E.21 coordinate floor or declared coordinate target, status payload, stop condition, bounded non-use, or how E.21 or another named quality pattern is applied.
slot, field, row, label, badge, mark, cue
These words alone do not settle which FPF kind or relation is meant.
Split:
- A.6.5
SlotSpecinside a current reusable episteme-constitutionRelationSignature; - actual participant of an obtaining direct relation;
- A.6.5
SlotSpecinside another current reusableRelationSignature; - participant designation inside a current assertion or relation-occurrence-description episteme;
- schema field;
- table row;
- row in a pattern body;
- publication label;
- provenance mark;
- status badge;
- pre-articulation cue;
- displayed cue;
- evidence marker.
A label, badge, mark, or cue may trigger review. It does not prove currentness, identity, authority, evidence, gate passage, deontic permission, or release authorization unless the source relation and the evidence or provenance relation selected by the relevant pattern are named by value.
Using the cue surface
During normal reading, keep the compact cue surface available rather than loading every detailed lexical table. A search term or grouping mark can locate a candidate span, but absence of a cue is not clearance and presence is not a defect. Read the complete natural span through F.19; open one detailed E.10 row only if an FPF wording question survives that reading.
Multi-span and corpus use
For a document or corpus, the selected review or editing method defines coverage and manages attention. Search may group likely candidates, but each accepted change comes from reading its natural span and produces repaired text or a blocker; E.10 adds no separate coverage or progress inventory.
Compare recurring FPF-governed heads across the bounded object, especially a selected name, heading, table column, schema field, coordinate name, status value, or reusable authoring term. Also compare a head that carries the local architecture across substantive claims, replaces another broad head, or carries a finding or accepted basis into the final wording. If the repeated head conceals different governed objects or relations, split or rename those uses and reread the affected spans. Ordinary polysemy remains acceptable when each local meaning is clear; repetition alone is not a defect. Check the replacement head as well, so that a new umbrella does not merely preserve the old ambiguity.
If the declared scope is the whole document or corpus, representative examples do not establish coverage. That coverage obligation belongs to the review or campaign that selected the scope; it does not turn E.10 into an inventory format.
Recovery and disposition
Closure rules
Closure is the repaired text, the concrete result of the one selected subject pattern, or a blocker. A pattern citation or trigger label alone is not closure.
After a wording change, reread the affected passage through F.19 as far as the instruction or conclusion that uses it. Check the subject, relation, scope and action against the original task. When the sentence provides one premise, retain the connection or open question at the point of use.
When the positive sentence settles the use, close with that sentence and return to the domain task. Additional evidence belongs only to a receiving review or decision that needs it.
Optional wording-repair note
Use these fields only when a receiving review or decision needs an inspectable wording-repair note:
BoundedTextSpan: the exact sentence, row, section, pattern version,DRRslice, or project text deliberately using FPF-governed terms, pattern references, relation names, or conformance claims under repair.TriggerSpan: the word or phrase that carries possible FPF-governed use.SelectedInterpretation: one applicable repair-path classification from this closed value set—ordinary no FPF-governed use, local head repair, register repair, morphology repair, context-word recovery throughE.10.D1, learning-word recovery throughE.10.LRN, interest-or-curiosity recovery throughE.10.INT, bare-role meaning recovery throughE.10.ROLE, relation-like precision restoration, episteme precision restoration, publication precision restoration, source-use relation or source-ref target recovery, durable naming, or not-triggered false positive.FinalWordingOrBlocker: the accepted local wording, the result returned by the selected repair or pattern, or the blocker that remains.StopBackToSubstance: once the final wording or blocker is written, return to the domain question that made the phrase matter. Further lexical classification is non-use unless another phrase still hides an FPF-governed claim.
SelectedInterpretation retains that closed value set. If none of its classifications fits the current subject-specific repair, use the selected subject pattern's result instead of silently extending the lexical form.
When a wording-repair note needs formal fields, record one plainIntent before the technical fields. Use E.10.ARCH for the fact, declaration, report, or representation branch only while that distinction remains unresolved. Keep triggerSpan, boundedTextSpan, selectedInterpretation, LEX.TokenClass?, register, USM.Scope?, EntityOfConcern and Description-episteme boundary and specification use?, patternRef?, and finalWordingOrBlocker; add patternRef only when pattern identity changes the result. Add an exact assertion, predicate, or ClaimGraph only when the current claim or a named later use depends on that identity. Otherwise name the concrete object and the ordinary claim. Do not use slotOrUsePosition as a union field for actual participants, A.6.5 SlotSpec values, participant designations, or representation places.
Problem frame
Intent. Provide a normative lexical cue and repair rule set that keeps FPF wording composable across sources and uses. Authors, reviewers, and tooling use the subordinate material only after E.10:0.2 has selected one unresolved lexical question:
- Meaning constraints between lexical strata (Kernel → Extension patterns → Local use → Instance);
- Twin registers (Tech and Plain) with safe synonyms;
- Naming morphology (allowed suffixes and style) for the kernel’s core objects;
- Minimal Generality tests (names are neither parochial nor vacuous);
- Ontology recovery rows for overloaded words (e.g., process, function, service);
- Conformance checks and minimal examples.
Scope. Applies to: (a) Core (Parts A–G), (b) Extension pattern specifications (CAL, LOG, and CHR), (c) local source or practice glossaries that claim FPF conformity, and (d) diagrams and prose in normative text. For Tooling and Pedagogy, apply it where wording deliberately uses FPF-governed meanings or claims FPF conformity, as §0.1 specifies.
Problem
- Polysemy drift. Process, function, service, agent, activity slide between structure, recipe, execution, and promise.
- Cross-local collision. A label such as Owner is assumed to have one global meaning even though distinct sources, practices, or schemes use it differently.
- Name-bloat and parochialism tension. Either hyper-specific domain names leak into core kinds, or vague umbrella names obscure invariants.
- EntityOfConcern and Description-episteme boundary and specification-use collapse. Authors mix EntityOfConcern (the thing under concern), Description episteme (how we describe it), and specification use (testable criteria, formality, acceptance, and harness-gated use of a Description episteme).
- Register soup. Tech terms bleed into Plain pedagogy and vice‑versa, inviting category errors.
Forces
Solution - compact cues with lexical routes
Apply the connected F.19 reading to the complete natural span. If ordinary meaning settles the issue, repair the text and stop. Only a surviving FPF lexical question opens the subordinate LEX-BUNDLE or ULR material.
LEX-BUNDLE and ULR (Unified Lexical Rules) name subordinate register, naming, morphology, and local rewrite checks inside the current E.10 pattern. They do not name a second pattern, a second ontology, or a second audit. The retained detail covers lexical strata and their meaning constraints, Tech and Plain pairs, token generality, naming morphology, overloaded FPF heads, and conformance of durable lexical choices. Use only the detail needed for the selected problem.
This subordinate material does not replace F.19, E.10.ARCH, a selected precision-restoration pattern, the concrete pattern for the recovered claim, or F.18. F.19 governs the ordinary semantic and pragmatic reading. When subordinate material conflicts with E.10:0.2, E.10.ARCH, A.3.4.P, A.6.F, C.2.P, E.24.*, F.18, or another named pattern, the current applicability table and the pattern that defines the claim control the repair.
Use the exact subject pattern as soon as the governed object and claim become clear. After the lexical repair, reread the changed sentence through F.19 and return to the substantive task.
Lexical strata and one-way meaning constraints
Rule V‑0 (Strata). Every technical token whose lexical stratum is being governed belongs to exactly one stratum:
- Kernel — admitted
U.*names, core relation kinds, and invariants (for exampleU.Holon,U.SystemRoleAssignment,U.Method,U.Work, andU.PromiseContent). - Extension patterns — CAL, LOG, and CHR exports (e.g., Sys‑CAL, KD‑CAL, Agency‑CHR) that extend but do not override Kernel.
- Local use — the exact source or practice boundary, effective scheme, local meaning statements, aliases, local kind distinctions and classification rules that the use needs; cite an F.9 Bridge only when an exact relation between distinct local senses obtains.
- Instance — concrete identifiers for admitted holder Systems, exact
U.SystemRoleAssignmentoccurrences, Work occurrences, and carriers.
V‑1 (Unidirectional meaning). The relation between these lexical strata is a constraint on meaning in the direction Kernel → extension patterns → local use → Instance. Each arrow says that the preceding stratum constrains the next; it does not describe their position on a page. A local source, practice, or scheme may add a narrower designation or distinction, but it does not silently redefine a constraining stratum's term; any actual relation between distinct local senses is stated separately.
V‑2 (Strata and authoring stances). The four lexical strata above constrain tokens. They are independent of a claim-bearing unit's stance (its CtxState pins such as DesignRunTag, ReferencePlane, and Locus). Strata answer “what words mean here”; stance answers “where this claim is situated” and which evidence-lane expectations apply.
V-3 (Citation style). When a local Tech designation changes interpretation or action, its first use names the source or practice provenance and effective scheme needed to read that use—for example, ReviewerSystemRole under the JournalReview-2026 definition. Reuse under another local meaning first compares the exact governed values; use F.9 only if a direct Bridge between distinct exact cells obtains. A suffix may serve as a locator, but it establishes neither kind identity, admission, nor assignment.
V-4 (Firewall). Abstract Core definitions exclude incidental implementation idioms under E.5.1. Exact concrete syntax may remain when a rule governs that publication form, protocol or grammar and the syntax is necessary for its conformance condition; name the subject and edition/use boundary. Runnable teaching programs remain in Pedagogy/Tooling, with conceptual references and Tech anchors as applicable. Mathematical expressions retain E.5.2 interpretation, operative-use and mapping duties.
Ontology Guards
Tech register ontology guards
Purpose. This section stabilises the Tech register of the kernel lexicon by enforcing head-anchored naming, explicit kind naming, EntityOfConcern and Description-episteme boundaries, specification-use morphology, guarded use of bare role, exact
SystemRolecompounds, and subject-specific recovery of Domain wording. It aligns with E.10.D1, F.4 SystemRoleKindDescription, A.2.5 SystemRoleAssignmentStateRelation, A.2.7 SystemRoleKindRelationStructure, F.11 Distinguish Method, MethodDescription, Work and Outputs, and F.17 UTS. Scope: Guidance is register-agnostic and applies across the FPF. Apply Minimal Generality and Domain Anchoring (MG-DA, §7) to proposed names.
Onto1 — Head‑anchoring (use Kernel heads + pass LEX.TokenClass, EntityOfConcern and Description-episteme boundary, and specification-use gates)
- Rule: The head noun names the recovered object or relation and makes its kind or use clear. A kind head such as
System,Holon,Work,Episteme,Characteristic,Method, orTransformationFlowStructurerequires its governing membership basis.Descriptioncan name an independently identified episteme in descriptive use;Speccan name that episteme when specification use is admitted. Neither suffix admits a new kind.Profile,Card,Pack, andDashboardretain the object or form defined by their applicable pattern.SystemRoleis allowed only as the common compound inside one concrete local system-role-kind designation such asReviewerSystemRole; bare role remains a recovery trigger rather than a kind head. - Figurative heads with obvious overload (“Tradition”, “family”, “process”, “function”) are not admitted in the kernel. Plain twins are admitted only with a one-to-one Tech mapping and declared
LEX.TokenClassfor the Tech token. They appear in the Plain register as one-to-one mappings to a Tech token, not in the Tech register.- Do:
IncidentDashboard,MethodSpec,TraditionProfile,TransformationFlowStructureDescription. - Don’t:
IncidentBoard,TDD Tradition,Production Process(kernel),Service Function(kernel).
- Do:
Onto2 — EntityOfConcern and Description-episteme boundary and specification-use morphology (ref. E.10.D2)
-
Rule: A term for the EntityOfConcern uses the bare head for the FPF kind under concern:
MethodorCharacteristic.…Descriptionmay name an independently identified episteme in descriptive use under C.2.1 and E.10.D2. The suffix creates no generic Description kind; a specialized dependent kind additionally needs its own membership rule.In particular, a claim-bearing episteme is
U.MethodDescriptiononly when its EntityOfConcern is one admittedU.Methodand it makes at least one substantive claim about that method as a way of doing.Algorithm, code, pseudo-code, recipe, procedure, diagram, or other expression form first remains source wording, a representation under its applicable pattern, or a publication expression; C.29 applies only when mathematical-lens use is current; none establishes that membership.A qualifying Description episteme appends
...Speconly after a named specification-use gate grants that use. ThusMethodSpecis available only when the same episteme passes both A.3.2 membership and the E.10.D2 specification-use gate; formal language, pseudo-code, or bundled tests alone settle neither condition. -
Formal-description guard: An episteme stating a formal mathematical or physical theorem, including a formal postulate theorem in physics, may also be admitted for a bounded specification use. Its formal language belongs to formality and publication-expression discipline; that use is admitted only under acceptance criteria, harness checks, normative invariants, measurable anchors, verification use, or another specification-granting condition named by value.
-
Extension: Apply the same morphology to non-method EntitiesOfConcern where appropriate:
TransformationFlowStructureDescription,TransformationFlowStructureSpec,SystemDescription, andSystemSpec. -
Do:
SamplingMethod-SamplingMethodDescription-SamplingMethodSpec. -
Don’t:
SamplingAlgorithm(when it is just prose),SamplingProcessSpec(head not signalling kind).
Onto3 — System-role kinds, assignments, and carrier-relation separation (ref. E.10.ROLE, A.2, A.2.1, F.4, F.5, C.2.1, C.2.P, E.17, E.24.PUB, A.10, and C.35)
-
Positive distinction: A system role is an exact local kind for entities already admitted under A.1 as
U.System. C.3 recovers it through the candidate domain, operative work-facing membership condition, intended member/non-member boundary, and continuity rule. A practice or source reference locates the definition or signals a comparison; it does not identify the kind. Its Tech designation ends in...SystemRole, for exampleReviewerSystemRole. The name creates no admission, assignment, agency, capability, or Work. -
Assignment rule: A system-role assignment is an obtaining occurrence of one directly declared species under
U.SystemRoleAssignment. The species declaration definesHolderSystemSlot, the exact local system-role-kind domain ofAssignedSystemRoleKindSlot, any other participant meanings, its predicate, applicability, and occurrence-identity rule. The occurrence supplies the actual holder System, assigned-kind value, any other participant values, and extent. A source, interpretation, taxonomy, scheme, description, or display is not automatically an assignment participant; name it separately only when the assignment claim depends on it. -
Readable example:
Under the JournalReview practice, TeamAlpha is classified under ReviewerSystemRole because it can supply the substantive review judgment required by that practice.AddReviewAssignment-42only when the assignment matters and both its directly declared species and obtaining occurrence are recoverable.If performed Work is current, point first to its independently complete A.15.1 occurrence basis. Add F.6 only for a separately claimed precise assignment-bound attribution. A short attribution sentence may omit only an assignment identifier unused by the receiving claim; it does not omit a performer, Method, time, containing System, assignment occurrence, or F.6 attribution from the recoverable attribution basis.
-
Carrier rule: Carrier is not a free holon or system kind. Recover the direct carrier relation: under E.17 and E.24.PUB, identify the physical or digital carrier independently and recover its actual
PublicationExpressionBearingRelationto the particular expression; identify any reusable form and its expression-conformance relation separately.If a reusable carrier-relation declaration is separately current,
PresentationCarrierSlotremains the declaration-localSlotKindof one A.6.5SlotSpecand is not the carrier or relation. Other exits are a file, transport, rendering, front-end, or access-carrier relation under E.17; evidence or source-currentness carriage under A.10 or G.11; generated or produced carriage under C.35; or a named episteme-symbol carrier relation independent of any system-role assignment. -
Source-word rule: Job titles such as reviewer, owner, and lead remain Plain or quoted wording until the current claim is recovered. Use
E.10.ROLEfor an ambiguous claim-bearing role. Use...SystemRoleonly for an exact local system-role kind, and preserve owner when an actual architectural, organizational, policy, source, or responsibility ownership relation is what the sentence states. -
Do:
ReviewerSystemRole;ReviewAssignment-42 : U.SystemRoleAssignment;LeanTraditionCarrieronly when its direct episteme-symbol carrier relation is declared. -
Don’t:
Revieweras a U-kind,ReviewerCarrierfor an assigned system, an unqualified...RoleTech head, orCarrieras an unstated system kind.
Onto4 — Recover what domain means in this use (ref. E.10.D1 and F.17)
- Rule: The word domain does not create a kernel kind, catalogue mark, family, bundle, or inheritance relation by spelling. Recover what domain names in the current claim—for example, a DPF subject, discipline, source-defined field, model domain, market, physical region, or policy extent—or keep it as ordinary prose when it carries no FPF-governed use.
- Local-meaning rule. When a durable domain expression carries source-local meaning, identify its exact source edition, effective
ReferenceScheme, local expression, local-sense claim, and any obtaining basis relation. Create an F.17SchemeSenseCellonly when a named receiver needs a stable address. Use F.9 only for an independently obtaining Bridge between distinct exact cells; state any proposed use separately. - Discipline boundary. Use
U.Disciplineonly when the claim satisfies the pattern that defines that kind. A domain label, shared vocabulary, or UTS row does not establish discipline identity. - DPF boundary. A DPF states its domain subject, intended audience and use, source basis, scope, and qualification window under
E.4.DPF; it need not inventDomainFamily,DomainBundle, or a list of Context identifiers. - Do: “The Clinical Safety DPF addresses adverse-event analysis and device-labelling decisions for the stated audience and scope.”
- Don’t: infer
ClinicalSafetyDomain,DomainFamily, orDomainBundleas a kind from that wording.
Onto5 — Always state what the term names
- Rule. The definition or first line of a gloss states the FPF kind or object named by the term—for example, a
U.Holon,U.System,U.Episteme,Profile, exact local system-role kind,U.Workas the admitted kind or a Work occurrence admitted under it,Characteristic, or direct carrier relation. - Do: “Kind named:
ReviewerSystemRole— the exact local kind whose admitted-system candidates satisfy the current substantive-review condition. Its member/non-member boundary and continuity rule are recoverable under C.3; the named review practice locates that definition. A concrete assignment names its directly declared species and one separately obtaining occurrence underU.SystemRoleAssignment.” - Don’t: “Reviewer — a person who …” (blurs the kind named).
Onto6 — Bans and ontology recovery hints (mirror E.10 § 9 L-rules; do not duplicate tables; not a substitution table)
process,procedure,workflow,function, oractivity-> first recover the wording family: change-situation wording appliesA.3.4.P; function-like wording appliesA.6.F. Possible recovered values includeU.Method,U.MethodDescription,U.WorkPlan, one dated Work occurrence admitted underU.Work, a separate episteme about it,U.Transformation, andTransformationFlowStructure. Choose among them only after naming the object, any obtaining method-side or other relation and its participants, the relevant declaration or representation use, or the claim kind and the pattern that defines it.traditionorlineage-> recover the exact historical-continuation, source-use, method, edition, derivation or provenance claim under its subject pattern. Under C.20 these may remain ordinary auxiliary values or C.3 project-local kinds; the label alone admits no public U-kind or Tech designation.domain-> apply Onto4: name the actual domain subject, source or practice boundary, effective scheme, discipline claim, DPF scope, or ordinary use that matters here. Do not inferDomainFamily,DomainBundle,ContextId, or a UTS row from the word.…CarrierRoleused for an assigned System -> start withE.10.ROLE; recover the holder System, local...SystemRolekind, A.2.1 assignment occurrence, and its declared species only when the passage asserts those facts. Recover carrier, source relation or source-local meaning, interpretation, publication, evidence-use, and Work claims through their own relations.- ambiguous owner wording -> recover the precise relation or other claim being made—for example, an architectural, organizational, policy, source-maintenance, responsibility, authority, commitment, or work-facing claim. Keep owner when that precise ownership relation is current; use a
...SystemRoledesignation only when the recovered object is an exact local system-role kind. - job titles (
owner,lead,champion) in the Kernel -> keep them in Plain or quoted wording until the claim is recovered; use exact...SystemRoledesignations only for admitted local system-role kinds. - Do:
ReturnsTransformationFlowStructureDescription;LedgerTeam is classified under LedgerCustodianSystemRole, with any exact assignment, responsibility, authority, source-maintenance, or interpretation relation stated separately when current. - Don’t:
Returns Process,TDD Tradition(kernel),Ledger Owner(underspecified).
Worked mini-examples across arenas. These names illustrate morphology only. Every ...MethodDescription presupposes one claim-bearing episteme whose EntityOfConcern is one independently admitted U.Method and whose claims pass A.3.2; every ...Spec also presupposes its subject-specific specification-use gate. The label establishes neither condition.
The Onto3 block above is the one bounded distinction and assignment example. The twelve rows below are morphology cues, not classification or assignment assertions. Read each candidate System label and candidate local system-role-kind label separately. Before asserting classification, pass C.3 and A.2; before saying an assignment obtains, admit the holder System and establish both the A.2.1 occurrence and its declared species. A schedule, place, office, desk, title, source or practice cue, taxonomy, scheme, or interpretation episteme supplies none of those facts by wording.
Checklist before minting a KernelToken
- Head noun signals kind (Onto1).
- EntityOfConcern and Description-episteme boundary and specification-use morphology correct (Onto2).
- If system-role-related or carrier-related: local system-role kind, direct assignment species, and carrier relation remain separate; holder-system admission is explicit and the direct carrier-relation pattern is named (Onto3).
- Any action-changing Domain wording recovers its subject, use, and applicable pattern; a durable local expression uses F.17 only when its source-local meaning or public term row is current (Onto4, Onto6).
- Object‑of‑talk declared (Onto5).
- SCR-LEX rewrites checked for current system-role-kind, direct assignment-species, and carrier-relation separation (Onto6).
Note on registers. A technical designation first names a value admitted under its own subject pattern. A Plain twin then names that same value under §6.2. Tradition or lineage wording follows C.20's exact-subject and relation recovery; the words alone identify neither an episteme about epistemes nor a public kind.
Onto‑Deon — Deontic lexicon guard (Core register)
Rule. Recover what standard names before choosing a Tech head: a normative document and its compliance use, a measurement reference, or another established subject. A speech act that institutes an obligation is a separate claim under its own pattern.
For interface and boundary invariants concerning things such as holons, interfaces, and ports, name the exact invariant, compatibility condition, compliance profile, acceptance specification, or interoperability profile by value—for example InterfaceCompatibilityCondition, ComplianceProfile, AcceptanceSpec, or InteropProfile. State any promise or commitment separately; naming does not make it a property of the thing.
For a normative document, identify the episteme, the criteria it states and the compliance use that gives them force. A metrological measurement standard can instead be a measuring system, material measure or reference material used to realize a quantity with a stated value and uncertainty (VIM3:5.1).
Rename a …Standard designation only when its head obscures the recovered subject or conflicts with the applicable naming rule. Keep that subject distinct from a description of it, any publication occurrence and any separately established duty.
Rewrite hints (Tech → Tech).
publication Standard → publication standard;
frame Standard → frame standard;
measurement Standard → measurement standard;
Method Interface Standard (MIC) → Method Interface Standard (MIS);
Boundary-Inheritance Standard (BIC) → Boundary-Inheritance Standard (BIS).
Rationale. The same word can name a normative account, a measurement reference or another subject. Recover its kind and use before naming it. Analyse deontic acts through the Gov lens; ReferencePlane retains {world | concept | episteme}.
Twin‑Register Discipline (Tech and Plain)
Plain twin (LEX). A registry entry pairing the authoritative Tech designation with a display-only Plain designation for one named value under one stated local meaning and effective ReferenceScheme: an admitted durable U-kind, C.3 U.Kind, Concept-Set row, imported signature symbol, or another value whose kind and definition are already known. The LEX registry checks the pairing under PTG (Plain Twin Governance) and identifies it by Twin-Map ID (LEX). Create an F.17 SchemeSenseCell only when stable reuse or another named receiver needs an exact address. “Plain twin” ≠ the Plain register (the register is where twins may be used; the twin is the 1:1 mapping).
Convention. In this spec, Plain (capitalized) names the register; plain twin (lowercase) names the 1:1 mapping entry.
Rule R-0 (Registers). Every Kernel and extension-pattern concept has a Tech designation used in testable semantic clauses and may have one Plain designation for a stated didactic use. The Plain designation is admitted only when it names the same value under the stated local meaning and effective scheme; it does not create another value, cell, kind, or relation.
Allowed pairs (normative table; examples)
R‑1 (Plain first-use). At first use in a section, show the Tech label and, optionally, the Plain twin after its governing claim condition is established; for a kind label, establish membership first: "…one U.Method (the how-to); and, when a separately identified claim-bearing episteme has that method as its EntityOfConcern and passes A.3.2, one U.MethodDescription (an account of that how-to, sometimes called a recipe)…"
R-2 (No unpaired Plain in CC). Conformance Checklists use Tech labels only.
A source or practice may use local aliases in its glossary. Each alias points to one Tech designation under an effective scheme and an explicit local meaning claim. Create a SchemeSenseCell only when a named receiver needs a stable address; use F.9 only when an actual relation between distinct exact cells is current.
A plain twin (reader-friendly label) preserves the named value, local meaning, scope, and reader expectations of the Tech designation; it is display-only and local to the stated source, practice, scheme, and use.
- Tech name (tech) — the canonical, kernel-conformant label used in normative clauses (for example
U.SystemRoleAssignment,TransformerSystemRole). - Plain twin (plain) — a didactic display alias permitted in expository prose and UI display only for the stated local meaning and use.
Principle: The Tech designation names the value; a Plain twin may not change that value or its stated local meaning. Locality comes from the named source, practice, effective scheme, and use. A Bridge is added only when its F.9 predicate obtains.
Plain Twin Safety constraints (normative)
CC‑TWIN‑1 - One‑to‑one and local. Each Tech designation has at most one plain twin for one stated local meaning and didactic use; that plain twin points to at most one Tech designation in the same use.
CC‑TWIN‑2 - Sense‑equivalence proof.
A plain twin names the same value under the same local meaning claim and effective scheme as its Tech designation. When an F.17 SchemeSenseCell exists for that use, both expressions resolve to that cell. The registry notes include at least one counterexample showing how the twin could be misread and why the stated use still passes.
CC‑TWIN‑3 - Head‑term discipline (HND). The plain twin preserves the head term of the Tech name or appends an explicit bracketed head on first use:
- A Plain twin for one exact local system-role kind keeps “(system role)” and names its Tech designation on first use. Bare role is not a Plain twin for a universal kind. When service or access still hides its object or relation, follow L-SERV and A.6.P:4.11a; after recovery, keep that object's or relation's head. Methods keep “(method)”,
U.Workas a kind keeps “(work kind)”, one Work individual keeps “(work occurrence)”, a separate episteme about it keeps “(work record)” only when its Tech name denotes that record, and Capability keeps “(capability)”. Examples:TransformerSystemRole→ “Transformer (system role)”,U.PromiseContent→ “post-op monitoring service promise (promise content)”; an exact access relation → “service access (access relation)”,U.Work-> work (work kind);PumpInspection_2026-07-22T0900-> inspection work occurrence;PumpInspectionRecord_2026-07-22-> inspection work record only when that Tech name denotes a separate episteme.
CC‑TWIN‑4 - Kind‑consistent. A plain twin does not map across Kinds (C.3). If its everyday interpretation can denote a different kind—for example, Tradition as organization, corpus, or field—it is admitted only with a bracketed head and a first-use local gloss (see CC-TWIN-7).
CC‑TWIN‑5 - Ambiguity stop‑list. The following base nouns are reserved and are not admitted as unqualified plain twins: Tradition, service, process, function, model, system, method, standard, library, dataset, evidence, activity, task, action. They are allowed only with an explicit head per CC‑TWIN‑3 and a first-use local gloss (CC-TWIN-7). (This list may be extended in the registry.)
CC‑TWIN‑6 - No cross-local relation by label. Plain twins are not portable by spelling. Reuse under another local meaning first recovers that exact value, scheme, expression, and local-sense claim. Cite an F.9 Bridge only when its direct relation between distinct exact cells obtains; names alone carry no authority, equivalence, or substitution.
CC‑TWIN‑7 - First‑use gloss.
At first occurrence in a document or screen, show a plain twin as “Plain twin [Tech designation] — local gloss”, for example:
“Transformer (system role) [TransformerSystemRole] — one local kind for systems already admitted under A.1 and eligible for the stated transformer assignments in OR_2025; classification creates neither an assignment nor Work. An assignment claim names both an A.2.1 occurrence and its declared U.SystemRoleAssignment species”.
CC-TWIN-8 - Normative and didactic placement. Use Tech names in Conformance Checklists, predicates, type signatures, and acceptance clauses. Reserve Plain twins for didactic use.
CC‑TWIN‑9 - Twin budget. At most one plain twin per Tech designation for one stated local meaning and didactic use. Synonym piles are non-conformant because they create uncontrolled vocabulary sprawl (see F.14).
CC‑TWIN‑10 - Registry entry and DRR.
Every admitted plain twin has a registry entry recording tech, plain, referenceScheme, localSenseClaim, sourceOrPracticeBoundary, didacticUse, head, SenseFidelity = {3,2,1,0}, ambiguity notes, counterexamples, and DRR id. A change opens a DRR.
CC‑TWIN‑11 - Tests.
Twin entries pass the Twin Harness (see F.15): Head term, Kind consistency, same value and local meaning, Stop-list compliance, and First-use gloss. When a SchemeSenseCell is current, the harness also checks the exact cell.
Minimal Generality and Domain Anchoring (MG-DA) — names neither parochial nor vacuous
Principle (MG-DA). A minted name is as general as necessary and no more, and its head noun is anchored to the FPF kind being named. First classify the NameToken using
LEX.TokenClass, then apply the guardrails corresponding to that class: kernel tokens unify across domains; discriminator tokens and existingContextTokenvalues make one recovered local source, practice, scheme, subject, or use legible from the name itself.ContextTokenis the existing token-class identifier; it denotes no Context object. The name and its definition should let the intended reader distinguish the governed value from a plausible alternative at the scope where the term will be used.
LEX.TokenClass (meta‑lexical; not a USM Scope)
Definition. LEX.TokenClass : NameToken → {KernelToken | ContextToken | DiscriminatorToken}.
This is a local lexical classification function on NameTokens with the closed value set {KernelToken | ContextToken | DiscriminatorToken}, used by the LEX registry and MG-DA checks. It is not thereby a U.Characteristic or a CharacteristicSpace; that CHR reading would require a separately named U.Characteristic with one declared CSLC scale.
It is not a USM scope and carries no truth or validity semantics.
KernelToken — Minimal Generality (MG‑K)
MG-K1 (Ground the shared meaning). In the DRR or Glossary note, state what the candidate designates, where the designation will be used and which defining property or relation is preserved in those uses. Test an admissible variation that could change the meaning or the decision. The defining rule may itself establish the full reach. Where a different domain supplies a needed premise, show how it applies there. For an admission or placement decision about a universal-core U-kind, apply A.8:3 and E.24.UK, including their three-domain test or an applicable previous result. A designation for an already defined object retains that object’s admission basis. Use a ContextToken only after the exact local source, practice, scheme, meaning, and receiving use are recovered. Use an A.19 CharacteristicSpace only when one named U.Characteristic, one declared CSLC scale, and the exact receiving use make that construction current; an assignment-state relation remains under A.2.5.
MG-K2 (No parochial nouns). Kernel names contain no domain nouns such as Ticket, Microservice, Patient, or Developer. Domain-looking wording is a recovery trigger, not a destination: it may denote a C.3 local kind, exact system-role kind, system-role assignment, system or architecture object, episteme or record kind, declared Characteristic and scale, source wording, ordinary qualifier, recovered local-use ContextToken, or another value whose kind and use are already defined. Bare role has no default Tech reading. SystemRole appears only inside a concrete local kind designation admitted through C.3 and A.2; lexical shape alone creates neither that kind nor an assignment.
MG-K3 (No vacuity). Avoid vacuous heads such as Thing, Event, Process, or Resource. Use existing U-kind heads such as U.Holon, U.Work, and U.Method.
MG-K4 (Intent after recovery). U-kind names and labels for an exact local system-role kind or its F.4 SystemRoleKindDescription encode recovered semantic intent rather than notation, implementation, or local-realizer accidents. Algorithm, hardware-form, and recipe-flavor wording is a recovery trigger, not one ontological family: name one U.Method, qualifying U.MethodDescription, qualified holder-ability claim under A.2.2, U.Mechanism, system or architecture object, A.19 CharacteristicSpace, A.2.5 SystemRoleAssignmentStateRelation, representation under its direct pattern, formal substrate, source wording, or another value only when the predicate for that value is satisfied. Do not use Capability, CharacteristicSpace, assignment-state relation, or mechanism as disposal bins for unlike cases.
MG‑K5 (Notation independence, SHOULD). The EntityOfConcern-side kind criterion is separable from any one notation or toolchain.
MG-K6 (Refactoring safety). If a name fails MG, record a DRR and apply F.13 Lexical Continuity and Deprecation rather than mutating it silently.
DiscriminatorToken and local-use ContextToken — Domain Anchoring (DA-D)
DA-D1 (kind anchoring). The head noun names the FPF kind or exact local construction being classified—for example Sense, Bridge, Characteristic, SystemRole, or another subject-specific head. A concrete local system-role-kind designation may use the SystemRole compound, for example ReviewerSystemRole; bare Role fails this test because it does not reveal whether the claim concerns classification, assignment, participation, declaration, representation, episteme use, or ordinary wording. Readers can answer “X of what?” without relying on an unstated container.
DA-D2 (Enumeration rule, not axis). An enumerated property is a CHR construction only when one named U.Characteristic is bound to one declared CSLC scale in a CharacteristicSpace. Otherwise recover the closed value set, classified kind, local classifier, state/status frame, source wording, representation under its direct pattern, example or alternative set, or another construction defined for that value. Avoid spatial metaphors (axis, dimension, plane, lane, tier, layer) unless the metaphor is a pattern-defined primitive in this spec.
DA-D3 (Enum clarity). If the term denotes an enumeration, the value set is small and closed, membership criteria are obvious from the definition, and the kind being classified is explicit in the name (e.g., SenseFamily, not bare Family, RowPlane or overly general Facet).
DA-D4 (Anti-recipe). Do not bake how-to or local methods into discriminator names. The way of doing belongs in one exact U.Method; a claim-bearing episteme belongs in U.MethodDescription only when that method is its EntityOfConcern and A.3.2's positive threshold is met. Use A.2.2 when the recovered claim concerns a holder's ability under stated conditions and bounds; that claim alone calls for no additional kind.
DA-D5 (Mapping discipline). A cross-local relation is not inferred from similar labels or token classes. Use F.9 only when its direct Bridge predicate obtains between exact distinct cells; discriminator names do not suggest global identity.
DA-D6 (Register discipline). Keep normative tokens stable; synonyms belong in the Plain register only and stay outside constraints and tests.
DA-D7 (Ban generic combinators). Reject vague composites like NameUseMode, NamingScope, RowFacet, RowPlane, or RowLane. Each candidate passes DA-D1 and DA-D3 for a kind-anchored head, explicit classified kind, and closed-value interpretation under its classification rule. Require a CharacteristicSpace only when one named U.Characteristic and its CSLC scale have independently been declared.
Global tests (apply after 7.2 and 7.3)
MG-DA-T1 (Reach of the designation). For a KernelToken, check the grounds for the common meaning under MG-K1. For a local or discriminator token, check that the definition distinguishes its intended value under the relevant scheme. A contrast is useful when the alternative would change identification or use.
MG-DA‑T2 (Object‑of‑talk). The head noun uniquely signals the subject area; avoid free-floating metaphors.
MG-DA‑T3 (Implementation-word recovery). Do not relocate mechanism- or implementation-looking wording by lexical category. First recover the object and rule; remove accidental implementation wording from the candidate token only after that recovery. Use U.Method, qualifying U.MethodDescription, qualified holder-ability claim under A.2.2, U.Mechanism, a system or architecture object, A.19 CharacteristicSpace, A.2.5 SystemRoleAssignmentStateRelation, representation under its direct pattern, formal substrate, source wording, or another value only when its predicate is satisfied.
MG-DA‑T4 (Enum clarity). For an enumeration, list the closed value set, the kind being classified, and the rule that classifies it. Add a CharacteristicSpace only when the enumeration is one declared CSLC scale for a named U.Characteristic; list shape alone does not establish CHR membership.
MG-DA-T5 (Collision and uniqueness). Before merge, perform a full-text search over the corpus and the Reserved-Names registry. A candidate colliding with an existing token used in another FPF sense is not admitted; rename it or raise a DRR to deprecate the prior token.
MG-DA‑T6 (Teaching swap). In the stated didactic use under §6.2, the term can be swapped in without caveats.
MG-DA-T7 (EntityOfConcern ground). The definition card states the EntityOfConcern-side kind criterion for membership; reviewers can check membership without consulting external narrative.
Compatibility with USM (how tokens and scopes meet)
To check a mint, rename or use of a token, identify its class and the conditions of this particular LexicalAct, including the register and publication setting required by the policy. AllowedScopes(c) supplies the policy for token class c.
For one use, test whether its U.ContextSlice belongs to a scope admitted by that policy under A.2.6. To authorize use throughout a proposed scope, show that the policy covers every proposed slice. For a policy that explicitly requires registration of the exact scope, also check the corresponding registry entry. An unresolved test supplies no permission to use the token.
The LEX registry defines AllowedScopes(c) (for example, KernelToken use in normative kernel constraints is admitted; Plain-register use outside a glossary is restricted; a locally scoped use under another scheme requires its own valid designation rule and does not gain a Bridge by alias alone).
Audit. Violations are flagged as SCR‑LEX‑Sxx (see acceptance tests below).
Lexical prerequisites for a durable token
Mentioning LEX.TokenClass, LEX.Reserved-Names, or LEX.AllowedScopes does not create the values needed to admit a durable token. Before a NameCard can become current, the selected FPFCoreReferenceScheme must resolve four things: (1) the U.NameToken; (2) its current TokenClass classification assertion; (3) the current reserved-name or other authoritative collision-set value and the collision result; and (4) the current allowed-scope policy and a passing use-scope assertion. A prose example, full-text absence, heading, registry-shaped table, or intended policy supplies none of them by implication.
Worked near-miss. Suppose SelectedRuleContentSubgraphDesignation, derivedUsingRuleContent, and evaluatedAgainstRuleContent are proposed as KernelToken candidates. Assembly-rule content can be a premise of a manufacturing derivation, protocol content a premise of a clinical derivation, and deployment-policy content a criterion for a cloud release decision. What would make a name travel across these uses is its defining relation: for example, a derivation used the selected content as a formal premise, irrespective of the field. Evidence for case facts, evaluation against a criterion and operational support have different relations. F.18:4.2d gives the three candidate meanings. Even if their spellings do not collide in the inspected corpus, corpus absence is not a reserved-name or allowed-scope result. Do not publish their F.18 NameCards or F.17 rows until all four prerequisites resolve; a placeholder card or row closes none of them.
Using these names as candidate designators in declaration content changes no predicate semantics and does not admit them as public names, enumerations, Characteristics, CharacteristicSpace values, relation kinds, or U-kinds.
Role-Precision Token Classes and Allowed Uses
The eight selected names below are KernelToken values under FPFCoreReferenceScheme. Each names one value already defined or constrained by its subject pattern; the lexical rule admits no kind, relation occurrence, declaration, judgment, description, structure, NameCard, row, or publication occurrence.
For all eight names, Plain wording that silently turns the token into a neighboring object is prohibited. Reuse under another local practice, source, scheme, or meaning uses that value's identity and designation rules; when two exact cells are compared, any Bridge and use claim remain separately admitted. A change to the named value, TokenClass classification, or stable allowed-use rule reopens only the affected NameCard and row. A collision or conformance result for one dated corpus or candidate is evidence for its publication decision; it is not a reusable lexical rule or a currentness participant in the public pattern.
SystemRole alone is not a universal token with its own FPF kind or a NameCard subject. It is common morphology inside a concrete local designation such as ReviewerSystemRole. C.3 and A.2 recover that kind through its system-candidate domain, operative work-facing membership condition, intended member/non-member boundary, and continuity rule; practice or source provenance only locates or prompts comparison of the definition. AssignedSystemRoleKindSlot, SystemRoleAssignmentSlot, and fields ending in ...SystemRoleKindRef or ...SystemRoleAssignmentRef remain declaration-local ContextToken uses; the token-class name adds no Context object, and the fields are typed by existing U.KindRef or U.RelationRef, not by newly minted RefKinds. J_kindUse remains local notation. None receives a public row merely because the spelling recurs.
Metaphor guidance (informative heuristics)
Prefer object‑anchored heads to metaphors. If a metaphor is unavoidable, ensure it is (a) explicitly defined by a pattern here, and (b) unambiguous within the NameClass. Example families (use sparingly):
- For level, tier or ladder, identify what is ordered and by which relation. A holon can have constituent and enclosing wholes. Models of the same question can admit different answers under B.5.MR. A functional control layer may group responsibilities. Use a gate or upgrade criterion when advancement is actually the relation in question.
- Separation metaphors (lane, track): only where parallel, non‑interfering flows are enforced by rules.
- Grouping labels (family, class): name the classified kind or grouping and use its governing membership rule. For a closed enumeration, state its value set and classification rule; a kind or family governed by another membership condition is not limited to an enumeration.
Short‑form and acronym discipline
SF-1 (First expansion). On first use, expand the term and place the short form in parentheses (e.g., “Minimal Generality and Domain Anchoring (MG-DA)”).
SF-2 (Uniqueness). Register short forms in the Reserved-Names list and perform the collision check (MG-DA-T5).
SF‑3 (Form, SHOULD). Prefer typographic separators (MG-DA) to fused acronyms (MGDA). Use the fused form only in code or identifiers where punctuation is disallowed, and only after registration.
Examples (illustrative, canonical)
For BusinessService wording, use U.PromiseContent when the recovered claim concerns promised content; for Function wording, use A.2.2 to state the qualified holder-ability claim when it concerns a System's ability; for NaturalProcess wording, use U.Dynamics when it concerns a law of change. Each recovered claim must satisfy its subject pattern. Replace ScheduleProcess with U.WorkPlan only when one episteme passes A.15.2: one present EntityOfConcern, one horizon, at least one PlanItem, and substantive coordination claims about possible future performed work. Otherwise retain the schedule representation, planning cue, or other recovered construction.
Do not mint ETLService at kernel level. Recover the ETL claim first: the way of doing may be one U.Method; a separately identified claim-bearing episteme may be U.MethodDescription only when that method is its EntityOfConcern and the A.3.2 substantive-description threshold is met. An ETL label, pipeline diagram, code expression, mechanism, work plan, dated Work occurrence, or API publication establishes neither membership. If a relied-on service use still hides another subject or relation, apply L-SERV and A.6.P:4.11a and name the recovered claim; the suffix alone requires no promise, access, acceptance, Work, or publication branch.
Acceptance and regression checks (LEX and USM)
SCR‑LEX‑S01 (TokenClass declaration). Every normative token has a declared LEX.TokenClass.
SCR‑LEX‑S02 (Collision and uniqueness). Full‑text + Reserved‑Names check passes (no other meaning in FPF).
SCR‑LEX‑S03 (kind anchoring). Heads name the FPF kind classified (DA‑D1).
SCR‑LEX‑S04 (Enumeration rule gate). Every enumeration names its closed value set, classified kind, and classification rule. Require a CharacteristicSpace only when one named U.Characteristic is bound to one declared CSLC scale; otherwise retain the local classifier, state/status value set, source wording or representation under its direct pattern, example or alternative set, local kind, or another construction defined for that classification.
SCR‑LEX‑S05 (USM compatibility). For each LexicalAct, apply the allowed-use test in §7.5 to the token class and the proposed use.
SCR‑LEX‑S06 (Slot and Ref suffix discipline). A token ending in …Slot names the declaration-local SlotKind inside one exact A.6.5 SlotSpec of one reusable RelationSignature. A token ending in …Ref names either a RefKind admitted by its direct reference pattern or a receiving-episteme field explicitly typed by that RefKind; the field remains designation or reference apparatus and does not become the participant or SlotSpec. No ValueKind or representation field may acquire either suffix by shape alone.
SCR-LEX-S07 (Manifest provides follows exact signature claims). If a SignatureManifest is present, its provides entry is used only when that signature's exact claim content G under C.2.1 states that the signature introduces public names for dependent use. The entry carries that claim content or visibly represents it; list membership alone establishes neither provision nor a consumer dependency. Include only names introduced by this signature under the patterns that define them, such as its own A.6.5 relation-participant SlotKinds and RefKinds whose direct reference patterns admit them. A RefKind defined elsewhere remains defined there, and membership in an A.6.1 operation-argument or result declaration list does not transfer its definition to the manifest. A mathematical operand, table column, tuple place, or other representation element becomes no provided SlotKind by shape; any reuse still needs its independently governed declaration and explicit correspondence.
RSCR‑LEX‑E01 (Banned generics). Reject tokens matching the banned combinators list (DA‑D7).
RSCR‑LEX‑E02 (Metaphor hygiene). If a metaphor is used, show the pattern that defines it; otherwise rename.
RSCR‑LEX‑E03 (Strategy token minting). Reject new Kernel tokens named Strategy or Policy as kinds. Recover the subject first: use a lens, flow, or composition inside G.5 when that is the construction, or use a subject-specific …Description or …Spec under the exact project profile and effective scheme when that construction's admission gate passes. (Prevents kernel overloading; aligns with C.22 “no minted Strategy head”.)
Morphology and Lexical Form (LEX.Morph)
Principle. Form follows the FPF kind being named. A token's morphology (suffix, prefix, and casing) expresses what kind of thing it names, respects MG-DA (Minimal Generality and Domain Anchoring), and passes LEX.TokenClass gates:
LEX.TokenClass(token) ∈ {KernelToken | ContextToken | DiscriminatorToken}. Morphological choices never override EntityOfConcern, Description episteme, specification use, publication faces, publication forms,PublicationUnits, carriers, renderings, or CHR:ReferencePlane semantics.
Casing and basic forms
M‑0 (Casing and categories).
Kind names and exact local system-role-kind labels: UpperCamelCase (IncisionOperatorSystemRole, MethodDescription).
Relations and verbs: lowerCamelCase (performedUnderAssignment, isExecutionOf, bindsMethod).
IDs and instances: flat with delimiters chosen by the exact local naming scheme or project profile, but never colliding with kind-name or system-role-kind-label forms (e.g., W#Seam134, ctx:Hospital.OR_2025).
Register discipline: normative tokens use the Technical register; Plain synonyms are allowed in prose only, never in constraints.
Reserved suffixes (gated by LEX.TokenClass, EntityOfConcern and Description-episteme boundary, and specification use)
Use tables as a whitelist. Rows indicate when a suffix is permitted and what it means. The EntityOfConcern and Description-episteme boundary and specification-use gate prevents EntityOfConcern, Description episteme, specification use, and publication-relation confusion; “Examples” are illustrative.
Suffix conventions and retained-family boundaries
Notes.
- See §8.2 for inadmissible KernelToken suffixes and §8.3 for reserved prefixes.
- CHR guard: the only token that may use the word plane is CHR:ReferencePlane.
- Axis and dimension metaphors are not selected FPF heads; use Characteristic only for one declared measured aspect. For an enumeration, name its closed value set, classified kind, and classification rule; use CharacteristicSpace only when that enumeration is the declared CSLC scale of the named Characteristic (see § 7).
Guards beyond suffixes
- Suffix choices should pass the MG-DA tests for the kind being named.
- Other morphemes also respect kinds. Use Space for a space construction defined by its subject pattern; an A.19
CharacteristicSpaceis a product of declared Scale value sets and need not have a geometric overlay. Prefer Set, Kind or Kit when only membership is intended.
L-EPI-PUB — episteme, publication, view, carrier, direct-relation, representation, and authority-reference discipline
-
Use
U.Epistemefor the claim-bearing unit.U.EpistemePublicationis a rejected kind name: when the selected edition is available as a published episteme, name or make recoverable its exactEpistemePublicationRelationoccurrence, particular expression, bounded use, and actual carrier, with a reusable publication form and conformance relation identified when its rules matter underE.24.PUB. The rejected spelling may remain only in an explicit rejection explanation or a negative test, never as a positive object, kind, reference, or field. -
Name reusable publication-form rules separately from the particular expression and episteme. Such rules can govern a
PreArticulationCuePack, an expression of aU.AbductivePromptaccount, a typed bounded projection, a partial normal form, or an endpoint-specific record. Identify the filled expression when its content is being supplied or compared. A publication form is not itself the governing FPF source. -
Name the episteme-level
U.View, reusable MVPK face rules, and constructed face expression separately.PlainView,TechCard,InteropCard, andAssuranceLanedesignate publication forms under E.17. A particular expression follows the selected form; its source or receiving episteme hasU.Viewmembership only under the exact E.17.0 conformance predicate. Neither the face designator nor the expression is the SCR or RSCR carrier. -
Name the carrier or rendering relation separately. A document file, screen or card can supply a physical or digital carrier when its independently identified actual state or realization yields the particular expression; a dashboard or transport format must be recovered under its own use before that bearing is asserted. Bearing alone establishes neither the
U.Epistemenor the claim or effect being relied on and supplies no rule for that claim. -
Name source-finding cues separately from source epistemes. A cue, badge, credential view, dashboard tile, heading, signature-looking mark, or generated explanation may help find a source; it does not by itself create an
authoritySourceReftarget, evidence relation, gate decision, assurance claim, exactU.SystemRoleAssignmentoccurrence, status assertion, Work occurrence, deontic permission, or Work authorization. -
Use an ordinary PatternID reference when a reader only needs to find the rule. Add
relationFunctionClaimRefand the defining or constrainingClaimGraphonly when admissible interpretation, comparison, migration, publication, or reuse depends on that exact rule identity. UseauthoritySourceRefwhen a non-pattern target such as an external standard, editioned register, DRR, gate decision, policy record, system-role-assignment register, or status register carries the relevant authority. Do not use generic sign, source, project-work, or container-placement wording as solution terms. -
When the current claim depends on a particular method, plan or Work identity, name the relevant P2W element: intended method family, selected method or method of work, one exact
U.WorkPlanbaseline, planned work, or one actual Work occurrence admitted underU.Work.For an admitted Work occurrence, name any separate claim-bearing episteme about that occurrence. Name any separately current direct resource-use, affected-referent, operation-application, measurement, evaluation, decision, delivery, acceptance, or receiving-use relation under the pattern that defines it; when a production-work, entity-inception, or production-completion claim is current, name one local A.15.PROD claim instead of implying a universal production relation.
Apply
A.6.P.WMRonly while one such Work-to-Method boundary relation remains hidden after generic relation recovery. Do not let genericaction,use,material,work result, orresult measurementhide that distinction. -
Use
C.2.Ponly while an unresolved episteme, publication, source-to-use, carrier or use-disposition distinction prevents selecting or safely using the direct pattern. It returns the needed concrete object, relation, claim or disposition. When that question is already clear, use its subject pattern directly or keep the sufficient ordinary sentence.
Publication face, form, unit, and carrier discipline - surface as trigger wording
- Definition.
surfaceis trigger wording, not a durable FPF Tech head by itself. When it has FPF-governed use, recover whether the sentence means publication face, publication form, publication unit, carrier, rendering, UI face, front-end face, physical surface, geometric surface, companion publication, projection material, carrier relation, or another FPF kind or relation named by value. - Allowed final heads: publication or carrier terms named by value, or deliberately ordinary physical or geometric
surfacewhen no FPF-governed use is carried. - Inadmissible final heads:
StructureSurface,MechanismSurface,PortfolioSurface, and any...Surfacethat hides a structural, mechanistic, measurement, review, assurance, explanation, comparison, or publication-unit object. - Preferred alternatives: name publication face, form, unit, carrier, and rendering; use
...Boundaryfor structural borders,...Viewfor episteme and view relations, and...Cardonly for a UTS or record unit.
L-Space - Disciplined use of Space
- Use Space for a measurement, state or other formal space whose construction is defined by the applicable subject pattern, including an A.19
CharacteristicSpace. Do not infer a Space from a set, portfolio or publication form. Publish a portfolio or archive in the set, view or card form its use needs. - Field-name and direct-declaration guard. In A.6.0 and A.6.1 declarations, write
SubjectKindandRangedValueKindas direct content fields. AddResultKind,SliceSet, andExtentRuleonly when their distinctions are current. A heading that merely wraps these fields is presentation, not another declaration component, and receives no Tech name. Use Space only when the governed value has the space construction stated by its subject pattern. Let the referenced C.3 kind, admitted durable U-kind, Concept-Set row, or imported signature symbol carry...Spacewhere appropriate; use...Setfor an ordinary set-valued universe. - A.19 topology and distance are separately declared overlays. Their absence does not disqualify a CharacteristicSpace. For a portfolio, publication form or ordinary set, use the name of that construction rather than adding Space without its defining basis.
L‑ROLE — guarded recovery from role
- Bare claim-bearing role is a lexical trigger with no default Tech reading. Apply
E.10.ROLE, write the ordinary sentence with its recognizable object and action or relation, and stop as soon as one exact object or relation and its direct pattern are clear. - Use
SystemRoleonly as the common compound inside one concrete local system-role-kind designation such asReviewerSystemRole. Recover the kind through C.3's candidate domain, operative membership distinction, member/non-member boundary, and continuity rule. Practice or source provenance is a locator and comparison cue; the spelling creates no system admission, assignment, agency, capability, responsibility, participation, Work, evidence use, or status. - A participant meaning or actual participant remains under its direct relation; a reusable declaration place remains an A.6.5
SlotKindor A.6.1 declaration; a tuple, table, formula, graph, diagram, schema, or call position remains under its representation pattern and explicit correspondence. None becomes a system-role kind by wording. - Preserve ordinary or quoted role when no FPF claim relies on it. Preserve owner when a precise ownership relation is current—for example, an architectural, organizational, policy, source, or responsibility relation. This rule forbids lexical cleansing as well as default formalization.
Inadmissible suffixes and the DevOps, Data Governance and Repository-Workflow Lexical Firewall
M-F (Inadmissible in Kernel tokens). KernelToken names do not use ...Function, ...Process, ...Task, or ...Activity. These are ambiguous or vacuous; recover the object through section 6 before naming it: one U.Method, one qualifying U.MethodDescription, one Work occurrence admitted under U.Work, or another accepted recovered value. The source suffix alone selects none.
M-FW (Tool and file markers). Abstract Kernel names do not acquire incidental tool or file suffixes such as …API, …JSON or …YAML. Keep those implementation designations in their local glossary or operational source. A rule governing a named concrete publication form, protocol or grammar may quote its necessary designation or syntax under E.5.1, with the edition/use boundary. Such a quoted subject designation is not thereby a new universal KernelToken. This distinction preserves exact subject rules without importing a data-management ontology.
Prefix discipline
M-P1 (Reserved prefixes). U. is reserved for U-kinds admitted under E.24.UK; Γ_ for algebraic operators; CAL, LOG, and CHR for pattern packages. A local glossary, scheme, source label, or use never mints U.*.
M-P2 (Edition and version markers). Use a model-use marker only when A.1.1's direct model-use relations or a selected BoundedModelUseStructure require it. Select one exact episteme edition only through a typed reference whose referent is that U.Episteme, with a narrow selector defined at its reference-pattern locator. Do not attach an edition selector to an EntityOfConcern-side Method, Space, system, bare Service, or their references.
When a use depends on a defining, describing, CG-Spec, service-description, or service-offer episteme, reference that episteme separately. A service-access publication, publication occurrence, publication form, and carrier retain their own references and do not inherit the episteme selector.
Tool and carrier versions remain separate. Authors may annotate local service labels for didactics only after every named value is recoverable.
Norms (edition, release, and version).
- edition — one exact
U.Epistemewith its own C.2.1 identity. A later episteme is related to an earlier one only when the exactEpistemeEditionRelationpredicate obtains; shared label, order, file version, or selector value establishes none.PhaseOfmay describe one unchanged episteme over a proper interval but never connects different episteme identities. - release — a separately governed publication or release occurrence, or the exact Work that performs it when that Work claim is current. Publication occurrence, publication form, and carrier remain distinct; release establishes neither episteme identity nor
EpistemeEditionRelation. - version — a tooling or carrier identifier for a file, package, code object, rendering, or other carrier-specific use. It is not an episteme edition, publication occurrence, or release claim and does not belong in Core EntityOfConcern names.
Property discipline. There is no universal <Thing>Ref.edition property. A direct reference pattern may define a narrow edition selector only for a governed reference whose referent is one exact U.Episteme. A Space, U.Method, formula, system, publication form, or carrier reference remains a reference to that object; pair it with a separately governed episteme reference when one exact defining or describing edition matters. A selector value identifies neither an edition relation nor historical continuity.
Morphology tests (apply with § 7 MG-DA)
M-1 (Kind and participation test). Establish the named entity's kind and the participation or use asserted in this sentence under their direct rules. EntityOfConcern names a participant, not a kind disjoint from epistemes or carriers: an episteme can concern another episteme or itself. Preserve the distinction among the subject, the account about it and its expression or carrier in the relation being asserted. Rename or split only where the designation hides that distinction.
M-2 (Classified-kind anchoring). The head noun names the classified FPF kind or exact subject construction: exact local system-role kind, U.SystemRoleAssignment occurrence, Method, Work, Characteristic, Capability, constraint claim, U.Commitment, publication form, service-access relation, service-offer record, exact source or practice boundary, local-use designation, or another direct FPF value. Bare role first uses E.10.ROLE; no free-floating metaphor, bare Service, bare Context, or bare Requirement head passes by lexical shape.
M-3 (Family congruence). Where eligibility clarity is needed, add the exact subject-specific characteristic or SystemRoleAssignmentStateRelation as a separate qualifier for the current value; do not hide either in a system-role-kind name. Do not turn standards, requirements, evidence, or status labels into ...SystemRole names, and do not fake families with bare metaphors such as RowPlane, senseFamily, or ...Lane.
M‑4 (Run and description split). Use Work only for executions. Treat recipe, code, diagram step, procedure, or document form as recognition evidence only: classify a claim-bearing episteme as U.MethodDescription only when its EntityOfConcern is one admitted U.Method and its claims pass the A.3.2 substantive-description threshold; keep the method, representation, publication form, plan, and Work occurrence separate.
M-5 (Kernel parochiality). KernelToken names carry no domain nouns. Recover domain markers under the objects and rules that define them. Use ContextToken only after the exact local source, practice, scheme, meaning, and receiving use are recovered; use A.19 CharacteristicSpace only after its named U.Characteristic, declared CSLC scale, and exact receiving use are current; use A.2.5 SystemRoleAssignmentStateRelation only when that direct predicate obtains. Lexical shape establishes none of them.
M‑6 (Vacuity ban). Avoid vacuous heads (Thing, Event, Process, Resource). Use established U-kind heads such as U.Holon, U.Work, and U.Method.
M-7 (Notation independence). The EntityOfConcern-side meaning survives notation and tool swaps.
M-8 (Collision and uniqueness). Before admitting a token, perform full-text and Reserved-Names checks; a token colliding with another FPF meaning is not admitted (cf. MG-DA-T5).
Alias hygiene
Aliases are permitted only in one named local source or practice glossary under an effective scheme and explicit local meaning claim. Each alias points to one Tech designation; it does not assert semantic equivalence or a Bridge. No global aliases.
Entry lexeme support and lexical-query discipline
Public first-entry scenario text, ToC query rows, local Problem-frame recognition text, or worked entry comparisons in E.11 may use one compact entry lexeme cue block when the lexical issue changes the first useful FPF entry.
That cue block should not be copied into every pattern body by default.
Keep it instead in:
- FPF
readmesection, E.11public-entry positions,- worked entry comparisons in
E.11, Table of Contentsquery rows,- or one bounded lexical-query record governed by
F.17,UTS, orF.18.
This block remains one editorial lexical-query set. It does not mint names, aliases, durable U-kinds, bridges, or semantic equivalences by itself. When visible, it should distinguish at least:
- canonical label,
- plain-language twin,
- domain alias,
- lexical-query cue,
- rejected cue,
- false friend or inadmissible synonym.
Minimal visible lexical-query shape may therefore use one compact field set such as:
Ordinary lexical-query support should stay compact:
- ordinary
Table of Contentsrows: prefer2-5query phrases; - ordinary
READMEscenario orE.11entry-distribution cues: keep only the most discriminating domain phrases and false friends; - fuller lexical sets belong under
F.17, F.18, and E.10only when one naming, alias, bridge, or collision claim exists.
Lexical support should increase entry precision, not maximize keyword recall. The same boundary should be kept explicit in lexical support:
lexical_hookis not one alias;- one alias is not one canonical name;
- one search cue is not one semantic equivalence;
- one
entry_orientation_labelis not oneRelationKind.
Language-specific query cues may be added as entry-lexeme support. They do not become canonical names, aliases, or semantic equivalents unless an accepted F.18 naming settlement admits that use; otherwise keep the phrase as ordinary local query wording. Such a practitioner phrase may help recover a canonical FPF pattern while remaining lexical-query support only.
Compatibility with USM (acts and tokens)
Token class and use. Determine the class of the token, the setting of its use and the applicable AllowedScopes policy. A claim made by a LexicalAct has its own claim scope. A capability to perform the Work can have a work scope.
For an individual use, test the slice for that use. For a standing permission, check coverage as set out in §7.5.
Example: use of a KernelToken in a locally scoped constraint is admitted only under the exact scheme and allowed-scope rule; an alias adds no Bridge. Keep logged performance assertions separate from the MethodDescription's method-side claims, even when one file presents both.
Acceptance and regression checks (LEX and USM)
- SCR‑MOR‑S01 (Suffix whitelist). Every normative token with a reserved suffix matches § 8.1 row semantics and passes EntityOfConcern and Description-episteme boundary and specification-use gates.
- SCR-MOR-S02 (Kernel exclusions). KernelToken names contain none of the inadmissible suffixes from section 8.2.
- SCR-MOR-S03 (Prefixes). Reserved prefixes obey § 8.3; no local glossary, source, scheme, or use mints
U.*. - SCR-MOR-S04 (Run and description gate).
Worknames the admitted performed occurrence. Actual participants and results retain their own assertions and relations; they do not become MethodDescription fields or the Method. - SCR‑MOR‑S05 (Collision). Full‑text + Reserved‑Names checks pass (no other sense of the token elsewhere).
- SCR‑MOR‑S06 (Object‑of‑talk). Heads pass M‑2; no bare metaphors as heads.
- RSCR-MOR-E01 (DevOps firewall). Tool and file suffixes stay in local glossaries or operational configurations; none leak into KernelToken names.
- RSCR‑MOR‑E02 (USM compliance). Apply §7.5 to the proposed use in its register and setting.
Autonomy lexicon (L‑AUTO )
Recovery cues in FPF-governed autonomy claims: unresolved “validity”, “actor” or “agent”, “kill switch”, “process” for behavior, and “envelope” used as scope. Apply the relevant recovery only when the claim is unclear; ordinary and already precise uses remain available.
Recovered values when the claim needs them: Scope (G) for epistemic scope; WorkScope for capability bounds; an independently admitted U.System when an actor's systemhood matters. When a precise Agent or performed-Work claim is current, use A.13 for the exact local agential kind and criterion, classification, obtaining assignment, any scope, working situation and window required by the use, and adequate core evidence; require a characteristic profile only when its Grade, autonomy, criterion-dependent, profile, or assurance use consumes it.
For a performed-Work claim, use A.15.1 after A.13 to independently admit the dated Work on the basis of its actual performer, Method, time, and containment facts. Add F.6 only when the wording or receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment; F.6 identifies neither assignment nor performer.
A missing or failed attribution alone leaves independently admitted Work intact; recheck any A.13 or A.15.1 premise invalidated by the same finding under A.15.1:4.1. Use a ...SystemRole designation only when that classification matters. Use SpeechAct for overrides and SafeStop instead of “kill switch”.
Named prefixes (policy and registry):
aut:for AutonomyBudgetDecl fields (e.g.,aut:action_tokens,aut:risk_bands);guard:for guard checks bound toAdmissibilityConditionsId;ovr:for override SpeechActs (ovr:PauseAutonomy,ovr:ResumeAutonomy, …).
Notes.
- Scope-sensitive guards preserve every selector declared for their admission use. Add a Gamma_time point, interval or policy when time changes membership; do not invent a temporal selector for an atemporal slice.
- Proper names of patterns and components that already include “Agent” or “Agency” (e.g., Agency‑CHR, Agent‑Tools‑CAL) are permitted as titled terms; recover a bare “agent” when its precise agency claim is unresolved.
LEX-CHR-STRICT — Reserve Characteristic for CSLC-measurable aspects
Intent. Prevent calling non-measurable objects (sets, statuses, scopes, policies, bridges, contexts, guards) “characteristics”.
Rule L-CHR-S1 (Reservation). Use Characteristic for a declared measured aspect with its CSLC Scale and admissible values. State a Unit where applicable. Declare a preference rule when the use evaluates desirability; describing or comparing magnitudes needs no preference rule (C.16 and A.17–A.18).
Rule L-CHR-S2 (USM). U.Scope, U.ClaimScope (G), and U.WorkScope are USM scope objects, not Characteristics or CHR components of a CharacteristicSpace.
Rule L-CHR-S3 (Status). Episteme statuses, SystemRoleAssignmentStateRelation occurrences or assertions, deontic statuses, and epistemic statuses are not Characteristics by label alone; each remains governed by its direct pattern.
Rule L-CHR-S4 (Lexical classifiers). Keep a lexical classifier or tag under its classification rule: a local classification function and value set, source wording, representation element under its direct pattern, example or alternative set, status or state-frame value set, local kind or classifier, or another construction defined for that classifier. Call it a U.Characteristic only when that characteristic and one CSLC scale are declared. Do not default the residue to Facet, attribute, or another umbrella kind.
Checks.
- CC-L-CHR-1.
scope characteristic(s)is banned in Kernel and local-use Tech wording. - CC-L-CHR-2.
CharacteristicSpacenearScopeis a cue to inspect the claimed relation. Reject wording that treats a scope as a CHR component. A scope may qualify use of a characteristic space while remaining a distinct USM scope object. - CC-L-CHR-3. Kind-preserving repair:
F–G–R characteristics→F–G–R componentsonly when the recovered kind is component rather than characteristic.
LEX-QA-1 - Using terms with the -ility and -ilities suffixes
Rule. Tokens ending with -ility or -ilities or widely used quality names (Availability, Reliability, Security, Safety, Scalability, Maintainability, Usability, …) are Quality‑Family labels, not automatically CHR Characteristics.
Authoring choice:
- Keep ordinary praise or quoted wording ordinary when it carries no FPF-governed claim; use C.16.Q if the evaluative meaning remains hidden.
- Use one named
U.Characteristicand its declared CSLC Scale when that one measure carries the engineering quality claim. - Use C.25 Q-Bundle-shaped claim content only when several differently typed contributors jointly determine the claim or receiving action. Include only the load-bearing measures, scopes, windows, mechanisms, statuses and evidence.
Rationale. Scope is set-valued (USM) and not a CHR measurement. Q-Bundle mechanism and status fields carry mechanism references, control presences, certification states, or other status values admitted by their specific patterns; they are not generic governance records or measurements. Claim scope, work scope, CHR measures, qualification window, mechanisms, status values, and evidence keep their own kinds even when one Q-Bundle authoring structure coordinates them. (A.2.6 § 6.2; A.6.1; C.16, A.18, and C.25).
Ontology recovery rows for overloaded words (LEX L-rules; normative)
What this section does. LEX L-rules standardise how we recover kind and use in Core and local uses when overloaded everyday words hide FPF concepts. What this section does not do. It does not restate naming (see § 7 MG-DA) or morphology, casing, and suffix rules (see § 8 LEX.Morph); it depends on them. Guards. Tokens are classified by
LEX.TokenClass ∈ {KernelToken, ContextToken, DiscriminatorToken}(§ 7.1). Only CHR:ReferencePlane may use the bare word plane. E.10.D2 keeps the EntityOfConcern, an episteme that describes it, and specification use of that episteme distinct; specification use needs a granting gate named by value. Publication faces, publication forms,PublicationUnits, carriers, and renderings stay separate.An enumeration becomes a CHR construction only when it names a
U.Characteristicwith one declared CSLC scale. Without that declaration, recover the exact construction instead of defaulting to a non-measurable attribute: it may be source wording, a field or other representation element under its direct pattern, an example set, unresolved alternatives, a status or state-frame value set, a local kind or classifier, or another value whose kind and definition are already known. It becomes an A.6.5SlotSpeconly inside one exact reusableRelationSignature.
Hard bans and ontology recovery rows (single table; normative)
Use this table as a recovery guide. “Ban” means the listed phrase is not accepted as an unexplained Tech term in Core prose, identifiers, or diagrams. It may remain as ordinary, quoted, or source-local wording when its use is clear; otherwise name the recovered value or relation. EntityOfConcern and Description-episteme boundaries, specification-use gates, and token gates prevent object, description, specification-use, publication-position, and TokenClass leaks (cf. § 8.1).
Recovery example. For “The process ensures quality,” first recover what process names and which quality condition is claimed. If it names a Method, recover the quality claim and its grounds under
C.16.Q,C.25or the direct subject rule; useB.3only for a selected assurance claim. Leave an unsupported guarantee unresolved. Identify a separateU.MethodDescriptiononly when the receiving claim needs it and A.3.2 membership holds.
Diagnostic examples, not substitutions
Use these rows only after the compact cue surface selects the named lexical problem. Accept a repaired sentence when its object, head kind, relation or claim, use, and scope are preserved, the applicable § 7 or § 8 rule passes, and the changed sentence passes the local F.19 reread. If an example would change the kind in the current sentence, split the claim or leave a blocker; do not copy it as a ready-made rewrite.
Acceptance tests (LEX‑AC)
A text passes LEX if all answers are Green:
-
Action-changing use recovered. Each use of context that changes interpretation or action names the exact source, scheme, scope, model-use structure, situation, frame, referent, or practice that matters. Ordinary, quoted, and already defined uses remain available.
-
Distinguish the EntityOfConcern, the Description episteme, and its specification use. EntityOfConcern, Description-episteme, specification-use, publication relation, and run-record uses are not conflated (cf. § 8.1 gates).
-
Promise, ability, and performance split.
PromiseContent(promise clause),Capability(ability),Work(performance) are not conflated. -
Agentive predicates and metonymy. A doubtful agentive or causal predicate—including a negated one—calls for the whole-span
F.19check. Retain ordinary, unambiguous metonymy; otherwise state the capable participant or the exact non-agentive relation. Add a denial only for a grounded plausible misreading whose correction matters. -
Scheduling hygiene. No actuals belong in a
U.WorkPlan. A performed occurrence is admitted as datedU.Workfrom its independent A.13/A.15.1 basis. A complete A.13/A.15.1/F.6 basis is required only when the receiving use additionally claims precise assignment-bound attribution; a missing F.6 link alone does not revoke independently admitted Work; recheck any admission premise invalidated by the same finding under A.15.1:4.1.Other direct facts, including affected referent, bindings, and resource use, remain separate. A short attribution sentence may omit only an assignment identifier unused by its receiving claim; the underlying occurrence and attribution remain recoverable. An assertion, description, log, or record about the Work is a separate episteme, not the occurrence.
-
Cross-local relation. When the text relates two different local senses, it identifies the exact F.17 cells and cites an F.9 Bridge only if that direct relation obtains under the applicable F.9 relation profile. When a receiving use is current, a separate C.2.1 claim says what action is proposed, its use direction, correspondence rule, tolerated loss, and polarity. That claim does not show that the action occurred.
For reliance on the bounded-use claim, preserve the A.10 evidence-provenance path for that same use. Ordinary evidence reliance requires the applicable local RelianceDisposition; an actual named assurance claim requires its B.3 result for the same bounded assurance use. Apply A.6.9 (RPR-XCTX) when published wording such as “same”, “equivalent”, “align”, or “map” still hides the relation.
-
MG-DA ok. New or refactored tokens pass § 7 MG-DA (anchored head noun; collision check; an enumeration names its closed value set, classified kind, and classification rule; use
U.CharacteristicandCharacteristicSpaceonly when the enumeration is the declared CSLC scale of that exact Characteristic). -
Morphology ok. Suffix, compound, prefix, and casing respect § 8 LEX.Morph (for example, concrete
…SystemRolekind designations,MethodDescription,Work, and reserved prefixes); bare...Roleis not a default Tech form. -
Banned tokens absent or recovered. No process, practice, function, task, or activity in Kernel senses unless the sentence applies the selected recovery pattern (
A.3.4.P,A.6.F, work patterns, method patterns,C.36.P, or another relevant pattern) and names the recovered value by value; no tooling or file suffixes in Kernel tokens. -
State gating present (when needed). Readiness of an assignment to a system role is expressed through
SystemRoleAssignmentStateRelationplus a separateSystemRoleAssignmentStateAssertionwhen an assertion is needed, not through vague “approved” or “ready” wording.
Coordination map (how LEX plugs into the rest of FPF)
-
With E.10.D1 — Recovering What “Context” Means in Use. E10-CTX-1: When context changes the statement or next action, recover what supplies the boundary—for example, a value, relation, scope, scheme, situation, or use defined by the relevant pattern. E10-CTX-2: For source-local meaning, state the exact source, expression, effective scheme, and local sense. Create F.17 cells when stable addresses are needed; use F.9 only for an obtaining direct Bridge between two exact cells, and keep every proposed use and reliance claim separate.
-
With E.10.D2 (EntityOfConcern and Description-episteme boundary and specification use and refinement discipline). Distinguish the EntityOfConcern from its Description episteme, then say whether that episteme has specification use. E10-EOC-DESC-SPEC-1..3 apply: the EntityOfConcern is named directly; Description suffixes name Description-episteme use; Spec suffixes name specification use on a Description episteme; a work assertion or description is a separate
U.Epistemeabout one Work individual and is not the occurrence; a state assertion is a separately governed claim about a state and is not that state; an evaluation-result episteme is neither the evaluated object nor an occurrence. Upgrade a Description episteme to specification use only when checkable acceptance or another specification-granting gate named by value exists. -
With E.10.ROLE, A.2, A.2.1, and A.15 (system-role-kind, assignment, Method, and Work alignment). Bare role has no default Tech reading. A local system-role kind classifies independently admitted systems; classification, assignment, and Work remain separate.
U.Methodis a way of doing;U.MethodDescriptionis a claim-bearing episteme about one admitted method that passes A.3.2. A performed occurrence is datedU.Work: A.15.1 admits it first from its actual performer basis, Method, time, and containing System; F.6 follows only for precise assignment-bound attribution. A compact attribution account keeps the Work basis and all performers named or recoverable and may omit only an assignment identifier unused by the receiving claim after attribution is established. Assertions, descriptions, logs, and records about the occurrence are separate epistemes. -
With F‑cluster (Unification) and UTS (F.17). Recover each source-local cell under its effective scheme. A comparison row may cite direct Bridges, losses, contrasts, and separately warranted use claims; the row creates none of them. UTS is the human-readable term publication.
Acts and tokens. Identify the class of the token. For a mint, rename or use, check the conditions of the act against the policy for that class (§7.5).
Conformance checklist (LEX‑CC)
- LEX‑CC‑1 (Triggered use, not spelling ban). A listed word opens repair only when its FPF-governed use hides a kind, relation, claim, or action; quoted, ordinary, and already precise uses remain available.
- LEX‑CC‑2 (Context wording). After
E.10.D1, each action-changing use of context names what supplies the relevant boundary, interpretation, or situation and makes the next action or stop clear. - LEX‑CC‑3 (EntityOfConcern and Description-episteme boundary and specification-use morphology). Usage passes § 8 gates (suffix, prefix, and casing), EntityOfConcern and Description-episteme boundary checks, and specification-use checks.
- LEX-CC-4 (Cross-local relation). Recover the exact relation being asserted between local senses. Cite an F.9 Bridge only when its direct predicate obtains between identified cells; any proposed use and reliance remain separate. Shared spelling establishes none of them.
- LEX‑CC‑5 (MG-DA). New tokens pass MG-DA tests, including full‑text collision and Reserved‑Names checks.
- LEX‑CC‑6 (Service, acceptance, and evidence). Service or access wording establishes neither Work nor acceptance. An acceptance claim names the selected criterion episteme, any evaluation or acceptance Work that applied it, the returned value or result episteme, and the acceptance predicate with its participants under the applicable acceptance pattern; delivery Work, evidence, or a positive display proves none by itself. Evidence use is a relation over an Episteme, target claim or use, the scope, polarity, temporal conditions and provenance required by the evidence use.
- LEX‑CC‑7 (USM compatibility). Apply the token-class policy to the individual use or proposed range of uses (§7.5).
- LEX-CC-8 (Naming discipline). If overload cleanup yields one local replacement phrase, the text records the repaired phrase and applicable local repair pattern. If cleanup yields one durable reusable name, the text first records the applicable F.8 decision for the already identified value, then completes an F.18 NameCard; public, Core-facing, durable, or cross-local publication also supplies the F.17 term row. Intuition-first labels, partial NameCards, and naming acts that substitute for value or kind admission are non-conformant.
Worked micro‑examples (short, cross‑domain)
Factory.
✗ “The process failed; the service restarted itself.”
✓ The compact strings PLC_17#ObserverRole:PipelineOps, CAB_Chair#ApproverRole:ChangeControl, and OpsBot#DeployerRole:CD_Pipeline_v7 remain quoted source cues only. Start with E.10.ROLE; they are neither system-role-kind designations nor assignment occurrences.
PLC_17, ChangeControlBoardSystem, and OpsBot are independently admitted systems. Use ObserverSystemRole, ApproverSystemRole, and DeployerSystemRole only where those classifications matter. For each precise dated Work occurrence, recover the actual performer through A.13 and let A.15.1 independently admit the occurrence. Add F.6 only when the example or receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment; F.6 identifies neither assignment nor performer. A missing or failed attribution alone leaves independently admitted Work intact; recheck any A.13 or A.15.1 premise invalidated by the same finding under A.15.1:4.1. Omit only assignment identifiers that the receiving example does not use. Source suffixes and interpretation epistemes remain separate.
Keep the log episteme, approval speech-act content, and fulfillment claim separate from those Work occurrences, assignments, and attributions.
Cloud.
✗ “The process owner approved; the API service deployed.”
✓ The source strings ProductLead#AuthorizerRole:Rollout_2025 and sCG‑Spec_ci_bot#DeployerRole:CD_Pipeline_v7 remain quoted cues. Recover ProductLeadershipSystem and sCGSpecCIBot independently; use AuthorizerSystemRole and DeployerSystemRole only for exact local classification claims.
For RolloutApprovalWork-F123 and DeployWork-F123, recover each actual performer through A.13 and let A.15.1 independently admit each dated Work occurrence. Add F.6 only when this account or its receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment; F.6 identifies neither assignment nor performer. A missing or failed attribution alone leaves independently admitted Work intact; recheck any A.13 or A.15.1 premise invalidated by the same finding under A.15.1:4.1. Omit only assignment identifiers that the receiving account does not use. Keep source titles, source or scheme cues, approval content, deployment results, and telemetry separate, and identify each through the rule for that value or relation.
RESTAccessMethod : U.Method names the exact API access method; RESTAccessDescription is U.MethodDescription only if its claim content substantively describes that method as its EntityOfConcern, and it is an access Spec only after the named specification-use gate. Any selected description edition is reached through a separate governed U.EpistemeRef, while file or carrier version remains separate. FeatureAccessPromiseContent states the acceptance condition; telemetry measurements provide evidence for the evaluation that the deployment fulfilled that promise content.
Research.
✗ “Dataset X proves the theory; the process is reproducible.”
✓ DatasetX supports claim C only under the applicable evidence rule, with the model-fit scope, polarity, provenance and temporal conditions needed for that inference. Preserve its exact source and independently established support/use relations. Use A.10 when the source-to-use account needs explicit recovery or replay; use B.3, G.6, F.10 or another subject pattern only for the result that pattern defines;
replication is recorded through evidence-use, source-use, status-use, or reproducibility-status relations under their respective patterns;
procedure wording first recovers one exact U.Method; a separately identified claim-bearing episteme is U.MethodDescription only when that method is its EntityOfConcern and its claims pass A.3.2, while re-runs are exact Work occurrences only on the A.15.1 basis.
Semioarchitecture.
✗ “projection has one meaning in routing and bridge prose.”
✓ A.16 keeps projection as a move name for route-bounded partialization; F.9.1 keeps projection as a bridge stance label. If one durable reusable replacement name is needed, recover the value and use that need a name, record the applicable F.8 decision, and then use F.18 plus F.17 when public publication is current; otherwise retain the source expression or local phrase rather than flattening both interpretations into one umbrella rewrite.
Related lexical rules.
Apply section 7 MG-DA (anchored head nouns; enumeration rule gate; U.Characteristic and CharacteristicSpace only when one named Characteristic has that enumeration as its declared CSLC scale; collision checks) and section 8 LEX.Morph (suffix, prefix, and casing). The only admitted uses of plane in the Core are CHR:ReferencePlane and the derived operators CL^plane and Phi_plane; policy flags do not introduce new planes. To distinguish pre-operational and operational states within ReferencePlane=world, use WorldRegime in {prep | live}.
Guarded-head cross-reference (normative lexical caution)
When one wording head already carries several FPF-governed local interpretations, lexical cleanup should prefer a guarded-head note over silent flattening. The note may record that the head remains risky, name the cited texts or patterns that define or constrain the local interpretations, and point readers to the local canonical interpretation in each cited text.
If cleanup reveals that no admissible existing token can carry the needed meaning, use the local repair pattern for one-off wording. If the change needs one durable reusable name, first recover the value and its kind and record the applicable F.8 decision, then use F.18 for the NameCard and F.17 when public, Core-facing, durable, or cross-local publication is current. Do not invent an ad hoc synonym or let naming settle an unresolved object.
This cross-reference is lexical only. It does not create a new repair-side definition site, does not establish global or cross-local equivalence, and does not overrule cited local definitions. It keeps overloaded heads from being normalized into one false global interpretation.
projection is the main current example: A.16 keeps it as a move name for route-bounded partialization, while F.9.1 keeps it as a bridge stance label. E.10 therefore uses deconfliction notes and explicit naming of the cited text that governs each local interpretation, not one umbrella rewrite that erases the distinction.
Bare claim-bearing role is another guarded head and has no default Tech reading. Use E.10.ROLE to recover the current object or relation; use SystemRole only inside a concrete local system-role-kind designation after that kind is independently admitted.
control, management, regulation, governance, steering
These words locate a possible ambiguity; they are not prohibited words or interchangeable names. Read the complete claim and recover its subject and use. A field name, a group of people, a governing act, an observation and an arrangement that influences what happens can require different explanations despite sharing the same word.
Apply C.43:4.1.4 when the control meaning remains unresolved. It recovers what is influenced, the maintained condition or intended change, the information and means, and the connection to actual action. Operations management can realize this control construction without changing its disciplinary name. Retain its organizational conditions; do not infer a machine-like actuator, an automatic feedback loop or a fixed cadence from the word control.
Use the recovered subject directly: C.30.LCA for a control-structure claim, C.32.LCA for constructing its architecture, C.11 for the needed local choice, and the applicable authority, commitment, measurement or evidence rule for those relations. A.6.P helps when a direct relation still cannot be stated. Use F.18 only when a reusable name must be settled after its meaning is known. A translation retains the source meaning, including the difference between checking and influencing; a lexical match alone establishes no equivalence.
Do not force a control interpretation onto a field title or an already clear sentence. Repair only the ambiguity that changes understanding or use, then return to the work.
Thinking terms and the whole being considered
When a thinking term identifies the subject of a holonic question, name the whole and the relation that the reasoning needs. System thinking can consider how Systems are constituted and contribute to larger systems. Methodological thinking can consider how reusable Methods are constructed and composed. Reasoning about performed work can consider how actual Work occurrences and their parts form a performance history. Use A.1, A.3.1/B.1.5 and A.15.1 for their different construction rules.
These are qualified applications, not exhaustive definitions of the familiar forms of thinking. An “operational” judgement can also concern capacity, commitments and service under OPS.23. Management and strategic reasoning can connect Systems, Methods, Work and Epistemes. Mathematical thinking can compose claim-bearing knowledge and proof Methods while also constructing mathematical objects and operations. Keep an object of a mathematical claim distinct from the Episteme making that claim.
Translate the intended subject and operation, not just the adjective. Russian «методологическое мышление» concerns methods and their grounds in this use; English “methodical thinking” can instead mean orderly procedure. “Systematic” does not by itself mean FPF's “system thinking.” For the Work application, “reasoning about performed work” identifies what happened and whose contribution is at issue. Preserve a source's different use when quoting or explaining it.
Library.Reference:1.2 develops the common construction and its limits. Apply this distinction with F.19 when authoring or repairing a consequential use. The adjective alone supplies neither holon admission nor a part–whole, level or layer relation.
present, represent, representation, representative
Recover what is being shown, described, modelled, instantiated, sampled or acted for, and which relation the current sentence asserts. The shared word can conceal different questions. A presentation can be an act of communicating or its material; a representation can be an expression used to reason about a subject; a representative can be a selected case, a statistical sample or an authorized participant.
When the intended participants and predicate are already clear, keep ordinary wording and use the subject rule. For a remaining ambiguity, E.10.ARCH:3 recovers them. C.2.P supplies source, content and publication distinctions; C.37 supplies representation choice; C.29 supplies mathematical correspondence and returned inference. Kind membership, composition, statistical inference and acting under an assignment retain their own governing rules.
When a sentence relies on a “representative” example or fragment, recover the claim it supports and the basis of that support. A case chosen to illustrate an operation and a sample used for population inference have different grounds. Use C.39.QR when the unresolved question is what the obtained result establishes for the originating assignment; use C.39.WA when that assignment requires the still-missing explanation of a whole subject. Select a useful case by its contribution to the explanation, while preserving the general proposition and its grounds.
Optional detailed lexical checks (informative)
Use this section only after F.19 and the compact cue surface have selected a lexical, register, morphology, or naming question.
- Sections 5 and 6 distinguish lexical strata and Tech/Plain register use.
- Section 7 tests token generality, domain anchoring, local source meaning, and acronym use.
- Section 8 governs morphology, suffixes, compounds, aliases, and entry lexemes.
- Section 9 contains recovery rows for an already-selected overloaded FPF head.
F.18governs a durable reusable name after the value and use are settled.
Open only the subsection that can change the sentence. Repaired text or a blocker, followed by the local F.19 reread, is sufficient. Create a DRR or other preservation evidence only when a separate receiving decision requires it.
E.10 conformance prompts (normative, concept-only questions)
Use only the prompt selected by the compact route. Do not answer all prompts for every wording repair; § 7 and § 8 retain their distinct conformance uses.
- Local-use prompt. When context changes interpretation or the next action, has E.10.D1 recovered the exact source, scheme, scope, model-use structure, situation, frame, referent, or practice that matters?
- EntityOfConcern and Description-episteme boundary and specification-use prompt. Does each sentence use the correct boundary (the EntityOfConcern named directly; Description-episteme use for descriptions; specification use only where a direct gate pattern grants it; run: actuals)?
- Token prompt. For new or renamed tokens, is
LEX.TokenClassdeclared and consistent with where the token appears? - Head-kind prompt. Does the head noun name what the phrase is about—for example, a Method, Work, Characteristic, Capability, constraint claim, commitment, publication form, service-access relation, service-offer record, interpretation, local kind,
U.Transformation,TransformationFlowStructure, authority use, source or practice boundary, or another exact object? If bare role is claim-bearing, useE.10.ROLE; a concrete...SystemRolecompound is admissible only for an exact local system-role kind. A narrowing qualifier alone does not answer this question. - Qualifier-claim prompt. If an adjective, participle, genitive, or comparative modifier carries a claim being made, comparison criterion, relation, or admissible-use boundary, has that use been restored explicitly rather than left inside the modifier alone?
- Relation and representation prompt. Does the sentence state its ordinary relation and participants? If the distinction among an obtaining fact, reusable declaration, claim or report, and representation remains unresolved, use
E.10.ARCH; if it is already clear, use the exact subject pattern without inventorying the other branches. - Support interpretation prompt. Write the concrete relation first. Keep ordinary or quoted
support; otherwise use the direct subject pattern,A.6.Ponly for an unclear predicate or participant,A.6.RCDonly after both are clear but no current pattern supplies the rule, andA.6.6for basedness. SectionE.10:0.2c.18carries the examples; do not recreate its alternatives in the repaired sentence. - Comparison-basis prompt. Does the basis satisfy the comparison or condition rule used? Preserve the distinct relations or values when that rule combines them; require homogeneity where the rule needs it, after restoring the heads and qualifier claims.
- Morphology prompt. Do suffix, compound, prefix, and casing pass LEX.Morph gates (for example, concrete
…SystemRolekind designations,MethodDescription, andWork), with no default bare-RoleTech reading? - Promise, ability, access, and performance split. Are service promise or acceptance content, service-access relation, Capability (ability), and Work (performance) distinct and defined by their patterns?
- Plan and execution split. Are planning cues and
U.WorkPlankept separate from performedU.Work? For each actual performer, does A.13 supply the local kind and criterion, classification, same obtaining assignment, any scope, working situation and window required by the use, and adequate core evidence, with a profile only when conditionally consumed? Does A.15.1 then independently admit the Work from its performance history, Method, time, and containing System? When precise assignment-bound attribution is claimed, does F.6 separately relate that already admitted Work through the same assignment? Do affected-referent, binding, resource-use, assertion, and description claims remain separate? - Predicate-compatibility cue. When evidence, a document, publication, pattern, representation, or method is the grammatical subject of an agentive or causal predicate, read the complete span through
F.19, including negation and modality. Retain clear ordinary metonymy; otherwise state the capable participant's action or the exact non-agentive relation positively. Open A.1, A.13, A.15.1, or F.6 only when precise agency, Work, or attribution is part of the claim. - Bridge prompt. When the text asserts an F.9 Bridge between local senses, are the exact cells identified and does that relation obtain under its applicable profile? When a receiving use is current, does a separate C.2.1 claim state the proposed action, use direction, correspondence rule, tolerated loss and polarity? Keep reliance, assurance and any performed action separate and open them only when current. Other relations between local senses use their own defining rule.
- Collision prompt. Were full-text and Reserved-Names checks completed, with no other meaning of this token anywhere in FPF?
- Naming-procedure prompt. If one durable reusable name is needed because no admissible existing token carries the needed meaning beyond one local repair, was the governed value settled first, was the applicable F.8 decision recorded, and were the F.18 NameCard and any required F.17 public term row completed rather than picking a label by intuition or filling publication apparatus around an unresolved object?
- Value-substitution prompt. After the repair, can the declared reader tell what use remains admissible, and did the repair preserve usability, affordability, semantic composability, fit with the rule governing the claim, and local action guidance? If not, narrow the repair, keep ordinary wording with a recovery note naming the recovered kind and use, or leave the issue blocking instead of optimizing for lexical purity.
Working order for precision repair on FPF-governed prose. Use the precision-before-coarsening rule in F.19:4; the selected lexical prompts here locate the unresolved head, qualifier, or comparison question.
Archetypal grounding: three examples (informative)
These examples show the repaired claim first. Open the exact Work, attribution, publication, or naming apparatus only where the receiving use needs it.
Healthcare (operating-room planning and work)
Source sentence: “The surgical process is scheduled at 08:00; the SOP approves the incision and the service documents recovery.” For this constructed case, the planning material identifies Patient_221, IncisionMethod, the 08:00 PlanItem and its reservations; the applicable policy supplies the approval and gate conditions. An approval record identifies ApprovalSpeechActWork-221 performed by QAApprovalSystem, and the service means the ward team.
Repair, taking service here to mean the ward team: “Schedule Incision_221 in OR_Case_221_WorkPlan for 08:00, using IncisionMethod. SOP_OR_v4 states the incision-readiness constraint. QAApprovalSystem performs the approval; record its speech-act content. Separately recover the applicable A.21 profile and required check results. Its GateDecisionResult states whether and under which conditions the planned run may proceed. The ward team records Patient_221's recovery in RecoveryRecord_221.”
Formal plan membership. OR_Case_221_WorkPlan is used as U.WorkPlan only after A.15.2 membership is established: its already identified present EntityOfConcern is Patient_221, its horizon is the bounded surgical-planning interval, and its PlanItem, Incision_221, substantively coordinates the intended surgeon classification and assignment conditions, operating-room resource reservation, planned start of 08:00, the U.Method IncisionMethod, and the incision-readiness target.
The WorkPlan cites IncisionMethodDescription, a separately identified claim-bearing episteme. That episteme is U.MethodDescription only because the method is its EntityOfConcern and its claims substantively describe how the method is carried out.
Any edition identity needed by the plan is selected through a separate U.EpistemeRef whose subject pattern supplies its rule; carrier version remains separate.
Actual approval Work. For ApprovalSpeechActWork-221, recover QAApprovalSystem as the actual performer through A.13 and let A.15.1 independently admit the dated Work. Add F.6 only when this account or its receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment; F.6 identifies neither assignment nor performer. A missing or failed attribution alone leaves independently admitted Work intact; recheck any A.13 or A.15.1 premise invalidated by the same finding under A.15.1:4.1. Add ApproverSystemRole only when that classification matters.
Separate source claims. If the source also specifies a monitoring promise or an access method, recover those additional claims separately: PostOpMonitoringPromiseContent states the promised monitoring and its vitals acceptance envelope. WardAccessMethod : U.Method names the exact access method; WardProtocol is U.MethodDescription only if it is a separately identified claim-bearing episteme about that method and passes A.3.2, while its publication form and carrier remain separate. These are additional claims, not replacements for recording recovery.
Manufacturing (assembly line)
Source sentence: “The welding function provides air-tight seams; the process costs 3 min.” In this constructed case, the available capability assessment identifies Robot_SN789 and its qualified Weld_MIG_v3 conditions; the run observations identify WeldWork-SN789-4711, its three-minute duration and the affected joint.
Repair: “Robot_SN789 can execute Weld_MIG_v3 within envelope E at measures M. The run WeldWork-SN789-4711 lasts three minutes and changes the workpiece joint. Use the recorded measurements to check the duration and the seam's air-tightness acceptance criterion.”
For one run, recover Robot_SN789 as the actual performer through A.13 and let A.15.1 independently admit WeldWork-SN789-4711 from its performer, time, Method, and containing-System facts. Add F.6 only when this account or its receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment; F.6 identifies neither assignment nor performer. A missing or failed attribution alone leaves independently admitted Work intact; recheck any A.13 or A.15.1 premise invalidated by the same finding under A.15.1:4.1. Add WelderSystemRole only when that classification matters.
Each bounded change of the workpiece joint is identified under A.3.4 before stating a work-to-change fact. If the Work first constitutes a distinct seam entity, A.15.PROD supplies its identity specification and inception boundary. Measurement-result epistemes remain separate evidence for acceptance and duration claims.
Treat source string WeldingCellContext as a quoted recovery cue. If it changes the claim, recover the exact source edition, plant practice, effective scheme, scope, or working situation that it denotes. When an assignment is being recovered, identify its participant meanings and values from its declared A.2.1 species and recover its uninterrupted obtaining interval.
Cloud and SRE
Source sentence: “The storage service wrote logs and the deployment process failed after 2 min.” In this constructed case, the event records identify sCGSpecCIBot's deployment attempt and LogWriterSystem's logging occurrence, including the deployment's failure after 120 seconds.
Repair: “sCGSpecCIBot performed DeployWork-r4711, which failed after 120 seconds. LogWriterSystem performed LogWritingWork-r4711.”
Admit each Work occurrence through A.13 and A.15.1; add F.6 only for a current assignment-bound attribution. Treat sCG-Spec_ci_bot#DeployerRole:CD_v7 as a source expression and recover CD_v7 separately. Use DeployerSystemRole or TransformerSystemRole only when the classification changes the claim.
Separate source claims. If the source also specifies durability and availability targets as a storage promise, recover them in ObjectStoragePromiseContent. If it supplies an access-method description, identify that separate episteme as S3_API_Spec_vX and preserve the method it describes.
If the source supplies only an intended cycle time, preserve that intended bound rather than assert a measured run duration. If it does not identify a performer, approval basis or particular occurrence, retain the supported statement and leave that dependent claim unresolved.
Bias-Annotation
Scope: FPF-governed wording-use repair; ordinary ungoverned language remains outside this pattern.
Conformance checklist
Use this checklist when changing an E.10 rule, a governed token, or a durable lexical decision. Ordinary sentence correction returns repaired text and needs no recorded checklist.
- The cue is treated as a locator, not a ban or verdict.
- The complete natural span has been read through
F.19; an ordinary repair stops there. - Any deeper route is the smallest pattern that owns the unresolved FPF kind, relation, source use, register, morphology, or name.
- The replacement preserves the intended object, kind, relation, scope, and action and does not substitute another umbrella head.
- The accepted text contains the concrete sentence or selected pattern result, not a menu of possible interpretations or a mandatory rejected overread.
- The changed sentence and its meaning-dependent neighbors pass the local
F.19reread. - For a reusable way of doing, the explanation uses method as its main Plain name under §0.2c.21a while preserving useful contextual synonyms and established expressions. Different words do not create different objects; any Method, Work, community, discipline or description distinction follows from the claim. Ordinary synonyms do not require separate Plain-Twin registration.
- A contribution/contributes phrase leaves its participants and the asserted action, content use, evidence use or comparison recoverable under §0.2c.13a; the word does not substitute for that statement.
- Goal, purpose, Method concern, expected effect and work-product wording preserve the claims distinguished in §0.2c.7–7a. The text states the wanted, planned or actual meaning being used, and supplies no participant, binding, produced entity, completed work or successful use from a label alone. Missing facts remain distinct from a missing defining rule.
Common anti-patterns and how to avoid them
Consequences
Positive consequences:
- Wording repair becomes ontology-first precision restoration rather than taste-based editing.
- New names, field names, and pattern prose stay composable with FPF kinds, slot discipline, and the patterns that define their claims.
- FPF can admit ordinary prose, source quotations, and local names without letting them become hidden ontology.
Costs:
- A quick lexical replacement often becomes a short ontological check.
- Some attractive phrases remain blocked until the relevant rule, relation, bearer, value set, or admissible use is named.
- Broad source wording sometimes needs a precision-restoration pattern rather than a one-word replacement.
Rationale
Wording mistakes in FPF usually matter because they hide an ontology choice, relation and participants, declaration slot, participant designation, representation correspondence, source-use relation, admissible-use boundary, or applicable pattern. A synonym replacement can make the sentence smoother while changing the claim. E.10 therefore starts with the connected F.19 reading. Only an unresolved FPF wording question opens a cue and the smallest pattern contribution needed to recover it.
The lexical rule content stays limited. It is not the ontology for evidence, assurance, work, gate, decision, publication, architecture, characteristic, temporal, role, method, mathematical-lens, or source-use claims. It only prevents wording from smuggling those claims in under broad heads. Once the recovered object or relation is visible, apply the relevant pattern or continue the ordinary task. Name an exact subject assertion or ClaimGraph only when truth, action, comparison, publication, reuse, or reliance depends on that identity; a PatternID may remain an ordinary citation. The fuller form System S used Method M while performing Work W remains correct when actor, Method, and Work identity are part of the claim, but it is not the default paraphrase of apply pattern P.
The conformance prompts are bounded so lexical governance does not become a corpus-wide purity ritual. A local repair should restore composability, reader action, and admissible use; when it cannot do that, the result is quote-only use, reduced-use cue, blocked use, or an incomplete rewrite rather than a more polished umbrella word.
SoTA-Echoing - lexical governance
Practice question. When wording carries an FPF-governed claim, what is the least costly authoring discipline that either closes one local repair or reaches the rule for the exact object, relation, claim, or use without turning a glossary into a second ontology?
Selected best-known line. Read the complete span through F.19, use a cue only for an unresolved FPF wording question, recover the governed object or relation before renaming, apply the direct defining or testing rule, preserve the remaining reader use, and stop. This line combines the current FPF precision-restoration patterns with established terminology and controlled-vocabulary practice only for the objects those sources govern.
Alternative or default. Synonym substitution, a central preferred-word list, and copied local warning catalogues are cheaper to start. They can leave the obtaining relation, participant, claim-bearing episteme, admissible use, or direct owner hidden while making the sentence look more precise.
Defect overcome and comparable effort. Clear ordinary wording exits after one scan. A load-bearing phrase takes one bounded recovery and one direct-pattern use; it returns an actionable claim or an explicit blocker. Compare its effort with searching the relevant local glossaries for the repair. The cost is a reduced-use or blocked result when the object cannot be recovered.
Mutation and source roles. The internal FPF row supplies the governing local architecture; ISO and W3C terminology or vocabulary sources constrain object, concept, designation, label, and publication uses; language-scent and accessibility sources inform cue preservation and discoverability. None of the external sources assigns FPF kinds or relation truth. The mutations are the trigger registry, direct-owner exits, phrase-level rewrite rules, entry projections, and Relations named in the rows below.
The intended practical result is clearer action guidance and semantic composability. Reopen only the internal rows whose defining pattern changes its kind, authority boundary, or recovery result; reopen the in-situ-cue choice only if stronger language-scent evidence changes cue usefulness for the intended readers; reopen the terminology, vocabulary-publication, or discoverability rows only if the cited standard changes the object, designation, label, publication, or navigation contribution used here. A source locator, publication-status, popularity, or unused-example change alone does not reopen E.10.
E.10 regression cues (concept-only “diff” triggers)
Re-review your prose when any of these happen:
- A source edition, effective scheme, or local meaning changes → re-affirm any Tech-and-Plain twin mapping and first-use gloss; recheck an F.9 Bridge, use claim, reliance claim, or acceptance wording only when that changed value participates.
- A system-role-kind or other governed name grows through coordination → use
F.19to ask whether a series is needed, then apply MG-DA only if the name still hides one kind, alternatives, a relation, or a bundle. Two members may already be excessive; a long required designation may be sound. Bare role first usesE.10.ROLE. - A slash,
and,plus,&, repeated pair, nested list, or modifier chain appears in FPF-governed wording → read the natural span throughF.19before editing the mark. Retain conventional notation and required sets. If an FPF kind, alternative, direct relation, declaration, or representation remains hidden, use the selected E.10 or subject route; do not infer the defect from the character or item count alone. - A
servicestatement broadens scope → use L-SERV and A.6.P:4.11a to recover the hidden subject, relation, and receiving use. Update only that recovered claim; do not apply a fixed reading list or rewrite every nearby service-related claim. - Recipes gain or lose steps -> first recover the exact
U.Method, the claim-bearing episteme, and the changed claim. UpdateU.MethodDescriptiononly when that episteme has the method as its EntityOfConcern and passes A.3.2; a code, diagram, recipe, procedure, or document-form change remains under the pattern for that representation or publication unless claim content changes. Never move the change into service labels or system-role-kind names. - An agentive or causal predicate is attached to a doubtful subject → read the whole span through
F.19, including negation and modality; retain established metonymy or state the capable participant or exact non-agentive relation positively. - A verb or relational noun loses a needed participant → restore the object, destination, source, bearer, or other operand when one value is not cheaply and uniquely recoverable; open a deeper FPF route only if that participant is governed.
- A generic head or support-headed compound acquires an FPF claim or bounded use → restore the head kind first. If
supportstill hides a direct predicate or participant, use the selected E.10 route; otherwise state the concrete sentence and apply its subject pattern. - Wording about a way of doing or a neighboring object changes → use the selected method-side rule in
E.10:0.2cor section 9 to recover the exact object or relation. Do not replace one umbrella withmethod,practice,mechanism,algorithm, orworkflow. - A declarative representation starts to sound imperative → recover the represented object and correspondence through
C.2.P.DRor the exact subject pattern, withC.29for mathematical-lens use. Treat a path, table, dashboard, formula, or pattern relation as action only when an action claim is established. - New token minted → ensure
LEX.TokenClassis declared and perform collision checks. If an enumeration is current, name its closed value set, classified kind, and classification rule; add aCharacteristicSpaceonly when the enumeration is the declared CSLC scale of one namedU.Characteristic. - Suffix drift (e.g.,
…Workon a plan) → fix via LEX.Morph. - A label is reused across local sources, practices, or schemes → recover each local meaning. Keep them distinct unless an F.9 Bridge obtains between exact cells; state any proposed use and reliance separately.
- A guarded head needs a new label → prefer a guarded-head note first; if no admissible existing token remains for one durable reusable name, settle the value and its use, record the applicable F.8 decision, and use F.18 plus an F.17 row when public publication is current.
Quick card
Read the complete natural span with
F.19. Use E.10 cues to notice an overloaded head, doubtful agentive or causal predicate, ungrounded contrast, missing operand or referent, grouping or list pressure, loaded qualifier, or source/representation ambiguity. Try the wording against the reader’s original task. When the claim and its grounds are clear enough for that task, keep or repair the text and stop. Where its subject, composition scale, model distinctions or entrusted choice shift, return through F.19:4.3.1.1. Open a subject rule for the distinction still needed. The result is repaired text or a blocker, never a mandatory form or catalogue of possible meanings.
- Use sections 5–9 only for the selected register, token, morphology, or overloaded-head problem.
- Use
E.10.ARCHonly while fact, declaration, report, and representation remain confused; use the subject pattern directly once the claim is clear. - Use
F.18only for a durable reusable name after its value and use are settled. - Treat every trigger as a cue rather than a spelling ban. Two coordinated members may already be needless; a long required set may be sound.
Closing notes (governance and purity)
- Notation-agnostic.
E.10is a wording-use governance pattern, not a scanner or template. Apply it in prose, sketches, or formal models. - Where checks belong. Convenience checks belong to Tooling;
E.10stays notation-agnostic. Conformance code belongs in SCR-LEX or RSCR-LEX as referenced above. - Acts and tokens. Determine the token class and test whether its policy allows the proposed LexicalAct in this setting (§7.5).
- Reserved “plane”. Only
CHR:ReferencePlaneuses the bare word plane. E.10.D2 is the EntityOfConcern and Description-episteme boundary plus specification-use gates, with publication faces, publication forms,PublicationUnits, carriers, and renderings kept separate; all other category talk is expressed as Characteristics in a CharacteristicSpace when scale semantics are declared.
One-line memory: “F.19 reads the claim; E.10 opens only the unresolved lexical branch.”
Relations
- Builds on:
F.19for whole-span semantic and pragmatic repair;A.7,C.2.1,E.17,E.24,A.6.0,A.6.5, andF.18for the distinction or durable name that remains after that repair. - Coordinates with precision-restoration patterns:
E.10.ARCH,E.10.DEVfor unresolved development or evolution wording,E.10.LRN,E.10.ROLE,A.6.P,A.6.P.WMR,A.6.RCD,C.2.P,A.19.SPR,E.10.MOVE, and the direct domain restoration pattern applicable to the current trigger; useA.15.PRODfor local production-work, entity-inception, and production-completion claims when that result family is current. - Coordinates with
E.10.D1: use it when context hides subject-defined content that changes the next action—for example, a scheme, scope, model-use boundary, situation, design-time or run-time referent, architecture, environment, domain subject, or local-practice use. - Coordinates with patterns for these objects: architecture, transformation, Work, evidence, assurance, gate, publication, source use, mathematical lens, Characteristic, temporal claim, exact local system-role kind and classification,
U.SystemRoleAssignment, Method, and direct relation when those claims are current. - Use: the concrete pattern for the recovered object, relation, evidence, authority, work, publication, or admissible-use claim whenever the issue is no longer wording precision.
E.10:End
Recover What “Learning” Means in This Claim
Type: lexical and ontological precision restoration (E)
Status: Stable
Plain name: recover what “learning” means here
Use This When
Use this pattern when claim-bearing wording says learn, learning, learned, taught, or trained, and the sentence does not yet reveal which participant, changed subject, Work, Method, result, evidence, or receiving use is meant.
First useful result. Rewrite or split the sentence so that each exact subject claim is recognizable, then continue with the direct pattern for that claim. A local repair normally ends with a recovered claim or ordinary non-use. It does not end with a generic LearningResult.
Cheap exit. Keep ordinary wording when no FPF inference or action depends on which sense is meant. If the changed subject, operation, result, evidence, and direct pattern are already explicit, use that pattern directly.
Not this pattern when. Use the relevant subject practice to choose a teaching or training Method, assess a capability, fit a statistical model, design an experiment, perform inference, measure a result, or decide what to do next. Use E.10.LRN to identify the claim that calls for that practice.
LRN remains in this PatternID because the ambiguous source wording opens the recovery. It is not a Tech designation for one governed process or value and is not a UTS-row precedent.
Problem Frame
The same word family is used for different situations:
- a person inquires, notices, remembers, or reorganizes an episteme;
- another System performs teaching, coaching, demonstration, or feedback Work;
- a person acquires a capability for later Work;
- a computing System performs parameter-estimation or optimization Work using an algorithm and returns a fitted model;
- an inference Method returns a posterior or approximate distribution;
- a query policy acquires data or reduces uncertainty;
- a probe decodes a representation from system-side phenomena;
- an organization or culture retains, reconstructs, selects, or loses a practice variant;
- a course, lesson, dataset, artifact, or guide is produced; or
- ordinary prose says only that someone found something out.
These uses can occur together without becoming one process. A teacher can teach while no capability is acquired. A person can acquire a capability without the sentence identifying a teacher. A model can be trained without a deployed System satisfying its capability claim. A posterior can be approximated without any human or machine capability changing.
Problem
When the umbrella word remains load-bearing, authors transfer evidence and conclusions from one subject to another without naming the change. A lesson is mistaken for a capability, an artifact for proof of authorship or transfer, and a benchmark improvement for generalization. Information gain is mistaken for competence gain, and model fitting for inference. An organizational slogan stands in for changed practice, and a repeated pattern for a universal atom of knowledge.
The repair must preserve familiar language while preventing those transfers. The direct subject patterns define capabilities, Work, Methods, epistemes, evidence, models, representations, cultures, and decisions. This pattern recovers the claim to which they apply.
Forces
Solution
Recover the current claim from its participants, subject, operation, result, and use rather than from the word learning.
- Bound the wording use. Locate the expression and read as far as the question or decision that gives it its meaning. An example of what one person did can supply evidence in a question about the team or its practice. Keep the requested subject and reach in view.
- Recover the grammatical commitment. Identify who or what is said to have taught, trained, learned, changed, produced, inferred, or acquired something. Grammar foregrounds a claim but does not fill missing causal participants or evidence.
- Name the changed subject. State whether the current claim concerns a named holder's capability, an episteme, a model and parameters, a probability distribution, a representation relation, an organization or population, a product, a Work occurrence, or another exact subject.
- Separate Work, Method, and result. Name inquiry, teaching, practice, training, optimization, inference, experiment, data acquisition, assessment, publication, or cultural-continuation Work only when it is current. Keep its performer, Method, inputs, and dated occurrence separate from the result attributed to another subject.
- State the evidence and any material transfer boundary. Name what was observed or assessed and the task, population, configuration, window, support arrangement and use conditions required to interpret that evidence or rely on the claim. Identify a stronger reading as unsupported only when independent local grounds make it plausible and the distinction changes use.
- Select the direct branch for each claim. When the decision requires several claims together, explain the connection that matters. A lesson was delivered and a team subsequently performed better. To attribute the improvement to the lesson, establish how the taught practice could change that team’s work and what supports the attribution.
- Stop after recovery. Return the repaired claim and direct pattern, or an exact missing-information, missing-governor, quote-only, ordinary-use, or blocker result. Do not create a generic learning record, role kind, process, progress scale, or causal relation.
Direct branches
Grammar does not settle the route
Passive and active grammar can also shift attention without changing ontology. “Was taught” foregrounds an intervention received. “Learned” foregrounds the attributed result. Neither wording alone establishes the full causal route.
Construction, public results, and patterns
Constructivism and constructionism remain source-local theories and Method families. Constructionism shares the learner-side construction emphasis and adds a design commitment to making something public or shareable. This commitment guides FPF Method and evidence design. A model, program, explanation, dance, diagram, or other external construction can expose distinctions and Methods for inspection.
The construction does not prove capability, authorship, independent use, transfer, or retention. Examine what the participant chose and performed in the construction, with the help then available. It can supply the needed evidence if it already exercises the claimed work under the relevant conditions. Use an adequate earlier performance or a further task for the capability, transfer or retention question still open. Reciting a poem or reproducing a pattern may establish a narrow remembered performance; it is not automatically action-guiding capability in a new situation.
An FPF pattern is an action-guiding episteme. In a named cultural case it can also be a retained, transmitted, selected, or reconstructed cultural variant under C.36. It is not the universal atomic unit of knowledge, a “meme” kind, the holder's internal state, or the holder's capability merely because it was copied or published.
Lightweight local result
Suppose a training record identifies a minimax optimization run, its resulting generator G-17, and sample evaluations T1–T3. The record supports this clarification.
No persistent record is required unless another use needs to inspect or reuse this recovery.
Worked Slices
Taught to swim and learned to swim
“A coach taught Lee to swim” opens a teaching-Work claim. Recover coach and Lee as Systems, the dated coaching and practice Work, enacted Method, conditions, and intended result.
A supervisor needs to know whether Lee can swim one length of this pool unaided today. A recent observed crossing can settle that question if the stroke, distance, support and conditions match. To say that Lee learned to do it during the course, establish what Lee could do at the start. A prior swimming record can supply that comparison. A successful crossing after coaching is compatible with Lee having been able to swim the length beforehand.
To credit a particular teaching Method with the change, examine the alternatives that could explain it under C.28. If a new aid and coaching were always used together, evidence can support their combined use while leaving their separate effects open. An assignment in open water needs grounds for transferring the pool result to those conditions.
GAN training
A GAN training run is optimization Work over generator and discriminator parameters under a declared objective and data basis. The run, trained model editions, generated samples, benchmark results, and a deployed system's capability are separate. Calling all five “what the network learned” loses the decision-relevant boundaries.
Variational inference
A variational-inference procedure selects an approximate distribution from a declared family by optimizing a divergence or bound against a target probabilistic model. It returns an inferential approximation and diagnostics. It does not establish capability acquisition.
Active learning
An active-learning policy chooses a query or observation because its expected result may improve a model or decision. Recover the current model state, acquisition option, expected information or decision value, cost, returned observation, update, and next decision separately. The data-acquisition choice is closer to inquiry or mining in this branch than to a claim that a holder acquired a capability.
Public construction
A participant builds and explains a working pump simulation. The artifact and explanation make some model choices inspectable. Assessment Work may use them as evidence for bounded claims.
Independent diagnosis of a changed pump configuration is a different representative task. Use adequate existing performance evidence for that capability or transfer claim. If its relevant conditions remain untested, a warranted test under the declared support conditions can supply the missing evidence.
Organizational learning
An organization publishes a “lessons learned” report. This establishes an available episteme. A supported claim that a new variant entered an institutional memory may finish at that narrower result.
A claim that people changed their practice needs the relevant Method, assignment and Work facts and evidence of the claimed change. Recover selection, retention or operating consequences only for the cultural claim that uses them. Publication alone does not establish changed enacted practice.
Bias Annotation
- Human-default bias: do not assume every use concerns a person's capability.
- Machine-anthropomorphism bias: do not translate model fitting into human cognition or agency.
- Substrate bias: biological and computing substrates can support several kinds of Work and result; substrate does not choose the branch.
- Artifact bias: public construction improves inspectability but does not prove authorship, capability, or transfer.
- English-grammar bias: active/passive voice and English lexical convention do not establish causal structure.
- Metric bias: a changed score or loss does not identify which subject changed or which stronger claim it supports.
Conformance Checklist
- Is the word family claim-bearing for the current use?
- Are participants, changed subject, Work or Method, result, and receiving use explicit enough to choose a direct pattern?
- Are teaching/training occurrences separated from capability, model, inference, or generalization results?
- Does the evidence identify the task or population and the conditions, support arrangement, window and locally grounded transfer boundary required by the claim or receiving use?
- Are information acquisition, belief/model update, statistical fitting, and capability change kept distinct?
- Are public constructions, recall performances, patterns, and cultural variants kept distinct from holder capability and internal state?
- Is source-local wording preserved as quotation where needed without becoming FPF ontology?
- Did the repair avoid a generic
Learning,Learner,Teacher,LearningProgress, orLearningResultkind or UTS row? - Does each recovered claim reach its direct rule, with the connection needed when several support one conclusion?
Common Anti-Patterns and Repairs
Consequences and Reopen Condition
Benefits. Education, human capability, machine learning, inference, inquiry, representation, organizational, and cultural claims can share readable prose without sharing false identity. Evidence stays attached to the result it supports. Downstream DPFs can specialize Methods without inheriting an ambiguous FPF process.
Costs. A load-bearing umbrella use needs one bounded recovery, and some sentences must be split. Domain methods and evidence still need their direct products.
Reopen this pattern when a recurring claim-bearing use cannot reach one direct subject pattern, ordinary non-use, or exact blocker with the recovery fields above; when cold readers still transfer evidence among branches after the repair; or when a direct pattern absorbs the same entry, action, first result, and stop without losing discoverability.
Rationale
The recurring transdisciplinary problem is lexical recovery, not a common learning substance. The same bounded action applies across these cases: recover who or what changed, separate Work and Method from result, state evidence and any action-changing transfer boundary, split unlike claims, and route each claim to its owner. That action changes practice while leaving every substantive ontology and Method with its direct pattern.
Familiar spelling is insufficient for one UnifiedTermRow. A durable public row is considered only for an independently governed value after its own naming and use tests. The umbrella word creates neither that value nor a Bridge among the branches.
SoTA Echoing
Recheck a source line only when a newer edition changes a distinction used by this Solution or when contrary evidence shows that the current branch boundary changes a practitioner action. Recency alone does not merge branches.
Relations
- Selected by:
E.10andE.10.ARCHwhen learning-word recovery is current. - Builds on:
F.0.1,F.0.2,F.1,F.17,F.18,C.2.1, A.3, A.15,A.2.2,A.10,A.6.3.RT,C.16,C.29, andC.36. - Returns to: the direct human-capability, Work, Method, statistical-model-fitting, inference, experiment/data-acquisition, representation, product, publication, organizational, cultural, evidence, or decision pattern selected by the recovered claim.
- Keeps outside: one generic Learning process, learner/teacher role kinds, teaching or training Method selection, capability assessment, model fitting, inference, experiment design, evidence qualification, causal explanation, and choice.
E.10.LRN:End
Recovering What Interest or Curiosity Means Here
Type: lexical and ontological precision restoration (E)
Status: Stable
Problem frame
Use this when a claim that something is interesting, in someone's interest or driven by curiosity guides a choice of work but leaves you unsure what to do. You may need to select an experiment, retain a promising construction, design an agent's exploration policy or understand why participants want different outcomes.
Begin by saying whose interest is involved and what the sentence proposes. “This experiment is interesting” might mean that its result would distinguish two models, that a participant wants to try it, or that the resulting technique could support further experiments. Each meaning can guide useful work.
The first result is a clearer sentence and a next action that follows from its meaning. Keep ordinary wording when its meaning is already sufficient. A musician expressing delight and a contract stating an interest rate can continue their respective work without this recovery.
Problem
The same word connects several different questions. An informative observation changes beliefs. A learning-progress signal compares what a learner can do. A promising stepping stone offers further possibilities, and an affected participant has something to gain or lose. In a joint project these relations often coexist.
Their differences matter when a phrase becomes an instruction. Asking an agent to maximize “surprise” may reward unpredictable observations instead of observations that help it distinguish its hypotheses. Requiring a promising idea to demonstrate progress toward a distant goal may discard the very construction from which another goal could emerge. Selecting an “interesting rhythm” leaves the choice unresolved until the listener, performer or musical use is known.
Forces
Solution
Recover the relied-on meaning, then continue the work that gives it value.
- Locate the claim that matters. Take the sentence together with the decision or activity it is meant to guide. Ask what would change if the object were judged interesting. If nothing depends on that judgement, keep the expression as it stands.
- Name the participant and relation. Identify who is engaged, affected, learning or choosing. Identify what draws attention, what might change, or what the participant stands to gain or lose. Use the distinctions in §4.1 only where they resolve the sentence.
- Preserve combined meanings. If a participant enjoys an experiment and the experiment could distinguish two models, say both. Let the corresponding motivation and inquiry methods supply their different contributions.
- Restore a needed comparison. When a score or optimization rule is proposed, name what it compares and under which model, experience or resources. Use §4.2 to resolve ambiguous appeals to surprise, progress or a Goldilocks region.
- Return to the action. Rewrite the claim so its recipient can choose, inquire, practise, retain a result or explain a concern. Continue once that meaning is sufficient. A shared decision can keep the clarification in its existing account. An ordinary sentence needs no additional form.
Recover the contribution being claimed
These are recurring uses of the wording. Select the ones needed for the sentence; several can apply together.
An information-seeking policy can be attributed to a human, an AI agent or a collective whose members obtain and use the information. For a claim about a person's experience, preserve that experiencer. A policy description alone leaves the question of subjective experience open.
A prospective continuation can be supported by a hunch, a newly available operation or a promising variation. Say how strongly that basis supports the continuation. Where its future destination is unknown, the next action can be to explore or retain the construction. C.18 supplies retention in an exploration archive; C.19 helps choose which directions in an active search pool to continue, retain or stop; C.40 helps develop problems and ways together.
Recover the model behind the comparison
For surprise, identify the quantity first. The surprisal of an observation, -log p(observation), concerns its probability under a selected model. Bayesian surprise concerns the change from prior to posterior beliefs about chosen variables. Before choosing a query, expected information gain averages what its possible outcomes would tell the agent about the variables of interest, using their probabilities under the model. A learning-progress signal compares successive predictive or constructive performance under a learning procedure. Those comparisons can disagree, as §5.1 shows.
In active inference, distinguish updating beliefs from selecting actions. Predictive coding uses prediction errors to update beliefs about hidden states and, in learning formulations, model parameters. Policy selection also depends on the information and outcomes expected from action. Recover the prior preferences when a source explains avoiding an outcome through its expected surprise. The source's account may couple epistemic and pragmatic value. Preserve that coupling together with the quantities and assumptions that give it meaning.
For a Goldilocks region, recover what varies, for which agent and use, and over which range. It might concern an attainable learning challenge, a listener's response to syncopation or a combination of rhythmic properties. Use the applicable account of the relation between those characteristics and the desired result. The useful region can move as the agent learns or as the task changes. In a search algorithm, a difficulty screen can select candidates for learning while novelty and transfer still govern which possibilities are retained.
When a quantitative answer is needed, C.16 supplies measurement and C.17 separates novelty, usefulness and sample surprise. Keep an adequate qualitative comparison when it already settles the action.
Combine contributions in a development project
A project can use information-seeking to distinguish alternatives, learning progress to choose a promising practice opportunity, and novelty or diversity to retain further possibilities. State the contribution of each rule to that project.
For example, a group can test the controller in §5.1 during a maintenance window. The researcher calls input 1 interesting because its response distinguishes two candidate relations. A colleague sees the query construction as material for other unknown interfaces. The maintenance team needs the controller returned to service before the window ends.
Recover their claims as three instructions: “Query input 1 to choose the command”; “Keep this construction available for exploration of other interfaces”; and “Schedule controller work within the available window.”
If the relation is supplied, the query becomes unnecessary. The potential use of the construction remains, and the window still constrains any remaining controller work.
C.11 compares formed options when a choice is needed. C.19 helps decide which directions in an active search pool to continue, retain or stop. A single agent's information-seeking policy uses the applicable subject method; choosing among formed queries can use C.11. C.40 supports the development of problems and ways. The field's methods supply the experiment, training activity or construction itself.
Archetypal Grounding
An informative controller query
A controller must choose an input that gives output 5. It has two equally plausible candidate relations, y=x and y=-x, and can first query input 0 or input 1. “Choose the more surprising query” leaves the criterion unclear.
At input 0 both candidates return 0. At input 1 they return 1 and -1 respectively. The second query distinguishes the candidates: after observing 1, command 5; after observing -1, command -5. The recovered instruction is “query input 1 to distinguish the relations, then choose the command.”
A separate fair coin produces an outcome with one bit of surprisal, but its outcome is independent of which relation holds. Querying input 1 yields one bit of information about that relation under the stated equal prior. Tossing the coin yields none. If the controller is already given the relation, it can select the command immediately.
The example constructs an information-seeking policy. Whether a particular controller can execute it is a further capability question.
Progress and a stepping stone
In a constructed example, a compressor learns a dictionary entry for a repeated substring in a fixed data block. Its total lossless description of that block, counting the dictionary and encoded data under the same convention, falls from 100 bits before the update to 80 afterwards. “The second version is more interesting” can now be recovered as “this dictionary update reduced the total description by 20 bits for this block.”
If the team is choosing more training, it considers the likely further improvement and the work required.
Another construction may deserve attention even though that measure has not improved. In Stanley and Lehman's Picbreeder example, a participant retained and developed an image resembling an alien face; a car-like image appeared through later changes. The first participant could act on the available image and its variation possibilities before knowing that destination. For further exploration, retain the construction and its means of variation. C.18 governs an archive used for that purpose.
These cases permit different choices: improving a known performance measure and preserving a possible starting point for new work.
Choosing an interesting rhythm
A workshop needs music for listeners who are familiar with 4/4 and are beginning to work with 7/8. “Choose moderately complex rhythms; they are the most interesting” hides both the response sought and the listeners' preparation.
Ask the facilitator what the material should help these listeners do and whose response matters. Suppose the workshop aims to help them maintain the beat and enjoy moving in 7/8. The facilitator can start with simpler 7/8 material and adjust after observing the group. A workshop for a skilled drummer’s coordination needs a different basis for choosing the material.
Spiech and colleagues found that Western listeners' groove ratings peaked at moderate complexity in 4/4, while simpler rhythms received the highest ratings in the less common meters tested. Here groove means the pleasurable urge to move. The study supports adapting the comparison to listener and meter. The workshop's teaching method supplies the exercise.
Bias-Annotation
The source traditions model different aspects of agency. A person's felt curiosity, an AI policy's objective and a collective's exploration practice call for different accounts of how the effect occurs. Use the shared recovery procedure to keep those subjects visible while comparing their contributions.
Rhythm examples also depend on musical experience. A result for Western listeners and selected meters provides a qualified starting point for that use; another musical practice may support another comparison.
Conformance Checklist
- The repaired sentence identifies the participant and the contribution that matters to the work.
- Combined meanings remain recoverable when both affect the action.
- A proposed score retains its model, comparison and receiving use.
- An exploratory continuation can proceed from its present basis while its eventual destination remains unknown.
- The reader can resume the relevant activity, or identify the particular unresolved meaning that prevents it.
- The recovery stops when ordinary wording is sufficient; further measurement or inquiry follows the needs of the activity.
Common Anti-Patterns and How to Avoid Them
Consequences
A clarified interest claim can guide an experiment, an exploration policy, a practice choice or a discussion of stakes. A project can combine these contributions while preserving what each one explains.
Recovery sometimes reveals that the needed information is absent. Keep that uncertainty attached to the claim. Obtain more information only when its possible contribution warrants the work, using C.11.DUA when the advice or demand itself needs examination.
Architectural Rationale
The common operation is to recover the relation hidden by a familiar expression. A single engagement score would lose the distinction between a participant's response, a model's progress and a construction's further uses. Conversely, treating these meanings in isolation would miss how one project coordinates them.
The procedure therefore identifies the participant and contribution first, and adds models or measurements only for the question that needs them. This also lets a promising new method be pursued from available clues while later uses are still being discovered.
SoTA-Echoing
Learning progress. Schmidhuber's historical formal theory of creativity and intrinsic motivation distinguishes improvement of a predictor or compressor from the unpredictability of its input. Adopt that comparison for progress claims in §4.2. The competing shortcut, fixed-model surprise, can reward an unlearnable signal; §5.1 makes the related information-seeking error actionable.
Information-seeking policies. Paprika describes strategic information gathering for task completion without an intrinsic-motivation reward. Its training curriculum uses a coefficient-of-variation heuristic for sampling tasks. Adapt the policy distinction in §4.1; keep that sampling heuristic separate from a measurement of acquired capability.
Active inference. Friston et al. (2017), §2, supplies the historical coupling of epistemic and pragmatic value. Their 2025 technical note extends information-seeking to distinctions among model structures. Adapt those contributions in §4.2: recover the model, expected information and preferences behind the action. This supplies a richer continuation than explaining all exploration as aversion to an unexpected observation; the selected formal account still determines its assumptions.
Open-ended search. Stanley and Lehman's Why Greatness Cannot Be Planned, Chapters 5 and 9, gives the historical argument for judging available stepping stones without requiring their final destination. Enhanced POET, §§2-3, combines a population-relative difficulty screen with novelty and transfer. Adopt the distinction between an attainable learning opportunity and retained possibilities in §§4.1-4.3. A progress-only selector can lose the latter; the particular search method must supply its generation, retention and resource rules.
Agent-relative challenge. Oudeyer's 2026 preprint review compares curiosity processes across timescales and examines conditions for an intermediate-knowledge preference. Adapt this conditional treatment of the Goldilocks region. It improves on an agent-independent midpoint rule by retaining the learner, environment and changing knowledge.
Rhythmic interest. Toussaint's The Geometry of Musical Rhythm, second edition, Chapter 41 and Epilogue, develops an account using several rhythmic properties and their combinations. Spiech et al. (2025) compare groove ratings across complexity and meter in three behavioral experiments. Adapt their listener-and-meter distinction in §5.3. It provides a more useful selection basis than a universal complexity optimum, while leaving rhythm construction and teaching to their field methods.
Reconsider these comparisons when a source provides a better account of the particular contribution, changes the conditions under which a signal works, or enables the same action with less inference or measurement.
Relations
- E.10 and E.10.ARCH locate this recovery when interest-family wording hides the relation needed for work. F.19 handles the surrounding prose.
- E.10.LRN recovers a remaining ambiguity about what learned or changed. E.10.DEV handles a remaining development or evolution claim.
- C.16 and C.17 supply measurement and the distinctions among novelty, usefulness and sample surprise.
- C.11 compares formed options. C.19 governs directions in an active search pool. C.18 supplies retention in an exploration archive. C.40 develops problems and ways together.
- D.1 and D.3 supply ethical-value and conflict analysis when affected interests raise those questions.
- E.23.CDI supplies capability development when that is the receiving task, separating the acquisition Method, its execution and the asserted capability change. Its entry conditions and the applicable field Method govern the intervention.
E.10.INT:End
Recovering What Development or Evolution Means in the Current Claim
Type: Part E precision-restoration pattern Status: Stable Normativity: Normative unless marked informative.
Use this when. After the F.19 reading and compact E.10 routing, use this pattern only while claim-bearing development, develop, progress, growth, maturation, adaptation, evolution, evolve, or lineage wording still hides what changed, what stayed identifiable, whether success or direction is asserted, whether the claim is actual or modelled, or which direct owner must be used—and resolving that ambiguity changes the claim or next action.
Primary EntityOfConcern. One bounded wording-use restoration whose development or evolution expression has an FPF-governed use.
First useful result. One short ordinary statement naming the exact changed or represented subject, the continuity or membership rule needed by the use, posture, any declared direction or value basis, the direct owner, and the next use or stop.
Cheap exit. Keep ordinary or quoted wording when no FPF inference or action depends on its exact sense. If the subject, claim, posture, and direct owner are already explicit, use that owner and stop.
Not this pattern when. Use the direct subject owners to choose an intervention, assess capability, explain biological or cultural mechanisms, build a dynamics model, select a programme, compare opportunities, or decide what to do next. E.10.DEV recovers the meaning needed to reach them.
Use E.10.LRN first when the unresolved expression is specifically learning, teaching, training, model fitting, inference, or information acquisition. Use E.10.MOVE when trajectory, route, path, or movement posture remains the action-changing ambiguity.
Problem Frame
The same umbrella wording is used for unlike subjects. For example, a person develops a capability, an organization changes its working arrangement, and an organism matures. A population evolves through membership and lineage relations. A model predicts a state history, an engineering search changes an archive or front, and a practitioner proposes a development programme. The claims may share readable language while differing in identity, evidence, method, and practical operation.
The repair should make the first useful subject claim visible. It preserves the source word when useful and returns each substantive question to the pattern that owns it.
Problem
Without a shared way to recover the claim, these mistakes arise.
- the umbrella word substitutes for identifying the current subject and direct owner;
- intervention Work is treated as the holder's actual change or as evidence that the intended result occurred;
- beneficial direction is asserted without a characteristic, scale, polarity, viewpoint, or evidence basis;
- a population or lineage is treated as one continuously developing holder;
- a modelled, predicted, proposed, or planned sequence is reported as actual history; and
- cultural evolution, engineering search, learning, and organization change silently borrow one another's mechanisms.
Forces
Solution
Recover the current claim from the subject and use rather than from the umbrella word.
- Read the expression in its intended use. Follow the wording to the decision or explanation that depends on it. A named change in one member can be relevant to a question about the organization. Establish what connects it to the result being claimed.
- Name the exact changed or represented subject.
- State continuity or membership. Name only the identity, reidentification, membership, generation, lineage, edition, or retention rule needed by this use.
- Expose direction or value only when claimed. Name the objective, characteristic, scale, polarity, viewpoint, and evidence needed by the receiving use. The word development does not establish improvement.
- State posture. Mark the claim's posture—for example, actual, observed, reconstructed, predicted, simulated, proposed, recommended, or planned.
- Use the direct rule for each claim. Distinguish independently actionable claims. If the recipient also needs a connection between them, recover what establishes it. A programme can develop a team’s ability through changed coordination even when its members’ individual qualifications stay the same.
- Stop after recovery. Choose one of six recovery outcomes—a repaired claim, ordinary non-use, quote-only use, missing information, the direct owner, or an exact architecture gap. State the next use or stop for the selected outcome.
Optional DevelopmentEvolutionWordingRecoveryLine
Write the ordinary sentence first. Keep the following temporary line only when another use must inspect or replay the recovery:
blockedOverread? states a rejected reading only when independent local evidence makes that reading plausible to the intended reader and deleting the boundary would change understanding, selection, safety, reliance, stop, or action. Use the temporary line for inspection or replay of the wording repair. When a substantive product needs admission, apply its direct owner’s rule. Omit fields the receiving use does not need.
Direct branches
Coordination with trajectory wording
The phrase development trajectory does not by itself require two full repairs.
- Start with
E.10.DEVwhen the action-changing doubt is which subject is developing, what remains identifiable, or whether improvement is asserted. If that repair also makes the path wording harmless, stop. - Continue to
E.10.MOVEonly when the trajectory itself still carries a separate claim about position space, order, segment identity, actual, predicted, or planned posture, or representation. - Start with
E.10.MOVEwhen the bearer and development claim are already clear and only the path or trajectory posture is unresolved. InvokeE.10.DEVafterward only if a distinct development or evolution ambiguity remains.
Return each result to the original question. One clarification may settle both expressions. If the proposed route and the development it is intended to produce concern different subjects or times, keep both and state the relation the conclusion needs. Neither child creates a DevelopmentTrajectory kind, and no recovery note is required for a clear local sentence.
Worked Slices
Different objects of development
Constructed wording for repair: Our development programme improves the product, develops Mira as an engineer, and advances our research.
Return a missing-information result to the programme author. First ask whether the sentence describes proposed work or work already performed and, separately, whether each effect is intended or observed. Then ask:
- Product: Which product and characteristic should or did change? What comparison makes that change an improvement, under which conditions?
- Mira: Which engineering Work family and conditions does the capability claim concern? For an observed increase, obtain the comparable qualification results through :5.2 before retaining that claim.
- Research: Which claim or problem formulation is meant? What revision is proposed or observed, and under what criterion does it count as an advance?
Change in human capability
Source wording: Mira developed as an engineer.
For the capability reading, recover the named Work family, supported envelope, attained bounds and evidence under A.2.2, with the qualification and time conditions needed by the use. To preserve developed, also recover the change over time from comparable qualification conditions and evidence under the applicable change owner. If that comparison is unavailable, return the missing comparison; state any supported current capability separately.
If the later question concerns an apparent loss, failed transfer, or condition-dependent expression, use current E.23.CAE only to return the qualified differential observation and disposition. That result supplies neither a trajectory choice nor development Work. If a dated intervention occurred, identify its Work and Method separately. Use E.23.CDI when the receiving task calls for capability development.
Organization change
Source wording: The organization evolved after the platform rollout.
If the claim asserts actual rollout Work, recover its performer and occurrence independently under A.13 and A.15.1.
For the separate organization-change claim, identify the organization and the changed working relations. Use the applicable organizational DPF or domain Method to establish which working relations changed and under what conditions, with A.3.4 for the actual change and the direct patterns for the other claimed relations. Recover the roles, assignments, interfaces, routines, capabilities, observed organization Work and effects, period, evidence, or cultural relations on which that claim depends. State any benefit under its declared basis and supporting evidence.
Population and archive
Source wording: The population evolved as the archive improved.
Split the claims. Recover the population relations the claim relies on—for example, membership, generation or lineage, variation, selection or retention, environment, or distribution—and their evidence under the applicable domain account. Keep a sufficient supported population claim; return only a definition, relation or evidential premise still missing for the receiving use.
Archive or front improvement uses the relevant C.17–C.19 contribution. Establish any additionally claimed reproduction or population relation under its own domain rules.
Model and plan postures
For The model predicts rapid development, use the model branch in :4.2 to identify the model edition and predicted state transition. For The programme proposes rapid development, identify the proposed claim and use its question-specific return in :4.2. Preserve each claim's predicted or proposed posture. Any claim that actual development occurred needs the applicable direct owner's evidence.
Development trajectory
Source wording: The development trajectory improved.
Start by asking what developed and what improved means. If the intended result is Mira's capability for review Work increased under the named measures and evidence, E.10.DEV returns that claim to the capability and change owners; apply their rules to the comparative evidence to establish the increase.
Use E.10.MOVE only if a separately relied-on trajectory representation or ordered path remains—for example, a planned sequence of interventions or a modelled state history. Otherwise stop with the capability claim instead of completing a second form.
Bias Annotation
- Human-development default. Select education or human learning when the recovered claim concerns it.
- Benefit-by-word bias. Tie any positive direction asserted through development, progress, growth, or maturity to a declared basis.
- Single-holder bias. Use membership and lineage for population claims and one-holder continuity for a continuously identified holder.
- Model-to-world bias. Preserve predicted or simulated posture; establish actual change under its own evidence rule.
- Candidate borrowing. Use admitted patterns for current coverage; keep a candidate pattern or future inquiry visibly provisional.
- Form completion bias. Open the second lexical child only for an independent unresolved ambiguity.
Conformance Checklist
- Is the expression claim-bearing for the current use?
- Is the exact changed or represented subject named?
- Is the needed continuity, membership, lineage, edition, or retention basis visible?
- Are intervention Work, Method, plan, result, evidence, representation, and later effect separated where current?
- Is any direction or value claim tied to the basis the use needs?
- Is the claim's posture explicit, using the distinctions needed by its direct owner?
- Does the repair reach one of the six outcomes in Step 7: the repaired claim, ordinary non-use, quote-only use, missing information, direct owner, or exact architecture gap?
- Is candidate or planned posture preserved? A candidate pattern supplies current FPF law only after admission; a current plan may be the subject of a claim while its intended result remains planned.
- For development trajectory, did the second child open only for a remaining independent ambiguity?
- Does each substantive claim return to its direct owner?
Common Anti-Patterns and Repairs
Consequences and Reopen Condition
Benefits. Familiar development and evolution language remains readable while holder, population, model, plan, archive, cultural, and learning claims retain their own identity and evidence. Readers get a positive first result rather than a prohibition or a list of PatternIDs.
Costs. Some sentences need to split, and a load-bearing direction or continuity claim needs its own basis. Domain mechanisms and evidence still require their direct sources.
Reopen this pattern when a recurring branch cannot reach a direct owner or exact gap at task-sized effort; when cold readers still transfer evidence between holders, populations, models, plans, or archives; when an applicable population account changes the boundary; or when a direct pattern absorbs the same entry, action, first result, and stop without losing discoverability.
Rationale
The reusable transdisciplinary problem is recovering the claim carried by the wording. Across these cases, recover the subject and continuity basis, separate Work and result, expose direction and posture, choose the direct owner, and stop.
E.10.DEV makes that recovery available through compact E.10 routing. Developmental science, evolutionary theory, organization change, guidance, and learning retain their direct owners.
SoTA Echoing
Practice question. When development or evolution carries a claim that can change action, what bounded recovery returns a usable direct claim?
FPF source contributions
Reopen only the affected route when its supplying pattern changes the entry, result or boundary used here.
Relations
- Selected by:
E.10when development or evolution wording remains action-changing because the subject, continuity or membership, posture, direction or value basis, or direct owner is not yet recoverable. - Builds on:
F.19,E.10,E.10.ARCH,A.3.4.P,A.2.2, currentE.23.CAE,A.3.3,B.4,C.17–C.19,C.27.TA,C.29, andC.36. - Coordinates with: current
E.23.CAEonly when the recovered capability branch needs its observation or disposition differential;E.10.MOVEfor a remaining trajectory or path posture;E.10.LRNfor learning-word recovery;C.36.Pfor remaining cultural-evolution wording; A.15 for Work or WorkPlan;E.23.CDIwhen capability development is the selected task; and each direct holder or domain owner selected by the repaired claim. - Keeps outside: domain ontology, governed by the selected subject owners, and framework admission under
E.4.PFAD.
E.10.DEV:End
Clarify Wording About Moves, Readiness and Trajectories
Type: Part E precision-restoration pattern Status: Stable Normativity: Normative for move-like, movement-like, readiness-like, route-like, path-like, and trajectory-like wording-use restoration.
Use this when. After the F.19 reading and compact E.10 routing, use this pattern only while move-like, movement-like, readiness-like, route-like, path-like, or trajectory-like wording hides the governed claim—for example, a demonstrated continuation, a prediction, readiness for a named Work, or an actual, planned, or modelled trajectory.
Primary EntityOfConcern. One bounded wording-use restoration within the scope above.
First output. Repaired wording, a truthful split, or a blocker. When later replay relies on the repair, retain its wording and interpretation basis in an existing account or use the temporary MoveAndReadinessWordingRepairNote in §4.1.
Not this pattern when. Use A.3.4.P first when the wording is primarily about a transformation or change situation. Use E.10.DEV first when development or evolution still hides the changed subject, continuity or membership, or direction or value claim; continue here only if an independent trajectory, route, ordering, posture, or representation ambiguity remains.
Use F.19 and the direct subject pattern when the current object is already known. Generic process, workflow, loop, or flow wording stays outside unless it independently carries one of the governed move, readiness, route, path, or trajectory claims.
Problem Frame
"Move" is useful in project conversation. It can mean a chess-like next choice, a first FPF use, a TameFlow MOVE, an architecture candidate, a language-state transition, a call-planning next action, a work-preparation item, or an ordinary action. "Ready", "full kit", and "work entry" can likewise mean source currentness, work planning, preparation work, gate passage, or performed work.
The defect arises when the word is used to choose the ontology. E.10.MOVE restores the object under wording repair and the direct FPF relation before any rewrite is accepted.
Problem
Without this restoration:
- FPF mints a false root
U.Move. - Pattern-use recommendations become performed work or work authorization.
- TameFlow
MOVEis imported as if it were an FPF kind. - Readiness labels become gate passage or work occurrence by appearance.
- Route, workflow, process, and path wording is repaired through taste rather than through the governed object.
Forces
Solution
Cheap ordinary use. When the governed value and its direct pattern are already evident, apply F.19, name the value, rewrite the phrase without changing the claim, confirm the remaining admissible reader use, and stop. Skip the repair note and disposition table for that case.
Open the fuller procedure only when the wording remains ambiguous, carries several governed values, imports a source term, or must be replayed later.
For a mantra, recover the pattern use before deciding that two phrases mean different moves. A question about a difficulty can refer to the Problem frame or Problem; a reminder of what to do can refer to the same use through the Solution. “Find what the trial establishes” can state the difficulty and the action intended to resolve it.
When the subject, intended result and conditions are clear, retain the useful wording. Clarify a missing participant or relation where the reader would otherwise choose a different pattern or action. This wording choice decides neither whether a mantra is local or long nor whether a surrounding presentation is a demonstrative slice of an independently qualified CGUS.
Restore the governed target before choosing replacement wording:
-
Name the exact
GovernedTextSpan, theClaimBeingMade, and theObjectUnderWordingRepair. -
Decide whether the wording is ordinary prose, a quotation, or wording relied on for an FPF-governed claim. Ordinary and quotation uses can close without inventing a technical target.
-
When the phrase is
mantra move, first ask which use is present. In a post-qualification A.22.CGUS demonstrative slice that shows pattern use, recover the exact E.11.PUAPatternUsePracticeContinuationDescription@Context: its proposed use, expected result and kind, PatternID and name, current condition, and continuation disposition. Keepmantra moveonly as bounded Plain wording for that shown continuation. A.22.CGUS supplies the structure and slice boundary; it does not create a universal displayed-row kind.For a Plain local mantra, name the bounded result and restore the move-like wording through that result's exact predicate or constraint.
For a Plain long mantra, recover the intended connected reasoning and the particular question now at issue. State the answer or blocker, the subject pattern that supplies it, and the other answers whose use depends on its conditions. A whole-account question may require comparing those answers rather than finding one final result or one missing map location. Do not invent a demonstrated row, collapse the long map into one pattern's Solution, or treat any branch as Work order.
-
When
move,movement,direction, or similar wording predicts a later evaluation result, recover the coordinate-and-scale-qualified prediction, its proposal basis, and protected tradeoffs underE.23. UseExpectedEvaluationResultChange@Contextonly when E.23's conditional account is needed for the receiving use. The prediction is not an operation, transition, movement, work occurrence, or proof of improvement. -
For every other governed use, name the exact recovered value or relation, its kind, and its subject pattern. For a relation claim, name the admitted direct predicate and actual participants. Add a
RelationSignaturereference only when an admitted reusable typed declaration is current and the receiving use needs that declaration. If the governed value is already clear, use its pattern directly. -
Distinguish the governed claims carried by one phrase and retain the connection the reader needs. For readiness to work together, identify the performances to be combined and the conditions they must satisfy together. Two crews can each be ready for a lift while their simultaneous lifts require the same crane. For a question confined to permission, a single authorization covering both crews can suffice.
-
Preserve
RemainingReaderUse: the repair is complete only when a practitioner can still tell what can be inspected, selected, evaluated, planned, performed, or returned to next.
MoveAndReadinessWordingRepairNote
The governed-value ref and kind ref are both present or both absent. BlockedOverread? states a rejected reading and appears only when independent local evidence makes the exact rival reading plausible to the intended reader and deleting the boundary would change understanding, selection, safety, reliance, stop, or action. The relation-signature ref is present only when an admitted reusable typed declaration is current and the receiving use needs that declaration. Otherwise a relation claim names the admitted direct predicate and actual participants without a signature ref.
A governed use has a non-semantic SubjectPatternLocator: an ordinary PatternID that identifies the pattern whose content defines, constrains, or tests the recovered value. Where the receiving claim needs a Method or MethodDescription, use the independent A.3.1 and A.3.2 conditions; admit any Method-use relation under its direct relation owner.
For ordinary prose or quote-only use, the disposition explains why no FPF object is claimed; the corresponding object positions may remain absent. The ...Ref fields carry references of the declared RefKinds; they do not carry the referenced values or kinds.
A retained repair includes the edition, source, context or time boundary needed for its later use, its interpretation basis and any consequential reopening condition. An existing account can carry this information; use the temporary note when its structure helps. Use G.11 only when actual refresh orchestration is current; the note records the boundary on its own currentness.
FinalWordingOrBlocker gives the wording or blocker for this bounded repair under its qualification and currentness conditions; a later change can reopen it. The note is a temporary wording-restoration aid; substantive results use their direct pattern's admission rules. Ordinary immediate repair need not materialize the note.
Trigger groups
After E.10 selects this pattern, use these cue groups to find the appropriate recovery branch while an action-changing ambiguity remains:
move,step,action,application,solution, andnext action;readiness,ready,full kit,work entry, andlaunch-ready;commit,committed,commit checkpoint, orcommitTriggerwhen they hide a route-adoption or work-entry transition;movement,direction, orshiftused for an expected evaluation-result change;route,workflow,process,path,trajectory,loop, orflowused for an unresolved claim about a path, ordering, or what it represents; use the direct exits below;- imported source wording such as TameFlow
MOVE.
The cue group locates a recovery branch. The recovered claim and its direct owner determine the governed-value kind.
Readiness exits
Stay in E.10.MOVE only while readiness, ready, full kit, work entry, or a similar cue still hides which governed value is meant. Once that value is recovered, use the direct pattern:
If the direct pattern and value were already clear, bypass this table and use that pattern.
No synonym closure
Recover the governed value and its subject pattern before closing a synonym replacement. Ordinary-prose or quote-only use closes when no FPF-governed value is claimed.
If responsibility is the remaining claim, name the admitted System, direct domain predicate, actual participants, and applicability, or identify the missing governor under A.6.RCD. An assignment is not a responsibility result. Individuate the responsibility-relation occurrence separately only when a named receiving use needs to distinguish that occurrence.
Trajectory wording recovery
Use this branch when trajectory or close path wording remains claim-bearing after any primary transformation wording has been recovered. The first result is an ordinary repaired claim or exact gap, not a trajectory record.
Ask only the questions the receiving use needs:
- What exact bearer or represented subject is positioned or ordered?
- What identity, continuity, membership, lineage, or edition rule matters?
- Which declared position, state, configuration or possibility space is relied on, if any? Name a defining episteme edition separately when the use depends on it.
- What is the ordering or reference domain—time, event, generation, plan order, graph order, or another index?
- What counts as a position, segment, branch, interval, generation, or edge for this use?
- What posture does the claim need—for example, actual, observed, reconstructed, predicted, simulated, proposed, recommended, or planned?
- Which direct pattern owns the resulting claim, what receiving use is allowed, and is any grounded non-use boundary needed under the
F.19plausible-intended-reader test?
These are recovery questions, not fields of a new Trajectory, TrajectoryAccount, relation head, Method, or mandatory card.
For development trajectory, open E.10.DEV first when the action-changing doubt is what develops, what remains identifiable, or whether improvement is asserted. Continue here only if trajectory still carries an independent claim about position, ordering, posture, or representation.
If the bearer and development claim are already clear and only path posture is unresolved, start here and open E.10.DEV afterward only for a remaining separate ambiguity. Do not require two notes or two full passes by spelling alone.
Wording-use dispositions
WordingUseDispositionValue is a local finite enumeration for choosing a repair branch. It is not a U-kind, relation kind, state frame, or claim about the project value being repaired.
Relation to A.3.4.P
Use A.3.4.P first when the claim is about a change situation or transformation-flow structure. Use E.10.MOVE only for the remaining wording-use question. If the same sentence also recommends a pattern use, claims readiness, or names a demonstrated continuation, split those claims and use the direct pattern for each.
Durable name repair
A durable name states the recovered subject value or relation; it does not retain an implementation head merely because the fields are typed.
These are repair demonstrations, not a global replacement table.
Archetypal Grounding - Worked Slices
Bounded mantra move
Source sentence: "The next mantra move is to compare the two patterns."
For the demonstrative case, recover one E.11.PUA practice-continuation description inside a named post-qualification demonstrative slice. The description states its proposed use, expected result and kind, direct PatternID and name, current condition, and continuation disposition. That PatternID locates the applicable pattern. If the pattern choice is unresolved, the description may point to a separate nested selection question.
Selected fields of an optional note; include BlockedOverread only for an observed or independently grounded misreading:
Expected evaluation-result change
Source sentence: "The repair should create an upward evaluation movement."
If the claim predicts a later evaluation result, restore the evaluation pattern, coordinate, scale, current result, one expected scale value, range, or closed direction, candidate proposal basis, and protected tradeoffs. Keep an ordinary prediction description when it is sufficient; use ExpectedEvaluationResultChange@Context only when E.23's conditional account is needed for the receiving use.
If those positions are unavailable, keep a provisional prediction description or use E.22 and E.23 to obtain the missing prediction basis.
Next FPF use
Source sentence: "The next FPF move is to check architecture."
If this is a project-local recommendation, recover its applicability, expected result, worth and boundary under E.11.PUR, and identify the architecture pattern being recommended. Materialize PatternUseRecommendation@Context only when a named later use needs the recommendation addressable.
With this sentence alone, the architecture, check question, and expected result remain unspecified. Return the blocker: "Specify which architecture is to be checked, which question the check must answer, and which result is required."
Once those are known, the recommendation may say "next useful pattern use" and name the operation on that architecture and its expected result.
TameFlow MOVE
Source sentence: "The MOVE is full-kitted and ready."
Preserve MOVE as imported source wording. Restore the exact WorkPlan and PlanItem for the intended performance, full-kit criterion, A.15.5 work-entry readiness result, and any actual gate decision under their direct patterns. Do not claim target Work occurred unless a dated A.15.1 occurrence is current.
Workflow diagram
Source sentence: "This workflow is the next move after problem framing."
If the diagram describes a transformation-flow structure or method description, use A.3.4.P, E.18, or A.3.2. If the sentence recommends the next pattern use, use E.11.PUR. If it demonstrates one continuation through a wider CGUS, use A.22.CGUS. Split the sentence when more than one claim is current.
For example, if the surrounding text identifies an admitted MethodDescription for heat-treating a shaft, the descriptive clause becomes: "This diagram describes the method for heat-treating the shaft." State any recommended next pattern use separately, with its object and expected result; return a blocker while those remain unspecified.
Evidence path
Source sentence: "Follow the evidence path to approval."
Use A.10 to recover the selected claim, exact sources, independently established evidence or provenance relations, currentness and bounded reliance. Identify separately the decision meant by approval: A.21 governs an applicable gate result; any authorization or commitment uses the pattern governing that exact relation.
Manufacturing operation
Source sentence: "The next move is to heat-treat the shaft."
If this names the reusable way of changing the shaft, recover the Method under A.3.1; use A.3.2 only if the claim depends on a particular episteme's MethodDescription membership. If it places a heat-treatment operation in intended work, recover the WorkPlan or PlanItem under A.15.2.
If heat treatment has occurred, recover the dated A.15.1 Work occurrence, affected shaft, method enactment, and result. If the question is whether that intended work can start, recover A.15.5 work-entry readiness.
If the receiving context does not select among them, return the blocker: "Specify whether this describes the heat-treatment method, plans the work, reports completed heat treatment, or asks whether the planned work can start."
Clinical readiness
Source sentence: "The patient is ready for discharge."
When ready hides a patient-state claim, use A.19.SPR to recover the patient as bearer, the clinical state frame or subject pattern, the current value or classification, its evidence and qualification window, and the practical discharge use. A discharge recommendation, accountable decision, work-entry condition, and completed discharge remain different claims under their direct clinical and FPF patterns.
Reopen when a local mantra is not CGUS
Initial sentence: "The next mantra move is: name the thing."
An initial repair classified the phrase as boundedDemonstratedContinuation. Inspection then shows that the enclosing text is A.6.P's local RPR mantra: a short rendering of the A.6.P Solution. It has no qualifying wider ConstraintGovernedUnfoldingStructure@Context, no post-qualification slice of an independently qualified CGUS, and no E.11.PUA practice-continuation description with the required proposed use, expected result, pattern, condition, and disposition.
That evidence overturns the initial disposition. Remove the demonstrated-continuation claim, retain the local RPR mantra as Plain didactic wording, use the A.6.P Solution and its direct relation-recovery guidance, and write: "Apply the first clause of the local RPR mantra: name the thing; then recover the relation or comparison."
The A.6.P locator and Solution establish neither a U.Method nor a U.MethodDescription. Establish a separate U.Method, a qualifying U.MethodDescription episteme, and any Method-use relation only if A.3.1 and A.3.2 independently admit them and the receiving claim depends on those identities.
Reopen the demonstrative-slice question only if a later qualified structure and slice actually show a complete E.11.PUA practice-continuation description.
Trajectory under changing constraints
Constructed wording for repair: Our architecture follows a trajectory under changing constraints.
Read the sentence through F.19 first. If it means only that the team expects to revise the architecture as constraints change, state that proposed action directly.
When an FPF inference relies on the trajectory claim, recover the exact architecture or system subject, the changing constraints and reference window, whether the sentence concerns actual changes to the architecture, editions of its description, a proposed evolution policy, or a modelled sequence, and the direct architecture, transformation, or model owner.
For a C.29 curve or ordered rendering, name its correspondence to the architecture and the losses allowed by that use; establish transformation and evidence claims under their direct owners.
Overlap example: The development trajectory improved. Start with E.10.DEV to recover the developed subject and the basis of improved. Open this branch only when a separately relied-on ordered path, model, plan, or representation remains. Stop when the recovered direct claim answers the question.
Bias-Annotation
Lenses: Gov, Arch, Onto/Epist, Prag, Did. Scope: FPF-governed move, readiness, route, path, and trajectory wording uses.
The method foregrounds Onto/Epist distinctions and direct subject ownership. The cheap ordinary-use path protects practical use and readability: recover only the distinctions that matter to the current claim and keep useful familiar wording. Recurring misuses and their repairs are in §8.
Conformance Checklist
Lowering and Reopen Conditions
Lower, block, or reopen the repair under any of these conditions.
- The governed text span, claim being made, or object under wording repair is not recoverable.
- The wording-use disposition is uncertain.
- The proposed wording changes kind or relation without an accepted subject pattern, or the subject pattern is missing.
- A change-situation claim was not separated from pattern-use or readiness wording.
- The repaired wording loses the remaining reader use.
- Changed source wording invalidates the recorded source-licensed use.
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits:
- FPF keeps friendly move, readiness, route, path, and trajectory language without letting it mint false kinds.
- A trajectory sentence reaches its direct claim owner or exact gap. Subject, posture, ordering, and representation are recovered separately where the use needs them.
- Pattern-use recommendation, P2W, work readiness, gate decision, performed work, transformation, architecture, and call planning stay separable.
- A wording review can locate unresolved move claims while preserving adequate uses.
Costs:
- When a later use needs replay, retain the wording and the interpretation or qualification basis that use requires. The temporary note is one possible form.
- Text may need to split one sentence into two governed claims when the original wording carried both change-situation and pattern-use meaning.
Rationale
Familiar move, route, readiness, and trajectory wording can hide different governed claims. E.10.MOVE gives a narrow restoration path: recover the governed text span, claim, bearer or represented subject when relevant, posture, and object under wording repair; classify borrowed or ordinary wording; name the governed FPF value; preserve the remaining admissible reader use; and apply the pattern that defines or constrains that value.
The mantra branch routes an admitted demonstrative use through one A.22.CGUS and its E.11.PUA continuation description, a Plain local use to its bounded result's direct pattern, and a Plain long use to the subject pattern of the current map answer or stop. Evaluation-movement wording uses E.23 for a separate prediction about a later evaluation result.
The trajectory branch separates the subject from posture, ordering, and representation and returns to the subject pattern or exact gap. E.10.DEV coordinates only when development or evolution still carries an independent ambiguity. Recommendation, transformation, readiness, gate, publication, choice, plan, and Work claims remain with their direct patterns.
SoTA-Echoing
The comparison separates direct-claim recovery, cue preservation, and imported source meanings.
Effort boundary. Each clear case takes the cheap exit. An ambiguous case opens only the branch whose question remains. If the subject or posture cannot be recovered, the cost is leaving the claim open and stating the precise gap.
The selected line is FPF's direct-claim recovery. The external sources contribute a domain comparison, a cue-preservation hypothesis, and imported practice meanings; the constructed architecture case is in §5.10.
Relations
- Builds on:
F.19,E.10,E.10.ARCH,A.3.4.P,A.22.CGUS,E.11.PUA,E.11.PUR,E.23,A.15.5, andE.24. - Coordinates with:
E.11.PUAfor thePatternUsePracticeContinuationDescription@Contextshown by a qualified practice continuation;E.11.PURforPatternUseCoordination@Context, onePatternUseOrderingRelation@Context, or the bounded total-orderPatternUseSequence@Context;E.10.DEVwhen development or evolution wording and trajectory wording carry independent ambiguities;A.1.STMfor connected system-thinking project reasoning and its long-mantra reminder;A.3.3,A.3.4,A.3.4.P,B.4,C.27.TA,C.29,C.17–C.19,C.22.2,C.11, A.15.2,C.36, andA.16.0for trajectory exits; andE.18,E.18.1,A.15,A.21,C.24,C.30,E.17,F.17,F.18,G.11, A.10, and each recovered value's direct subject pattern.F.18governs a durable-name decision;G.11governs refresh orchestration only when currentness, edition, telemetry, freshness, or decay is the actual claim. - Selected by: E.10 compact routing when move, readiness, route, path, or trajectory wording still has an unresolved FPF-governed use after the
F.19reading and no direct subject pattern has resolved it.
E.10.MOVE:End
Architecture for Recovering Ontological Meaning in Wording
Type: Architectural (E) Status: Stable Normativity: Normative unless marked informative
Plain name. Wording ontology repair architecture.
Intent. Organize recurring wording recovery so that subject patterns explain their own objects, rules and useful actions, while authors can reuse a shared recovery rule.
Use This When
Use this pattern when a recurring FPF-governed wording problem survives the F.19 reading and compact E.10 routing, and a shared recovery rule is needed for the hidden object, relation or use. In DPF authoring, enter through E.4.DPF only when recurring domain wording obstructs a named reader use. The shared method remains in FPF; the domain entry stays with its DPF claim.
Failure to recognize. Subject patterns repeat local wording-recognition tables and teach repair doctrine instead of their own governed object, invariant and first useful move. Readers then lack one stable recovery rule.
First useful move. Decide whether F.19 and compact E.10 routing can close the wording, whether a subject rule already defines or tests the recovered claim, or whether one shared applicability row is needed. Add that row only for recurring subject distinctions, a stable recovery shape and a useful reader result that no existing rule supplies; §1 gives its content and §6 its admission criterion. Cite a PatternID to locate its rule, and identify an exact ClaimGraph only when the receiving use depends on that identity.
Not this pattern when.
- If a sentence is repaired locally under
F.19with compactE.10routing, stop there. - If the governed object, exact direct relation, or claim-bearing episteme and its defining or testing rule are already recoverable by value, state the claim under that rule and cite the pattern containing it.
- If the wording problem is phrase-level apparatus around an already recoverable kind, use
F.19rather than creating a new wording-use restoration row.
Problem Frame
Precision restoration in FPF is ontology-first, not word-substitution-first. A recurring wording family is important only when it hides a stable governed object, direct relation kind, actual participant or relation-participant meaning, declaration-local SlotSpec, assertion-side participant designation, claim kind, publication-use relation, source-relation disposition, mathematical lens, or neighboring-pattern boundary.
Problem
Without a shared distribution architecture, subject patterns collect first-stage repair catalogues and lose their object focus. The same false friend is then repaired differently in architecture, characteristic, evidence, publication, method, relation, and state-family patterns.
Forces
Solution
Use F.19 for whole-span repair, E.10 for compact recognition and routing, and E.10.ARCH only for the ontology that remains unresolved. Once the object, relation, or claim is recovered, use its defining, constraining, or testing rule and cite the pattern that contains it. Add an applicability row only when recurring wording hides a stable recovery shape and a useful surviving reader use that no existing rule carries.
Primary EntityOfConcern and applicability-row scope
The primary EntityOfConcern for this pattern use is the pattern-local authoring and publication architecture of WordingUseRestorationApplicabilityRow rows. The project object exposed by the wording is recovered under its own defining or constraining rule.
A row may use semanticAreaBaseConcept, semanticArea, semanticAreaSenseFamily, and ontologicalNeighborhood as author-facing routing coordinates. Its minimum semantic content is:
- the recurring wording use recognized by
E.10; - the exact governed entity, value, episteme, obtaining direct relation, or representation exposed by that use;
- the exact claim or use being made;
- the rule that defines or constrains that object, relation, or representation, and the pattern ID that locates the rule;
- the repaired wording; and
- the admissible reader use that survives.
A negative boundary is not row-minimum content. Add one only when independent local evidence makes the exact rival reading plausible to the intended reader and deleting the boundary would change understanding, selection, safety, reliance, stop, or action. State it once and in the smallest useful form.
Declaration, designation, reference, publication, or representation fields are optional and appear only when the current repair needs them. A reusable relation declaration names its RelationSignature and A.6.5 SlotSpec values. A current assertion or relation-occurrence-description episteme may carry participant designations. A publication form or C.29 representation element names its represented object and explicit correspondence. An E.24 onticSlotRelation appears only when durable ontic settlement is itself current. None of these optional objects becomes a field of the governed entity merely because the authoring row cites it.
WordingUseRestorationApplicabilityRow names this authoring/publication unit. An author uses it to select and explain a repair; project-kind admission and project authority remain with their defining subject rules. Ordinary engineers do not fill it. They receive the shortest practitioner-facing sentence that identifies the governed object, direct relation or claim, and remaining action-facing use.
WordingUseRestorationApplicabilityTable is the pattern-local publication table of such rows. It publishes rows for lookup; row membership supplies no semantic parentage or project authority.
semanticAreaBaseConcept is the Base concept, source wording span, or already settled row cue by which an author first recognizes the candidate semantic unit.
semanticArea is the Part-F semantic unit used by one wording-use restoration row: one Concept-Set row, one UTS row, or an explicitly bounded row-set whose rows remain sense-uniform enough for one recovery architecture.
semanticAreaSenseFamily is the Part-F senseFamily or FPF kind named by value-family discriminator used to select one sense- or kind-specific recovery.
ontologicalNeighborhood means the FPF applicability neighborhood around that named semanticArea: the exact governed objects, admissible adjacent objects and relations, subject patterns, use boundaries, and optional declaration, description, publication, reference, or representation objects needed by the current repair. Membership follows applicability to the current repair, independently of textual or file proximity, index order, topic or domain grouping, work organization, and pattern numbering.
pattern nest means a numbering or placement grouping such as A.6.*, C.16.*, or C.30.*. One applicability row may point to a realization pattern in one pattern nest.
Distribution architecture
The patterns divide the work as follows.
- The author applies
F.19to the whole span and uses the compactE.10cues to locate an FPF-governed wording use. Close it locally when the object and predicate are recoverable; otherwise select the defining or testing rule, controlled precision-reduction pattern, durable-name application, or fail-closed non-use disposition. - The
E.10.ARCHtext contains the shared recovery algorithm and theWordingUseRestorationApplicabilityTable. - The author applies the bounded entry, realization pattern, or direct subject rule selected through
E.10:0.2aand the applicability table below. The selected content unpacks the wording for one namedsemanticAreaand itsontologicalNeighborhood. UseA.6.P.WMRafter generic relation recovery only when one Method and Work boundary claim remains hidden; useA.6.RCDonly after exact participants are recovered and no current direct relation or already admitted local relation-bearing claim closes the receiving claim. - Add an applicability row, and a realization pattern only if needed, when repeated FPF-governed wording hides a stable primary-EntityOfConcern use field set, a stable recovery shape, and a useful remaining reader use that no existing subject pattern carries.
E.8defines pattern-form and placement requirements for wording such aspattern nest, and requires authoring prose that usesontologicalNeighborhoodto expose the relevantsemanticAreaBaseConcept,semanticArea, andsemanticAreaSenseFamilyrather than treating neighborhood as the semantic unit.- An author or reviewer uses
E.19to check that authored pattern hosts preserve this distribution and do not keep rival first-stage repair doctrine.
This architecture keeps E.10 compact. It also keeps subject patterns centered on their own primary EntityOfConcern values, decisions, characteristics, structures, mathematical lenses, consequences, and worked uses.
EntityOfConcern and recurring hidden-field distribution
EntityOfInterest, EoI, EoIClass, describedEntity, DescribedEntityRef, and primary described entity are active repair triggers. FPF-governed wording must recover the EntityOfConcern-family use named by value, publication-unit primary-EoC or ordinary primary-subject use, or local FPF kind, then rewrite to EntityOfConcern, entityOfConcernRef, EntityOfConcernRef, EntityOfConcernClass, publicationUnitPrimaryEntityOfConcern, publicationUnitPrimarySubject for the E.17.AUD.OOTD primary-subject use, or the local FPF kind named by value.
For the non-claim-bearing case, use the kind named by value or plain topic or subject without inventing an EntityOfConcernRef. If no use is recoverable by value, the wording remains quoted source or trigger wording and cannot be used for reliance.
-
C.2.1defines episteme identity and the identifiedEntityOfConcernparticipant ofEpistemeConstitutionRelation;A.6.5definesEntityOfConcernSlotonly inside a reusable constitutionRelationSignature; direct reference rules define or constrainentityOfConcernRef,EntityOfConcernRef, and related reference use. -
C.2.Pcarries episteme, publication, source-wording, and source-relation precision restoration when the sentence still hides source wording, claim-bearing episteme, publication construction, publication-form construction, project-side reliance, pattern-application wording, or use or non-use disposition. -
F.18carries durable naming, selected head settlement, and source-string and durable-name discipline after the kind under repair and use are recovered. -
C.2.Pdefines andE.17.AUDuses the union fieldpublicationUnitPrimaryEntityOfConcern: the carried EntityOfConcern or the non-claim-bearing kind, topic, or subject value.E.17.AUD.OOTDgoverns its separate localpublicationUnitPrimarySubjectfor one bounded publication unit with one carried move and one outside-work boundary.Only when the unit carries one identified episteme
Eand the subject is the exact entity that its claims concern maypublicationUnitPrimarySubject = EntityOfConcern(E)be stated. Neither field adds a participant toEpistemeConstitutionRelation, and the fields are not aliases. -
A.6.3, its retainedentityOfConcernRef-preserving specializations, andA.6.4carry preservation or retargeting of the EntityOfConcern across episteme morphisms. -
When an evidence, assurance, gate, work, decision, architecture, characteristic, mathematical-lens, or project-side claim is already recoverable, use the rule that defines or constrains that exact claim or its admissible-use boundary.
This selected-family case is the standing example for recurring hidden-field architecture. When a new hidden-field family recurs, do not add local warning prose to every pattern. Use an existing defining or testing rule when one fits, add one applicability row when shared recovery is needed, and justify a new realization pattern only when the hidden field set, recovery apparatus, and remaining reader use recur across FPF-governed texts.
Ontic-Level and Facet-Level Restoration Distribution
Use this distribution before adding or specializing a wording-use precision-restoration pattern.
E.10 is the shared recognition scan. The author uses it to identify an FPF wording-use problem and choose the first applicable restoration or the rule that already defines or tests the recovered object or claim. E.10.ARCH supplies the distribution rule. A specialized restoration pattern contains only the defining content for stable ontological recovery in one selected ontic, semantic area, or facet.
When the governed object, exact direct relation, claim-bearing episteme, representation use, or claim kind is already recoverable by value, use its defining, constraining, or testing rule directly and cite the pattern containing that rule. A clear A.3.4, A.6.F, C.29, E.18, C.30, A.15, A.10, gate, decision, publication, or evidence claim needs no restoration detour merely because a familiar trigger word appears.
For bare claim-bearing role, use E.10.ROLE to recover the sentence's work-facing or use-facing object and stop when one exact object or relation and its defining rule are clear. Use A.6.RSIR only for the narrower relation-signature-interface-assignment-slot cluster, including direct-relation, declaration, interface, operation, and representation branches recovered under E.10.ROLE. After selection, the recovered object's rule defines the content and its pattern ID locates that rule.
For claim-bearing learn, learning, learned, taught, or trained wording, use E.10.LRN only while the participant, changed subject, Work, Method, result, evidence, or receiving use is hidden. Return the corrected or split claims to their direct subject patterns.
For claim-bearing development, develop, evolution, evolve, progress, growth, maturation, adaptation, or lineage wording, use E.10.DEV only while the changed or represented subject, needed continuity or membership, posture, direction or value claim, direct owner, or receiving use is hidden. Use E.10.MOVE afterward only when a separately relied-on trajectory, route, path, order, posture, or representation remains ambiguous.
Use an ontic-level restoration pattern only when recurring wording hides a candidate durable ontology unit whose primary governed subject kind, stable identity, core direct relation, named neighboring direct relations, and subject patterns need joint recovery before wording repair. The restoration recovers the exact current governed objects and direct relation uses; it does not treat declaration-local SlotKinds or assertion-side participant designations as parts of the ontic.
Use E.24.CD only when recurring wording exposes a candidate subject that may need an E.24 ontic-introduction decision: a potential primary governed subject kind, stable identity, core direct relation, named neighboring direct relations, and action-facing gain that no existing rule carries.
Use E.24.PUB only when the repair must distinguish ontic, ontic-description episteme, publication form, view, record, card, table, schema, data-structure expression, rendering, or source relation.
If A.22, A.19, C.30, A.3.4, C.2.1, or another pattern already contains the defining or testing rule for the recovered object or claim, use that rule directly and cite E.24.CD or E.24.PUB only for the relevant thin boundary.
Use a facet-level restoration pattern only when one recurring facet cuts across several ontics or subject patterns and has its own stable ambiguity. Function-like wording under A.6.F is the standing example: function wording may point to transformation behavior, performed-work action or another actor-side claim under an exact direct governor, a separately typed architecture or other influence source under its exact relation, mathematical function, module allocation, capability, quality, role, work, method, evidence, assurance, gate, or decision. That facet is too broad to duplicate inside every ontic-level restoration pattern and too specific to leave as ordinary prose.
Do not create one precision-restoration pattern per relation-participant meaning or declaration-local SlotKind. A separate restoration pattern is justified only when the same ambiguity recurs across several patterns, changes the selected FPF kind or direct relation use, and would otherwise force repeated first-stage repair prose. Otherwise use the rule that defines the direct relation and apply A.6.5 only when reusable declaration is current.
When both an ontic-level restoration pattern and a facet restoration pattern are applicable, apply them by recovered question, not by word order. The ontic-level pattern asks which candidate subject, governed objects, core and neighboring direct relations, and subject patterns are current. The facet pattern asks how the overloaded facet word is assigned after that recovery. For example, transformation wording that includes function, functional, or functioning may use a transformation-ontic restoration pattern to recover U.Transformation, TransformationFlowStructure, its exact participants and direct relations, or FunctioningRef?; detailed function-kind discrimination remains with A.6.F.
A conforming specialized restoration pattern states:
- the ontic, semantic area, or facet-neighborhood under repair;
- the recognition wording family selected by
E.10; - the recovered object, any direct relation use, any current claim-bearing episteme or representation, the rule that defines or constrains each, and the pattern ID that locates that rule;
- the defining, constraining, or testing rule to use directly when the value is already recoverable;
- any facet restoration pattern whose ClaimGraph defines or constrains a narrower recurring ambiguity;
- the temporary recovery product and the retained user-facing move after wording repair.
DPF-local application
When authoring or maintaining a DPF, use this shared method only when the E.4.DPF entry condition holds: recurring domain wording obstructs a named reader use. Identify the domain object or relation, the relied claim, and its use boundary in the affected DPF subject pattern; use E.10.ARCH for the recovery method and F.19 for the final plain technical rewrite. Semantic synthesis and wording restoration remain separate: restoration preserves the meaning of an identified claim in the current DPF edition; it neither establishes nor revises that claim.
Write the resulting domain entry beside the DPF claim whose wording it restores. The entry identifies the recurring wording, the relied DPF claim, the DPF edition in which that claim is current, the recovered domain object or relation, the surviving reader use, the repaired wording, and the condition that reopens either the entry or its relied claim.
Add a negative or non-use boundary only under the F.19 plausible-intended-reader test; it is not a required entry field. Do not copy the domain trigger into the FPF-wide applicability table.
One isolated wording failure stays a local F.19 rewrite, a compact E.10 route, or a direct rewrite under the DPF claim's defining or constraining rule.
Identify a separate DPF-local profile only when several entries are consumed together by a named maintained use. Identify the profile as a claim-bearing episteme, state its receiving use and edition, and keep the constituent entries independently revisable. The profile can be published as a table. Existing local profiles remain local applications of this method.
If the attempted repair exposes a missing or unstable domain distinction, stop the wording route. Return to the DPF source, semantic-synthesis, or content decision that can establish the missing claim; only then resume the wording repair.
Shared recovery algorithm
Use this five-step object-preserving order only for the ontology question that remains after the F.19 and compact E.10 pass:
-
Bound the wording and use. Name the exact text span or publication unit, the recurring wording, its sentence function and register, and the working claim or use that makes it consequential. Keep
semanticAreaBaseConcept,semanticArea, andsemanticAreaSenseFamilyas author-facing routing coordinates rather than as candidate project ontology. -
Recover the exact subject and claim. Identify what could bear the property the reader has to explain or act on. The wording may concern one entity, a kind under its membership rule, or how a practice repeatedly produces a result. A passage about one occurrence can help establish any of these when its grounds reach that far. For an obtaining direct relation, identify the relation and its actual participants. For a row, report, or card that states the claim, identify the claim-bearing episteme. When representation is current, identify both the represented object and the representation use without treating the representation as obtaining. When metonymy or compression is plausible, keep both the literal and intended candidates explicit until their defining or testing rules and exact relations or claims distinguish the uses or rule one out.
-
Bypass restoration when the subject and predicate are clear. Once the object, relation, assertion, description, publication, representation, work, result, structure, or architecture claim has a clear predicate and governing rule, state or use that claim and stop the restoration detour. Use the PatternID to locate the rule; identify an exact
ClaimGraphonly when the receiving use depends on it.If actual authoring or repair Work is part of the claim, recover each exact performer and obtaining assignment through A.13 and A.2.1, then let A.15.1 independently admit the dated Work. Add F.6 only when this architecture account also consumes precise assignment-bound attribution. State Method, MethodDescription, and responsibility claims separately only when they matter.
-
Add only receiver-needed apparatus. Add a reusable
RelationSignatureand A.6.5SlotSpecvalues, relation-occurrence identity, participant designations, designators, references, publication forms, or C.29 representation elements only when a named later use needs that additional object. Keep each item under the rule that defines its identity or use, and cite the pattern containing that rule. Use an E.24onticSlotRelationonly after durable ontic settlement makes that exact relation current.If the same entity participates in several direct relations, is designated in several assertion or occurrence-description epistemes, or corresponds to several representation elements, keep those uses distinct under their separate predicates. Shared entity identity does not merge the relations, epistemes, designations, or representation correspondences.
-
Write the shortest usable sentence. State the governed object, direct relation or claim, and admissible action-facing use so the working reader need not consult the applicability row. Add a guard only when the
F.19plausible-intended-reader test warrants it. Try the repaired sentence in the receiving work. A decision that needs the installation record and an applicable approval, for example, depends on their concerning the same installation. Carry that connection, or what remains open, into the wording.
Perform a terminology-source audit only when source ontology can change the governed object, direct relation kind, participant meaning, actual participant kind, declaration-local SlotSpec, assertion-side participant designation, exact use, admissible use, or the defining or testing rule selected for the claim. Stable ordinary prose remains ordinary. A source tuple or argument place remains representation-side until an explicit correspondence to a declared SlotSpec is current.
For unclear collective-work or methodology wording, use E.10:0.2c.7 for result, output, outcome or work product; :0.2c.7a for goal, purpose or Method concern; and :0.2c.13a for contribution. A.6.P recovers the predicate and participants; C.2.P handles an unresolved source/content/use distinction. Continue under the rule for that claim, including the conditional Method/Work branch below. Preserve clear ordinary or locally defined uses. Missing facts and an absent defining rule are different stops. Add a new restoration pattern only if §6 identifies a recurring question the existing methods do not answer.
Method, work, and P2W claim-rule constellation in wording restoration
Use this branch when one source label, project handle, or project concern points to changing, producing, selecting, deriving, controlling, or maintaining an EntityOfConcern rather than to one typed FPF value.
For the naming choice between method, practice, culture and other contextual synonyms, apply E.10:0.2c.21a. A.3.1 recovers the reusable way from the actual claim. When that meaning is clear, use the chosen explanatory name and return to F.19. The choice of synonym calls for no further restoration.
Do not name a new recovery object. Recover each current value and claim separately: one locally used U.Method; a U.MethodDescription episteme that describes that method; exact method-side relations and, only when a named use depends on their organization, the A.22-selected structure locally designated MethodRelationStructure; a mechanism or formal-substrate declaration; a mathematical-lens or other representation use; a U.WorkPlan; dated U.Work; an actual transformation; a production, inception, or completion claim; a changed-referent relation; a measurement, evaluation, choice, or decision result; an evidence, source, gate, publication, temporal, delivery, or acceptance relation; a selected structure; an obtaining C.30 ArchitectureRelation with its actual holon and structure participants; a separate ArchitectureClaim episteme; an architecture-description episteme only when that use is current; or another object named under its defining or testing rule. A.15 keeps system-role classification, assignment, Method, MethodDescription, WorkPlan, dated Work, and F.6 attribution separate. Unresolved role wording still uses E.10.ROLE.
When wording concerns a relation among methods or method families, recover the relation itself—for example, serial or parallel composition, guarded choice, iteration, refinement, substitution, decomposition, parameterization, family membership, selection, or fallback. Use A.3.1, G.5, or the rule that defines or constrains that relation. If a named use depends on how several such relations are organized, use A.22's criterion to select the structure and call it MethodRelationStructure locally; that name creates neither a U-kind nor another relation.
A graph, algebra, tuple, or other notation may represent those relations or their selected structure. Keep its correspondence to the represented subject clear; use C.29 for a separate mathematical representation or transfer question.
A claim-bearing episteme that describes the relation remains separate from it. U.MethodDescription still means an episteme that describes one Method. Do not classify one value as both U.Method and U.Mechanism unless the defining rules for those two kinds independently admit both claims.
The authoring note may record the affected entity; the exact source or practice boundary, effective scheme, model-use structure, situation, scope, or frame when it changes the claim; a change or maintained-condition claim; any current state or delta predicate; and the exact objects and relations exposed by the wording together with the rules that define or constrain them. This is a wording-repair note. Keep each project-side value under its own defining or testing rule and preserve its identity separately; the pattern ID remains a locator.
Treat input, raw material, epistemic source data or source material, output, result, outcome, deliverable, handoff, and work-name wording as triggers only while their exact relation is hidden. Once clear, bypass E.10.ARCH: use A.3.2 only for a description episteme about one exact method; A.15.2 for an intended participant or use in planned work; A.15.1 and the exact resource or participation pattern for a dated Work occurrence; A.3.4.P and the direct transformation pattern for an actual transformation participant; A.15.PROD, the measurement or evaluation pattern, or the delivery or acceptance pattern for the exact result claim; and the applicable representation rule when the word names an argument, tuple component, graph element, or other representation place.
Use C.29 only for a separate question about choosing, transferring or relying on the mathematical representation. Use C.2.P only while an unresolved epistemic source or source-to-use distinction prevents selecting or using the direct pattern. Keep physical material under its direct physical governor.
If the exact method/work-boundary relation is still hidden after generic relation recovery, apply A.6.P.WMR. Its result remains exactly one family: a positive or governed-negative direct subject-relation claim; an exact A.6.1 operation-application binding; an exact local A.15.PROD or A.6.RCD claim; or reason-specific non-assertability under A.6.P.WMR:4.3.
A failed known predicate such as EpistemeUsedByReviewWorkAsReference is factually unsupported; an unavailable ETL receiving-use fact under a known rule is missing-information; an absent relation kind and defining ClaimGraph or declaration for the health-effect claim between Patient_8472 and HE-8472 is missing-governor.
In these three examples, only the missing subject rule identifies a future relation rule or declaration need. Known inapplicability and missing constructor semantics retain the other continuations defined in A.6.P.WMR:4.3. Classification, a generic result label, a type-correct designation, planned use, or inferred opposite polarity does not close the branch.
Durable naming follows the governed value. F.18 may name a performed Work occurrence only after its A.15.1 occurrence basis is established, and it names neighboring production, measurement, evaluation, delivery, and acceptance results separately. If a proposed U.* name merely repeats a declaration-local SlotSpec label or a participant meaning stated in a direct-relation rule, keep it local unless E.24 supplies durable identity, action-facing gain, and the exact relation involved. If repeated Method, Work, or process material is proposed as durable ontology, its E.24 ontic decision and, for any public U.* kind, E.24.UK admission plus the pattern containing the kind's defining rule must precede current citation.
Applicability table
Use this reference table to select or author a shared recovery rule for the unresolved question. Apply Required recovery apparatus to the current branch under the receiver-needed rule in section 3, step 4. The trigger and product columns list alternatives; they do not require every listed object or result in one repair.
Architecture source-word recovery note. When architecture prose says that a source, document, view, ADR, diagram, dashboard, model, publication face, or source-return condition carries an architecture claim, do not mint an architecture-local Source kind. Use C.2.P first only while source expression, publication construction, carrier-relation construction, source relation, project-side reference, or non-use disposition is hidden. After recovery, use C.30.P's architecture or structure rule, C.30.AD's architecture-description and source-return rule, C.30.ASV's structural-view adequacy rule, C.32's architecture synthesis or decision rule, or the exact rule for a current evidence, assurance, gate, Work, decision, publication, or currentness claim.
Direct known claim-rule use
If the governed object, exact direct relation, claim-bearing episteme, representation use, or claim kind and its defining or testing rule are recoverable by value, use that rule directly and cite the pattern containing it. Use A.6.P.WMR only while the exact Work and Method boundary relation or exact non-assertability result remains hidden. For non-assertability, retain the reason and continuation defined in A.6.P.WMR:4.3, including a known rule's inapplicable result and unavailable constructor semantics. A missing subject rule and a missing derivation basis are different repair questions.
The selected entry conditions in sections 2 and 4 govern any remaining wording ambiguity.
Admission and extraction criterion
Add or retain a WordingUseRestorationApplicabilityRow when it meets all these conditions.
- the wording recurs across FPF-governed texts or project text deliberately using FPF-governed terms, pattern references, relation names, or conformance claims;
- the hidden primary-EntityOfConcern use field set is stable;
- the recovery apparatus or field set is stable enough to teach;
- repeated in-place repair distracts from the subject pattern's primary EntityOfConcern and first useful move;
- a useful remaining reader use survives the repair and the row helps recover it;
- no existing subject pattern carries the row without duplicating repair-only doctrine inside subject patterns.
Use an existing subject pattern when it already supplies the rule and practical guidance needed for the recovered object or claim. Record its PatternID as subjectPatternLocator and state what the rule defines, constrains or tests.
Extract material from a subject pattern when it consists solely of wording-recognition lists, false-friend rows, anti-umbrella prose, or repair fields that must run before the subject pattern can state its own invariant. Leave a narrow first-use cue or subject-pattern relation in the subject pattern.
Keep material in the subject pattern when it states the subject pattern's own invariant, worked case, conformance condition, characteristic construction, structural construction, mathematical lens, source-return condition, or user action.
First-use references in subject patterns
Subject patterns keep the minimum local first-use cues needed to resolve independent hidden questions about the EntityOfConcern, relation, claim, or field, then name the selected precision-restoration pattern through ordinary references or Relations. They do not turn that reference into local reference boilerplate, and they do not copy:
- the full
E.10wording-recognition table; - this shared algorithm;
- the
WordingUseRestorationApplicabilityTable; - broad false-friend lists whose only job is first-stage repair;
- past placement or repair history written in place of current architecture prose.
A thin pointer is acceptable when it helps the working reader choose the right first move. For example, subject patterns can give these directions.
- Use
E.10.ROLEwhile bare claim-bearing role hides its work-facing or use-facing object; return to the object's rule once it is clear. - Use
C.30.STRATwhile a stratification source label hides the FPF kind, relation, or claim-use; return the recovered claim to its defining or testing rule. - Use
A.6.P.WMRonly while an exact Method or Work boundary relation remains hidden after generic relation recovery. UseC.2.Pfirst for an epistemic source side and bypass restoration when the direct rule is clear.
The full routing conditions remain in E.10:0.2a and the applicability table in section 4. A subject pattern keeps only the pointer needed for its current ambiguity.
Name and placement discipline
semanticArea is the selected Part-F Tech term for the semantic unit used by a wording-use restoration row. Plain speech may say "semantic area" or "meaning area" only as a gloss for that declared Part-F row or bounded row-set.
Tech prose must resolve a distribution cue to its applicable routing coordinate: semanticAreaBaseConcept, semanticArea, semanticAreaSenseFamily, entityOfConcernUseFields, ontologicalNeighborhood, or a subjectPatternLocator for the defining, constraining, or testing rule. Add a realization pattern when one is needed.
pattern nest is allowed for ID and placement grouping such as A.6.*, C.16.*, or C.30.*. It is not a semantic parent relation and not an authority relation.
Examples and near misses
Read each wording with its stated use. Apply a route only while that FPF question remains unresolved. Clear ordinary wording closes under F.19.
Archetypal Grounding
Bias-Annotation
This pattern blocks semio-bias in two directions. It prevents subject patterns from becoming patterns about descriptions, records, and wording guards. It also prevents word-replacement bias by requiring recovery of the ontological neighborhood, the defining or testing rule and its PatternID locator, and the admissible reader use before a new term is selected.
Conformance Checklist
Selecting a recovery pattern selects guidance. Kind admission, the truth of an exact claim, performed Work, and attribution are established under their direct rules. The checks below apply that shared condition to particular recovery families.
Relation-use recovery rule. When wording hides a positive or explicitly restricted direct-relation claim, first name the direct predicate, its actual participants and the claim's polarity under the rule that defines it. Establish obtaining separately when the claim affirms that the relation obtains. Individuate one U.Relation occurrence only when a named receiving use needs to distinguish it. Add a reusable RelationSignature and A.6.5 SlotSpec values only when reusable typed declaration is current. Add participant designations only inside a current assertion or relation-occurrence-description episteme; the designations do not replace the actual participants.
A filled project row that states the claim is a claim-bearing episteme. A field, edge, diagram element, or table cell remains a publication form or C.29 representation element unless a defining rule establishes another object and an explicit correspondence. Use an E.24 onticSlotRelation only after durable ontic settlement makes that direct relation current. A mathematical tuple or argument position stays representation-side until an explicit correspondence relates it to a declared SlotSpec.
Using E.10.ARCH creates none of those objects; the author returns to the rules that define, constrain, or test them, with the PatternIDs serving only as locators.
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits. Wording-use restoration stays distributed and coherent. Subject patterns stay centered on their objects. Recurring hidden-field families get one recovery architecture instead of many local catalogues. DPF authors reuse that architecture without moving domain meanings or trigger lists into FPF.
Costs. Authors must decide whether the current case is local E.10, a subject pattern, an existing restoration row, or a new row with a stable recovery shape.
Risks avoided. The main risks are semio-bias in subject patterns, lexical substitution without kind recovery, and pattern-nest or placement language masquerading as semantic-area architecture.
Rationale
Repair-only trigger prose migrates into subject patterns and begins to compete with their primary EntityOfConcern and first useful moves. One symptom is a Solution dominated by warnings about descriptions and publications. F.19 removes a proposed rhetorical boundary when it lacks independent local ground for a plausible intended-reader mistake or makes no difference to understanding or use.
A warranted guard may still be misplaced. Keep it with the rule for publication pragmatics, description pragmatics, admissible use, or the neighboring subject claim it qualifies. The subject pattern's Solution remains about its own EntityOfConcern and first useful move.
The workable split is whole-span repair in F.19, compact recognition and routing in E.10, shared ontological recovery here only when needed, and local realization only where a named semanticArea has stable row identity, a stable field set, an ontologicalNeighborhood, and a useful remaining reader use.
SoTA-Echoing
Practice question. When the same wording problem recurs across subject patterns, what distribution keeps the repair reusable while letting practitioners reach the exact object, relation, claim, and defining or testing rule without filling an authoring form?
Selected best-known line. Use ISO 704's object, concept, definition, and designation discipline for terminology work; use SKOS only when a controlled-vocabulary or knowledge-organization object is current; and distribute each FPF claim to its direct subject rule through one shared trigger scan and one shared recovery architecture. Keep a local realization only when a stable recovery shape and remaining reader use recur.
Alternative or default. A central preferred-word list makes labels consistent but cannot establish obtaining relations, actual participants, claim-bearing epistemes, or direct rule ownership. Copied local warning tables stay close to each subject but duplicate recognition, drift independently, and displace the subject pattern's working problem. The selected line keeps the useful locality of direct owners and the reuse of one shared entry.
Bounded comparison and accepted trade-off. Take the section 0 case in which MethodDescription_NormalizeCustomer says “input x is CustomerRow_17” beside normalize(x). Give each distribution the same source sentence, formula, and one applicable recovery entry. A preferred-word list can standardize the label input, but that alone leaves open whether it names an argument representation or a declared participant slot. Both a correct subject-local table and the shared architecture can recover Arg_x and its correspondence to CustomerRow_17, then state the recovered representation relation.
They yield the same sentence: “In the worked formula normalize(x), argument x represents CustomerRow_17.” Because that use is already clear here, both approaches can stop at the sentence; C.29 is needed only if a separate mathematical representation or transfer question remains.
For an unresolved recurrence, a local table offers a nearby lookup; the shared entry may require a cross-reference. In return, the shared distribution keeps the recognition and recovery rule in one maintained place, while copied tables require their copies to be kept consistent.
Adopt the shared entry with direct-known-rule bypass, accepting that possible extra lookup to avoid duplicated rule maintenance. Preserve a local realization when its stable field set and remaining use justify it. This qualitative comparison establishes the distribution choice for the stated recurring case; it claims neither measured speed nor a benefit from centralizing the subject rules themselves.
Lower only the affected row when E.10 can close it locally or a subject rule absorbs the same recovery at equal or lower effort.
Relations
-
E.4.DPFdefines the DPF entry condition and the placement rule. Each domain entry remains beside the DPF claim whose wording it restores. A separate local profile exists only for a named maintained multi-entry use and can be published as a table. -
Use
E.10to recognize and close local wording issues or select the applicable row; useE.10.LRNonly while claim-bearing learning wording hides the changed subject, Work, Method, result, evidence, or use; useE.10.DEVonly while development or evolution wording hides the changed or represented subject, continuity or membership, posture, direction or value basis, direct owner, or receiving use; useE.10.ROLEfor bare claim-bearing role, which has no default Tech reading. -
E.10.ROLEprovides the thin first entry from bare claim-bearing role.A.6.RSIRrealizes first-level recovery for the narrower relation, signature, interface, assignment, declaration-slot, operation, and representation cluster only until the defining or testing rule for the recovered object is clear. -
A.6.Prealizes the shared algorithm for generic relation construction and retained relation specializations. AnA.6.P.WMRapplication records exactly one result family for a current Work and Method boundary claim: an exact direct subject-relation claim, positive or governed negative; an exactA.6.1operation-application binding; a localA.15.PRODclaim or another local relation-bearing claim selected underA.6.RCDdisposition 2; or exact non-assertability with its reason and continuation under A.6.P.WMR:4.3. A missing subject rule may identify a needed future relation rule or declaration; missing constructor semantics retain their separate derivation question. UseA.6.RCDonly for the residual needed-claim derivation and relation-kind admission question after exact participants are known and no lighter current rule closes the receiver. -
A.6.Frealizes function-like kind and relation recovery. -
C.2.Prealizes source-expression, episteme, publication, and FPF-governed-use recovery. -
C.2.P.DRrealizes declarative representation and imperative-metaphor overread repair. -
A.3.1defines one exactU.Methodand the method-side relations within its scope. When a named use depends on organization among several such relations, use A.22's criterion to select the structure, which may be locally designatedMethodRelationStructure. Enter through the method-like wording row only while the actual object or relation remains hidden. -
A.3.2defines membership for aU.MethodDescriptionepisteme that describes one exactU.Method. -
A.6.0defines reusable signature identity andA.6.5definesSlotSpecdeclarations;C.29develops a mathematical representation and its correspondence when a representation choice, transfer or reliance question needs that work;A.6.1defines operation application andE.20defines governing-definition assignment. An exactonticSlotRelationexists only after its E.24 durable ontic settlement is current. -
Use
A.15.2for planned work,A.15.1for dated Work, andA.10for a descriptive account of sources, independently established support relations, currentness and bounded reliance that method-like or path-like wording may otherwise hide; useA.15.PRODfor the local production-work, entity-identity-inception, or production-completion claim when that exact WMR result family is current. -
E.18defines graph paths, path slices, flow valuations, and graph relations over a selectedTransformationFlowStructurewhen the graph claim is current. -
C.30.Prealizes architecture and structure wording recovery. -
C.30.STRATrealizes stratification and source-label wording recovery before the recovered claim is handled under its defining or testing rule; the trigger family is in the applicability table. -
C.16.Prealizes characteristic and scale wording recovery. -
C.16.Qrealizes quality characterization and evaluative characterization wording recovery. -
E.10.MOVEresolves ambiguous move-, readiness-, route-, path-, or trajectory-like wording and exits to the direct pattern; afterE.10.DEV, it opens only for a remaining independent path ambiguity. -
A.19.SPRrealizes state-family wording recovery only while the exact object, state frame, value, or direct rule remains hidden. -
Use
F.18for durable reusable naming after the kind under repair or relation is known. -
F.19owns the whole-span semantic and pragmatic reading and the final plain technical rewrite; deeper recovery opens only for the FPF question that remains unresolved. -
E.8states the pattern-form and placement rules. -
E.19checks distribution preservation during review and refresh. -
E.11states the entry-distribution rules for broad or old-term cases across README scenarios, ToC query cues, local Problem frames, and worked entry comparisons inE.11.
E.10.ARCH:End
Recover What “Role” Means in This Claim
Type: Lexical and ontological precision restoration (E)
Status: Stable
Plain name: Recover what “role” means here
Use This When
Use this pattern when a sentence uses role to make a claim but leaves the object or relation unclear.
A system role is a local kind whose candidates are independently admitted
U.Systemindividuals, such as a person, team, organization or non-human technical object. Its membership condition distinguishes a stable, assignable, work-facing contribution under A.2 and C.3. The name creates no admission, assignment, agency, capability, or Work.
A system-role assignment is one exact occurrence under U.SystemRoleAssignment. The bare word role identifies neither object.
First useful result. Rewrite the ordinary domain sentence so that its recognizable object and action or relation are explicit. Select the pattern for that object or relation. Stop there unless the receiving claim needs a technical designation, exact occurrence, predicate, assertion, or reference.
For example:
- “Alice is reviewer” may remain ordinary recognition prose;
- “Alice holds the review assignment for this manuscript” makes an assignment claim current;
- “the report plays a role in approval” normally needs an evidence-use, source-use, reliance, or other direct relation, not a system-role assignment;
- “the first role in this tuple” normally points to a representation position, not a system-role kind.
Not this pattern when. Keep ordinary or quoted wording unchanged when no FPF claim relies on the word. When the object and its governing pattern are clear, use that pattern. Use A.6.RSIR when the unresolved question is specifically about participation in a direct relation, a relation declaration, an interface, or a representation position.
Problem Frame
Readers meet role in organizational, engineering, mathematical, software, documentary, and ordinary language. The same spelling may point to a local classification of systems, one assignment occurrence, participation in a relation, a declaration slot, a position in a representation or organization, functioning, capability, Work, authority, responsibility, use of an episteme, or no technical claim at all.
A definition given elsewhere cannot determine what role means in every sentence. Replacing every occurrence with SystemRole would turn unrelated participation, slot, representation, evidence, and ordinary-language claims into a new ontology.
Problem
A cold reader needs to answer two questions at the point of use:
- What exact object or relation does this sentence claim?
- What useful action or conclusion depends on making that distinction?
If the text answers neither question, a fluent rewrite can preserve the word while changing the claim. A mechanical replacement can introduce a further error by assigning a report, schema field, relation participant, or diagram position to a system-role kind that was never intended.
Forces
Solution
Use the sentence's intended claim, not the trigger word, to select the result.
- Quote or locate the bounded phrase only when its source identity matters.
- Write the ordinary sentence the reader should understand, naming the recognizable object and action or relation.
- Select one branch below from that recovered claim. The examples are non-exhaustive; they illustrate result families and do not define a new role taxonomy.
- Apply the selected pattern only as far as the receiving claim needs. Add a Tech designation, occurrence identity, predicate, assertion, reference, evidence, or assurance only when omitting it would change truth, action, reuse, or reliance.
- For one recovered claim, stop when you have selected its particular object or relation and the applicable pattern. Otherwise identify the claim that lacks a governing rule (
missing-governor), state what information is missing, or return the quote-only or ordinary-non-use result.
The same phrase can carry several claims. Give each its direct rule and keep the connection the reader needs. “Alice reviewed the manuscript under her assignment” depends on the work and the assignment being hers for this manuscript. Ambiguity alone leaves it open which claim is meant.
Use A.6.C Contract Unpacking for Boundaries only when its contract-like boundary-language trigger holds; otherwise apply the direct rule for each recovered claim. Do not create a multi-claim record or an umbrella role object.
Boundary with A.6.RSIR
E.10.ROLE starts from the ambiguous word and recovers the sentence's work-facing or use-facing object. Use A.6.RSIR for the narrower question of direct-relation participation, reusable declaration, interface, operation declaration or binding, and representation position.
If the recovered claim leaves a direct-participation, reusable-declaration, interface, operation-declaration-or-binding, or representation-position question unanswered, apply A.6.RSIR to that question. If it recovers a system-role kind, assignment, capability, Work, deontic relation, evidence use, another direct object, or ordinary non-use, apply that object's direct rule and do not apply RSIR. Neither pattern duplicates the other's subject rules.
The Ordinary Wording as a Result
Ordinary repaired wording is enough. This example assumes that the source establishes the stated evidence use. Include blocked overread only when independent local evidence makes that rejected reading plausible and the boundary changes the receiving use:
No separate repair record is required unless another named use must inspect or reuse the decision.
Archetypal Grounding — Worked Slices
Alice Is Reviewer
“Alice is reviewer” may stay as readable recognition prose when no technical classification claim is consumed. If only the local kind matters, identify ReviewerSystemRole through C.3 and A.2. If a technical classification judgment matters, use A.2 and C.3.2 and keep Alice, ReviewerSystemRole, the current KindSignature edition, the context slice, and the true | false | unknown result recoverable.
If assignment identity matters, first name the declared assignment species JournalReviewAssignmentRelation and establish that its direct predicate obtains for the actual participants, including Alice as holder, under the stated applicability. Only then identify ReviewAssignment-82 as the occurrence and state its extent. A roster, appointment label, description, or evidence item may support the assignment claim but does not make the assignment obtain.
If performed Work matters, first recover Alice's A.13 core for this action and independently admit ReviewWork-82 under A.15.1. Because this branch also says that Alice performed the Work under ReviewAssignment-82, F.6 afterward relates the already admitted Work to that same assignment.
Classification, assignment, Work, and attribution remain separate. None follows merely from the ordinary sentence. A short projection may omit an assignment identifier unused by its receiving claim only when every relation the claim consumes remains recoverable.
A Report Plays a Role in Approval
The report is an episteme. For the evidence-use reading established by the source, write “reviewers used Report-R as evidence for ApprovalClaim-C” and keep the support and limits needed for that claim recoverable. Use A.10 when the source-to-use account remains implicit or the receiving use needs it replayed. Use B.3 only when an actual named assurance claim is current, including when an applicable material-reliance threshold requires it.
API Provider Role
“The API role is provider” does not yet reveal which claim is meant. It may hide several claims, but the wording alone does not establish that; recover the intended claim before selecting and applying a rule. First ask whether a provider System is current and whether its classification under a local provider system-role kind matters. If the assignment itself matters, name its declared assignment species and the obtaining occurrence separately; do not infer either from root-family typing or the word role. Provision, service, declaration, interface, schema position, publication, promise, and access claims each use their own pattern.
Only when provider Work is current, recover every precise performer's A.13 core and independently admit the dated Work under A.15.1. Add F.6 afterward only if this provider account also needs precise assignment-bound attribution through the same obtaining assignment.
Passive Test Article
A passive test article may independently pass A.1 and be classified under TestArticleSystemRole. If an assignment claim matters, name its declared assignment species before identifying an occurrence. Neither classification nor assignment makes the article an agent or performer.
The source claim is TestArticle-7 participates passively in TestWork-9 during TestInterval-9; its intended participant order is article, Work, then applicability interval. No current pattern supplies a direct passive-participation predicate with those participants, applicability, and occurrence identity, so the current result is the A.6.RCD missing-governor for that exact attempted claim.
Any tester or test-rig Work first reuses every precise performer's A.13 core and is independently admitted under A.15.1; add F.6 only when the tester account also needs precise assignment-bound attribution. A short projection may omit an unused assignment identifier, but it keeps the article, Work, interval, and missing relation recoverable.
Bias-Annotation
Lenses: Gov, Arch, Onto/Epist, Prag, Did. Scope: Universal for claim-bearing uses of role that enter this pattern; ordinary and quoted non-uses remain outside it.
The pattern favors Onto/Epist precision, which can tempt an author to expand every sentence into technical apparatus. Writing the ordinary sentence first, adding only distinctions used by the receiving claim, and stopping when the applicable direct rule is clear preserve Prag and Did usefulness, while the separate branches preserve Gov and Arch boundaries.
Conformance Checklist
- Is role being used in a claim, rather than only in ordinary or quoted wording?
- Does the repaired ordinary sentence name the recognizable object and action or relation?
- Was the result selected from the recovered claim rather than from the word alone?
- If a system-role kind is current, are its candidate domain, work-facing membership distinction, member/non-member boundary and continuity rule recoverable through C.3 and A.2? If a technical classification claim is current, are the candidate’s independent A.1 system admission, exact kind, current
KindSignatureedition, context slice, andtrue | false | unknownC.3.2 judgment separately recoverable? - If an assignment is current, can you identify both the assignment occurrence and its declared species? Does that species' direct predicate obtain for the actual participants, including the holder, under the stated applicability, and is the occurrence's extent recoverable? Is supporting evidence kept separate from whether the assignment obtains?
- Are participation, declaration slot, operation binding, interface place, and representation position kept distinct?
- Are functioning, capability, Work, deontics, access, authority, responsibility, evidence use, and results kept under their direct relations?
- Does ordinary prose stay ordinary when no technical distinction changes the receiving claim?
- Does the repair add a qualifier only for a live alternative and avoid a fixed formal expansion?
- Does the reader have the particular object or relation and its rule for each claim, and the grounds for connecting them when the assignment calls for that conclusion?
Common Anti-Patterns and How to Avoid Them — Role-Word Repairs
Consequences — Reopen Condition
Benefits. The intended object is stated at the point of use while natural practitioner language remains available. Technical names distinguish system-role kinds from relation slots, evidence uses and representation positions.
Costs. A claim-bearing ambiguous sentence needs one bounded interpretation before it can be reused. Some old compact Tech names must be retired or made explicitly historical.
Reopen this pattern only when an actual bare-role use cannot reach one exact object, relation, ordinary non-use, or missing governor without duplicating A.6.RSIR, or when repeated cold readers still infer system admission, assignment, agency, capability, participation, or Work from SystemRole morphology.
Rationale
The recurring problem is word-sense recovery, not a missing universal role category. For each use, recover the claim being made, then use the direct pattern for the recovered kind, assignment, participant, declaration place, representation position, or relation. The trigger word neither supplies that ontology nor makes the recovered claim obtain.
The pattern supplies an entry and a stop rule, while C.3, A.2, A.2.1, A.6.RSIR, A.6.5, C.29, A.10, A.15, and other direct patterns retain their respective predicates and identity laws.
SoTA-Echoing
No external source governs this design, and FPF imports neither an external upper ontology nor its terminology. Two current lines provide the main SoTA comparison. The gUFO account and OntoUML role documentation and specification test anti-rigidity, external dependence, and the connection to a base kind. Engineering-function work tests the separation among function, behavior, and capability. FPF uses both lines as bounded comparators and adapts their tests to keep holder identity, changing classification, assignment, capability, functioning, and Work distinct.
Other comparisons have narrower uses. FPF adapts Toyoshima's role facets as diagnostic questions about position, specification, and potential. It uses DOLCE, BFO, and CCO as bounded comparators for dependence and the distinctions among persistent entities, processes, roles, dispositions, capabilities, and functions. It keeps DnS as lineage for separating descriptions from entities participating in described settings. These comparisons supply distinctions and counterexamples, not FPF kinds or assignment identity.
Within FPF, E.10 and E.10.ARCH define trigger recognition and recovery distribution. A.2, C.3, and C.3.2 define system-role kinds and classification judgments; A.2.1 defines assignment occurrences; A.6.RSIR defines direct-relation, declaration, interface, and representation recovery; and F.19 constrains the final plain wording.
These patterns support an entry that starts with the ordinary sentence. They cannot supply a missing direct relation or establish that the recovered claim is true. Reopen this choice if one of those patterns changes that boundary, or if a stronger current comparison exposes a case in which the thin entry cannot preserve both plain language and whether the recovered relation obtains.
Relations
E.10.ROLE:End
Conceptual Prefixes policy & registry
Status: Stable
Intent. This compact, notation-neutral registry lists conceptual prefixes, the short labels that identify Core vocabularies and constructs in use. The minting policy governs the introduction of new prefixes.
Policy (normative).
- Purpose. A conceptual prefix aids reasoning by identifying the construct or vocabulary in use (for example, a meta-type, calculus operator or relation family). It does not name files, serialisations or APIs.
- Anchoring. Every prefix MUST name the Core patterns that define its constructs, operators or vocabulary, or govern admission to its namespace. List these anchors in the registry and cite them in Relations. For a family with several definitions, name the supplying patterns for the relevant members.
- No tool lock‑in. A prefix MUST NOT imply a particular notation or machine binding (see E.5.1–E.5.2).
- Minting rule. New prefixes are introduced by a DRR (E.9) that demonstrates (a) cross‑pattern need, (b) non‑overlap with existing prefixes, (c) alignment with Pillars P‑1/P‑5.
- Scope. Prefixes are globally reserved within the Core; domain patterns MAY mint local shorthands only within a named local glossary and its effective scheme and MUST NOT collide with this registry.
Registered conceptual prefixes (Core).
U.— namespace for U-kinds admitted under E.24.UK, whose dispositions cover root and dependent kinds. Spelling, dependence or reuse alone does not establish U-kind admission. Anchors: E.10:8.3, M-P1, reserves the namespace; E.24.UK:4.1–4.3 governs admission. Each kind retains the identity or membership rule in its defining subject pattern.Γ_— calculus-operator notation, for exampleΓ_epist,Γ_ctxandΓ_time. Anchors: E.10:8.3, M-P1, reserves the prefix; B.1.3:4.2 defines the synthesis and compilation operators; B.1.4:3 defines the optional notation for contextual and temporal aggregation. For other operator variants, cite their definitions. For a system aggregation or delimitation decision, use B.1.2:4 to recover the separately governed results and any missing basis.ut:— Universal relation family (e.g.,PartOfsub‑relations). Anchor: A.14 (Mereology) — informative alias vocabulary.tv:— Trace & Validation vocabulary (CT2R‑LOG):tv:AliasOf,tv:groundedBy. Anchor: B.3.5 (CT2R-LOG).ev:— Evidence-account vocabulary, used for source and support labels in a descriptive account. Each support or assurance claim retains its direct governing rule. Anchor: A.10 / B.3.mero:— informative trace labelsSumTrace,SetTraceandSliceTracefor C.13'ssum,setandsliceconstruction accounts. A materialized trace is a C.2.1 episteme; these labels introduce no additional kinds. Anchor: C.13.
Conformance Checklist (E.10.P).
- CC‑LEX‑P.1 New Core text SHALL NOT introduce an unregistered conceptual prefix.
- CC‑LEX‑P.2 Keep the prefix's meaning and applicable defining anchor recoverable. Cite the anchor where the prefix or member's meaning would otherwise be unclear; an adequate earlier definition or reference can be reused.
- CC‑LEX‑P.3 Incidental examples that expand a prefix into a concrete URI or machine syntax MUST mark the expansion informative and locate it in Tooling/Pedagogy. A pattern governing that concrete syntax may include necessary examples under E.5.1; the conceptual prefix alone establishes no machine binding.
Relations. Uses E.10:8.3 for namespace reservation, E.24.UK:4.1 for U-kind admission, B.1.3:4.2 and B.1.4:3 for the listed Γ operators, B.1.2:4 for system aggregation and delimitation decisions, and A.14, A.10, B.3, B.3.5 and C.13 for the remaining registered vocabularies. Applies E.5.1 (Lexical Firewall) and E.5.2 (Notational Independence); depends on E.9 (DRR).
E.10.P:End
Recovering What “Context” Means in Use
Type: Method pattern Status: Stable Normativity: Normative when context carries meaning needed by an FPF claim; informative for quoted source wording and ordinary prose that already makes its meaning clear.
Problem frame
Use this pattern when the word context can change a statement or the next practical move, but the statement does not yet say what supplies the relevant boundary, interpretation, or situation.
What goes wrong if missed. A reader treats a source edition, reference scheme, local sense, claim scope, model-use boundary, working situation, architecture, environment, or domain boundary as one generic Context object. The sentence then hides the distinction that should decide what to inspect, compare, change, or stop doing.
First useful result. One repaired statement names the value, relation, claim, scope, situation, or use that supplies the missing meaning and, when useful, cites the pattern that defines, constrains, tests, or supplies a method for it. The reader can then take the next subject-matter action or stop.
Not this pattern when. Keep context when it is a quoted source term, an ordinary word whose meaning is already clear, or an established Plain designation whose value, relation, scope, situation, or use is already named under the pattern that defines or tests it. Use that pattern directly; E.10.D1 does not require a second wording pass.
Problem
The word context is useful because it points toward locality, but it does not say which locality matters. Terminology work uses source schemes and local senses. Domain-driven design uses a model boundary and relations among model uses. Claims use scopes and qualification windows.
Architecture work distinguishes a described holon, actual subject relations, a selected structure, an obtaining ArchitectureRelation, and an ArchitectureClaim; viewpoint, environment, and operating-condition claims introduce further distinctions. A pattern's Problem frame describes a recognizable situation. A DPF has a domain subject, audience, source basis, and local qualification conditions.
Treating these uses as one Context participant, ContextId, or two-part SenseCell(Context, LocalSense) hides the distinctions supplied by A.1.1, A.2.6, C.2.1, F.0.1, F.17, and F.9. Replacing that proxy with another universal container preserves the failure under a new name.
Forces
Solution
Start with the sentence and the practical use that makes it matter.
- Mark the phrase containing context and state what a reader would do differently under another interpretation.
- Use the smallest branch in
E.10.D1:4.1that settles the receiving question. If a phrase combines different conditions, recover the ones needed together. For example, restarting a pump may require both the authorized procedure for this site and the operating conditions that procedure permits. - Apply the named subject pattern and recover its value, relation, claim, or situation. For source-local meaning, reuse an adequate current
F.0.1result or applyF.0.1only when that meaning remains unclear. Do not create a generic Context participant as an intermediate step. - Rewrite the sentence with the recovered content and state the next action or stop.
- If the same defect recurs across framework contributions, use the shared method in
E.10.ARCH. Keep this pattern as the word-specific branch and keep the recovered content in its subject pattern or DPF.
The result is the repaired statement. Create a separate record only when a named later use needs its identity.
Positive recovery branches
When to retain the word
E.10.D1 defines no U.BoundedContext, generic Context, universal ContextId, or two-part SenseCell(Context, LocalSense). A source or subject pattern may define a value whose established designation contains context; keep that designation and its defined meaning. A DDD bounded context is the Plain retrieval name for the A.1.1 BoundedModelUseStructure, not a universal semantic-locality container.
Do not ban anchor, domain, design, run, or context by spelling. When a subject pattern defines the word's current use, preserve it. When the word hides the claim being made, recover that claim and rewrite the sentence. A source-local expression remains quotable even when its ontology differs from FPF.
An F.1 Source-Cut Card is a memory aid for one retained source edition and its answer-changing claims. It supplies neither local meaning nor source authority. A SenseCellAddressRef designates one identified F.17 cell; the address is not the cell and does not create a Context object.
Short working script
Name the meaning, the grounds for this use and what they allow the reader to do. When several conditions must hold together, show how they apply to the same case.
The bracketed words are prompts, not a public schema. Delete them in the final prose.
Archetypal Grounding
Source-local meaning. A draft says, “In the maintenance context, service includes scheduled inspection.” The author finds an existing F.0.1 result for the current MaintenanceGuide-2026 edition: its exact F.17 cell says that service includes scheduled inspection, and its LocalSenseBasisRelation names the supporting claim episteme. That result is adequate for this sentence, so the author reuses it and writes, “In MaintenanceGuide-2026, service includes scheduled inspection,” with a citation to the cell.
If the source-local meaning were still unclear, the author would apply F.0.1 first. F.1, F.9, and F.0.2 remain closed unless source selection, a cross-local relation, or comparison of several source ontologies becomes a live question.
Model-use decision. A change note says, “The controller change stays inside the press-control context.” If the decision asks only whether PressControlModel-5 applies to Press-3 within the stated claim scope, the engineer states that ModelApplicabilityRelation and stops.
If release review depends jointly on model applicability, actual assigned-Work use, fixed-content coherence, applied constraints, and one selection-use frame, the engineer selects their A.1.1 BoundedModelUseStructure. The word context does not decide between those branches.
Claim boundary. A review says, “The comparison is valid in this context.” Name the compared bearers, comparison basis, applicability conditions, evidence basis, and intended use. When those conditions and the calculation settle the question, stop. If the receiving rule requires A.2.6 scope membership, identify the required scope and slice and apply its guard; retain any qualification window required by that use.
Ordinary non-use. A source quotation says, “Context mapping is collaborative.” If the current claim is only that the source uses this phrase, keep the quotation and cite the source. Open A.1.1, F.9, or another branch only when the receiving text relies on a model-use structure, semantic relation, or other recovered content.
Bias-Annotation
Conformance Checklist
A use of E.10.D1 conforms when it meets these conditions.
- the sentence and the action-changing ambiguity are named;
- the recovered meaning supplies the conditions the receiving use needs, including their connection where several apply;
- the repaired sentence names the relevant value, relation, scope, scheme, situation, or use and the contribution of any cited pattern;
- source-local meaning reuses an adequate current
F.0.1result or appliesF.0.1when that result is absent or inadequate, then cites a basis relation only when it obtains; - an A.1.1 structure is selected only when the organization of its direct facts changes the decision;
- a claim boundary states the applicability conditions and grounds needed by its use, applying A.2.6 scope and membership facts when the receiving rule requires them;
- an F.9 Bridge is opened only between different semantic-context projections, and its bounded-use claim remains separate;
- architecture wording distinguishes an actual selected structure and obtaining
ArchitectureRelationfrom a negative, unresolved, candidate, or expectedArchitectureClaim; - viewpoint or view wording identifies the candidate episteme and viewpoint edition and uses the E.17.0 positive, negative, or unresolved conformance result;
- environment, operating-region, or operating-condition wording returns to the subject claim and names the fact or condition that changes it, or returns an unresolved wording result;
- quoted source wording and already precise ordinary wording remain available; and
- the result gives the reader a practical next move or a truthful stop without creating a universal Context kind, field set, or card.
Common Anti-Patterns and How to Avoid Them
Consequences
The repair removes one convenient universal alias. Authors must sometimes name several values that the old word compressed, and legacy ContextId, U.BoundedContext, and two-part SenseCell fields need semantic repair rather than mechanical renaming.
The author can then use the method supplied for the subject question. Source-local meanings remain traceable to schemes and basis epistemes; model-use boundaries remain engineering structures; claim scopes retain their applicability conditions. Working situations remain readable. Obtaining ArchitectureRelation occurrences remain distinct from ArchitectureClaim content. Environment, domain, and publication claims keep their own participants and tests.
A local repair can stop after one sentence, while recurring problems can reuse E.10.ARCH without copying a second ontology into every DPF.
Rationale
Context wording can make locality visible, distinguish a working situation from a source-local meaning, and reveal the need to relate different local meanings. One universal U.BoundedContext would erase distinctions that the subject patterns define separately.
Positive recovery is preferred to a forbidden-word list. A spelling check can find candidates, but only the receiving claim tells whether the phrase hides a scheme, scope, structure, situation, or other value. Naming that content opens the next practical move; banning the word does not.
SoTA-Echoing
These comparisons support the recovery branches for the wording uses named here. They do not show that the branch set is complete for every use of context or that this method dominates every alternative. Reopen the method when a subject pattern defines a better distinction, a recurring use of context needs another productive branch, or a current practice source changes one of these action-bearing separations.
Relations
- Apply
E.10to recognize a local wording problem and make the smallest local repair. ApplyE.10.D1when context hides content that changes the statement or next action. - Apply
E.10.ARCHwhen the same consequential wording problem recurs across framework contributions. That pattern supplies the shared restoration method;E.10.D1supplies this word-specific branch. A.1.1defines the direct model-use relations and the decision condition for selectingBoundedModelUseStructure.A.2.6defines claim scopes, context slices, and their membership facts.C.2.1identifies claim-bearing epistemes and their effective schemes.F.0.1supplies source-local meaning recovery, reuse and stop. Its ordinary result is a plain answer tied to the source passage; an exact F.17 cell and a separately obtaining basis relation are conditional additions.E.10.D1recognizes the wording use and returns the repaired sentence; it does not repeat that recovery method.F.1is used only when source selection is live.F.0.2is used only when several source ontologies must be compared for one receiving claim. Neither follows automatically from a source-local wording repair.F.17definesSchemeSenseCell,SenseCellAddressRef, andLocalSenseBasisRelation.F.9defines semantic-context projection, direct Bridge truth, separate bounded-use claims, and reliance boundaries; useF.9only when the receiving claim needs that cross-local relation.C.30defines the obtainingArchitectureRelationand the separateArchitectureClaimform. Use the actual relation only when its predicate holds; use claim content for a negative, unresolved, candidate, or expected architecture statement.E.17.0defines viewpoint identity, the directEpistemeViewpointConformanceRelation, its readable positive, negative, and unresolved results, and the resulting same-epistemeU.Viewmembership.- For environment, operating-region, and operating-condition wording, use the pattern that defines or constrains the subject claim. When the affecting fact or condition cannot be recovered, keep the wording result unresolved rather than inferring architecture or viewpoint content.
- Apply
F.19for the connected reading and final plain rewrite. UseE.10.D1only for a consequential context ambiguity that remains. ApplyF.18only when the repair creates a durable reusable designation.
E.10.D1:End
EntityOfConcern, Description Episteme, and Specification-Use Discipline
Status: Stable Type: Definitional pattern Normativity: Normative, notation-agnostic
Problem frame
Use this pattern when one passage names an exact entity and also speaks of a description, specification, view, diagram, publication, file, dashboard, model, evidence item, assurance result, gate result, or decision around it. The wording fails when it makes one of those neighboring objects stand in for the entity, the episteme, or the authority for the next action.
Begin with the receiving use:
- What exact work, decision, comparison, inquiry, preservation, teaching, publication, or other use needs the description?
- What is the next unresolved question or choice for that use? Do not invent one for a use that has none.
- What exact claim content is being used?
- What exact
U.Entityis theEntityOfConcernof that claim-bearing whole? - Which effective
ReferenceSchemesupplies the designation and interpretation rules that make those claims readable about that entity?
The last three answers recover the C.2.1 EpistemeConstitutionRelation. A request for that identity can end here. If the recipient needs to decide something else, recover the additional facts on which the decision depends and, where several must hold together, establish that they concern the relevant edition, participants or work.
Not this pattern when the live question is already an exact evidence path, assurance claim, work occurrence, gate decision, commitment, Bridge, publication occurrence, representation correspondence, or state fact. Use its direct governor. Return here only if the wording also obscures which entity is being described or which episteme carries the claims.
The working distinctions are:
- the EntityOfConcern is the independently identified entity about which the selected claim-bearing whole makes its claims;
- a description episteme is an ordinary
U.Epistemeused to carry descriptive claims about that EntityOfConcern; - a describing use names the receiving use and may select one exact viewpoint when that selection changes what is read or checked; selection changes neither episteme identity nor conformance;
- specification use is a checkable use of a description episteme, not a third peer ontology class;
- viewpoint, view, claim scope, model-use structure, grounding, evidence, edition, publication, carrier, and representation remain neighboring objects and relations.
The reader can now say what is described, which claim-bearing episteme is being used, what the receiving use needs next, and where any additional claim is governed. A formal-looking file, card, suffix, approval, or diagram gains no ontological or practical authority by appearance.
Problem
- Entity-description collapse. The EntityOfConcern is identified with the episteme, diagram, card, file, dashboard, or work record that says something about it.
- Record-shaped constitution. A local tuple, filled card, context record, or field list is treated as what makes the episteme exist.
- Specification inflation. Detailed or official-looking prose is called a
...Specalthough no checkable claims and no exact harness or validation relation are present. - Neighbor collapse. Viewpoint, view, claim scope, model-use structure, grounding, evidence, edition, publication, carrier, and representation become fields of one omnibus description object.
- Use-free qualification. Context, scope, structure, currentness, or publication machinery is required without naming the receiving use that needs it.
- Agency and authority leakage. A description, standard, card, approval label, or publication is said to perform work, authorize action, or establish a world-side fact without its direct relation.
Forces
Solution
For the current passage or artifact:
- Name the receiving use. State the exact work, decision, inquiry, comparison, preservation, teaching, publication, or other use and what it needs next.
- Recover the episteme constitution. Identify the exact claim content G, exact EntityOfConcern, and effective
ReferenceScheme; test whetherEpistemeConstitutionRelationobtains under C.2.1. - Classify the expression or use. Decide whether the current object is the claim-bearing description episteme, a specification use of it, an assertion about another object, a publication form, a carrier, or a representation. Do not infer the answer from a suffix or medium.
- Open only a needed neighbor. Add empirical grounding, viewpoint, view, claim scope, model-use structure, evidence, edition, specification evaluation, publication, form, carrier, currentness, or representation only when the named receiving use depends on its direct relation.
- Stop at the smallest sufficient result. Name the receiving use, the recovered episteme and its EntityOfConcern. Add the relations on which the answer depends. For example, deciding from an approval and a use record whether an operator used the approved instructions requires both to identify the same edition.
A short statement usually suffices.
For
<receiving use>, episteme<E>carries claims<G>about EntityOfConcern<T>under scheme<R>.<The additional relation or combination of relations>supports<this conclusion>under<these conditions>.
If no neighboring relation is needed, omit the second sentence. If the C.2.1 triple or the required direct governor cannot be recovered, return that exact blocker instead of filling a generic context field.
Core recovery discipline
EntityOfConcern
EntityOfConcern is the one exact independently identified U.Entity about which the selected claim-bearing whole makes its claims. It may be a system, work occurrence, method, episteme, direct relation occurrence, characteristic, structure, pattern, or another admitted entity. It is neither a universal object bucket nor the authoring target merely because the author is editing it.
A ClaimGraph may designate several other entities as participants in relational, comparative, negative, counterfactual, or modal claims. Those designations do not by themselves create a joint EntityOfConcern. Select a relation occurrence, collection, or structured whole only after its direct pattern independently identifies that entity.
Description episteme
In “revise the description,” the word description can designate the same U.Episteme E identified under C.2.1: its claim content G states claims about its EntityOfConcern T under its effective ReferenceScheme R. The word description supplies no additional kind or identity criterion. When the instruction concerns selected claims, wording, layout or describing work, identify the object of that action. C.2.1:4.0.5 and :6 explain these readings and the specification-use boundary.
Descriptive claims may assign designations, characterize the entity, state criteria, or describe structure or behavior. A diagram may express those claims under the effective scheme and separately serve a representation use under §8.4. Their expression does not by itself establish parts or properties of the EntityOfConcern; recover those subject-side facts under their direct patterns.
For one named describing use, state the exact viewpoint P it selects when that selection changes interpretation or action. Keep the episteme, its EntityOfConcern, the use, and P distinct. The selection is not an episteme identity discriminator and establishes neither viewpoint conformance nor U.View membership.
Specification-use admission
Use a ...Spec name only when the receiving use depends on specification force and satisfies the applicable conditions in C.2.1:6. That section supplies the exact episteme, checkable claims, checking means and reliance-relevant viewpoint conditions. When those conditions are absent, use description for the account and present proposed criteria as proposals. A suffix alone supplies no specification force.
Model-use structure
A BoundedModelUseStructure is selected only when the receiving assertion, calculation, interpretation, comparison, or other use depends on the organization of admitted model-applicability, model-use, and coherence relations governed by A.1.1. The receiving use designates that exact structure through its direct relation. The structure is never a constituent of description-episteme identity merely because the episteme is used inside it.
If a proposed dependent relation species requires one exact model-use structure as an identity-bearing participant, its own pattern must declare that participant and its obtaining and identity rules. E.10.D2 supplies no generic context relation as a shortcut.
Episteme about an episteme
When an episteme is described, use C.2.1's ordinary recursion. A review episteme about another episteme has that earlier episteme as its EntityOfConcern and its own ClaimGraph and effective ReferenceScheme. Self-description is also possible when the episteme's identity remains recoverable; then its EntityOfConcern designation resolves to itself. Publication and representation uses remain separately governed, and self-reference supplies no assurance by itself.
Naming discipline
Default suffix. Use ...Description when naming a description episteme for a practitioner-facing use.
Reserved suffix. Use ...Spec only when the specification-use conditions above obtain. Do not use it as a synonym for detailed, official, approved, formal-looking, or stored in a schema.
Entity names. Name the EntityOfConcern by its independently governed kind and identity: one exact local system-role kind, Method, System, ArchitectureRelation, Characteristic, PromiseContent, Work, Episteme, or another exact kind. Append Description, Spec, View, Publication, Form, Carrier, or Representation only when that neighboring object is what the name designates.
Relation language. Prefer the direct governing verb: a description carries claims about an entity; a publication occurrence makes an edition available; a carrier bears a form; a representation corresponds under a scheme; evidence supports an assertion; an admitted system performs work. Do not turn those verbs into one generic description link.
Ambiguous role language. When source wording says that a description, source, standard, requirement, evidence item, publication, dashboard, or view “has a role,” recover its exact evidence-use, source-use, standard-use, requirement-use, publication-use, assurance-use, or gate-use relation. For a claimed Work use, name the exact premise, governed reference, decision-use relation, or A.6.1 operation-argument binding and its actual participants.
If the claimed use needs another relation and no direct governor supplies its predicate and participants, return missing-governor rather than inferring a universal description-to-Work or episteme-to-Work relation.
Open one exact occurrence of a directly declared U.SystemRoleAssignment species only when an independently admitted U.System is assigned to one exact local system-role kind for the bounded work; an acting holon is eligible only after that exact entity has independently passed U.System admission for this claim.
Invariants
D2-1 (Direct constitution). Every description episteme is identified through the exact C.2.1 <ClaimGraph, EntityOfConcern, effective ReferenceScheme> constitution; no local record or tuple replaces it.
D2-2 (Entity-description distinction). Recover the independently identified EntityOfConcern and the claim-bearing episteme without treating a description as its subject merely by appearance. They may be the same episteme in a C.2.1 self-description; different constitution triples identify different epistemes.
D2-3 (Specification is a use). Specification force requires checkable claims and a named harness or validation relation. When viewpoint selection affects reliance, preserve or update the named describing use and its exact selection. Specification is not a peer class or label effect.
D2-4 (Conditional neighbors). Grounding, viewpoint, view, claim scope, model-use structure, evidence, edition, publication, carrier, currentness, and representation enter only through their subject patterns when the receiving use depends on them.
D2-5 (Stable identity across use). Changed description use, viewpoint selection, harness, evidence, publication, form, carrier, rendering, or representation does not by itself change episteme identity.
D2-6 (Work and authority separation). An account of Work identifies its performing System under A.15.1. Epistemes carry claims and may be used in that Work under independently governed use relations. Permission, acceptance, assurance and world-side claims retain their direct governing rules; a description or display supplies none by label.
D2-7 (No label-only sameness). Identical labels across schemes, scopes, viewpoints, model-use structures, or contexts establish neither the same EntityOfConcern nor the same episteme. Use the governing identity and Bridge rules.
D2-8 (Representation separation). A tuple, card, graph node, schema field, notation token, file, or UI element may participate in a representation or publication of a recovered object; it is not that object by position or appearance.
D2-9 (No generic context relation). E.10.D2 defines no U.EpistemeSlotRelation, positive DescriptionContext value or tuple, BoundedContextRef constituent, mandatory context recursion, or universal description relation. The old names may appear only as explicitly rejected source wording.
Recovery decisions
Neighboring use routing
Open a neighboring object only after naming the receiving use and recovering the description episteme. The same episteme can participate in several of the uses below. Each use retains its subject pattern, participants, obtaining condition, and identity.
Describing use, viewpoint, and view
For one named describing use, state that the use selects one exact U.Viewpoint episteme P when that selection changes what is read or checked. This selection says from which concern-bearing viewpoint the already identified episteme is being read for that use.
That selection:
- does not change the C.2.1 identity of the episteme being read;
- does not establish
EpistemeViewpointConformanceRelation; - does not admit or remove same-individual
U.Viewmembership; - selects no receiving view and performs no A.6.3 viewing construction;
- may change between two describing uses while the episteme remains unchanged.
Call the same episteme a U.View only when it conforms to at least one exact U.Viewpoint episteme under E.17.0's fixed membership rule. Direct authoring and A.6.3 source-to-receiving construction can produce an episteme but grant no view membership. A rendering, publication form, or carrier-borne display is not a view by appearance.
If one use selects several viewpoints, keep each identified. Use C.13 only when that use depends on the plurality as a collection and A.22 only when it depends on their organization; do not overload one context qualification.
Scope, model use, grounding, evidence, and currentness
Use A.2.6 when the receiving use depends on the exact claim scope and its context-slice membership. Use A.1.1 when the receiving use depends on one exact BoundedModelUseStructure. Neither scope nor structure becomes a description constituent merely because a table displays it.
Use C.2.1 empirical grounding only when claims must be mapped to exact observation, intervention, measurement, or test relations involving one grounding holon.
Retain an adequate evidence-provenance basis. Use A.10 when the source-to-use path is still implicit or a replayable account is needed. Use B.3 when an assurance claim is made or its material-reliance threshold is met.
Evidence, assurance, or an evaluation result can support an assertion about a description or its specification use; none makes the subject-side claim true, changes the EntityOfConcern, or mutates the description episteme. State the exact validity or reliance window when that receiving use depends on one.
Retain an adequate currentness basis for the description edition, evidence path, harness, viewpoint, publication or other neighbor. Use G.11 when a changed premise or condition may alter that basis and the receiving use needs a refresh decision. A currentness judgment applies to that exact object or relation.
Where claims cross reference schemes, first recover the exact F.17 source and receiving senses and the obtaining F.9 Bridge needed by the direct use. A separate current C.2.1 claim states whether that Bridge is suitable for the named bounded use, direction, correspondence rule, and loss tolerance.
For reliance on that claim, preserve the A.10 evidence-provenance path for that same use. Ordinary evidence reliance requires the applicable local RelianceDisposition, while an actual named assurance claim requires its B.3 result for the same bounded assurance use. A Bridge, profile, card, or shared spelling is neither a licence nor proof that comparison, translation, or work occurred.
Edition, publication, form, and carrier
Changed ClaimGraph, EntityOfConcern, or effective ReferenceScheme identifies another episteme under C.2.1. When the receiving use also claims continuity between two epistemes, use the exact C.2.1 edition relation. A version label, file history, publication order, shared name, or collection membership establishes neither another episteme nor edition continuity.
Use E.24.PUB to distinguish these publication-side objects and relations:
E.24.PUB defines PublicationExpressionRelation(E,X,U), PublicationExpressionBearingRelation(C,X) and exact EpistemePublicationRelation(E,A,U,X,C). A selected reusable F adds X/F conformance. Its form-mediated relations preserve their quantified uses with the same X as witness for availability.
Rendering, printing, uploading, indexing or access-control Work remains separately governed. Plain “published episteme” names contingent participation in a publication occurrence.
For an inspection card, distinguish the organized claims that constitute E, the filled expression X, reusable layout rules F and the sheet or file C. A publication occurrence can make E available through X and C to a maintenance team for one bounded use. Recover the object needed by the question and its actual predicate.
Representation
For a representation use, recover the identified expression, its correspondence to the subject, and the interpretation and operations supported by that correspondence. A.6.3.RT and A.6.3.RT.OE explain expression construction. C.29 supplies additional grounds for mathematical inference or transfer. What can be returned to the subject follows from those operations and their conditions.
A diagram therefore has distinct branches:
- if its exact ClaimGraph, EntityOfConcern, and effective ReferenceScheme satisfy C.2.1, the selected claim-bearing whole is an episteme;
- if that same episteme conforms to an exact viewpoint, E.17.0 may admit it as a
U.View; - its particular signs and arrangement supply expression X; reusable rules can supply F, and representation correspondence is checked for the intended operation;
- a screen, sheet or file can bear X under its selected carrier identity;
- a publication occurrence may make one selected episteme edition available.
No branch follows from visual appearance, generation history, a heading, or a repository path.
Work, status, and authority
Only admitted systems perform authoring, evaluation, revision, publication, viewing, query, rendering, and use work under the corresponding work relations. The resulting episteme, publication, carrier, trace, or evaluation result does not perform that work.
Epistemic and deontic statuses over epistemes are not SystemRoleAssignmentStateRelation occurrences, system states, or runtime facts about the EntityOfConcern. A gate verdict, permission, commitment, acceptance, requirement use, standard use, source use, or Work authorization needs the pattern that defines, constrains, or tests that claim. Neither a description nor its publication grants those effects by label, approval mark, or availability.
Archetypal grounding and bias annotation
System case. A service-interface description carries claims about one exact system interface under its effective scheme. The interface is the EntityOfConcern. A selected viewpoint for a safety review, a conformance harness, a publication to operators, and a deployment gate are four separate uses; none belongs in episteme identity.
Episteme case. A DRR, pattern, safety case, source set, or model episteme can itself be the EntityOfConcern of another episteme. A review note about it uses ordinary C.2.1 recursion. Its dashboard, PDF, publication, evidence path, and review work remain separate regardless of which one the reader first encounters.
Card and diagram case. A filled card or diagram can be a claim-bearing episteme when its C.2.1 constitution is recoverable. Its layout can separately be a publication form or representation, and its file can be a carrier. Filling or displaying it makes no subject relation obtain.
The dominant bias is substitution by the most visible object: a reader sees a file, diagram, dashboard, card, label, or status and lets it replace the independently governed entity, episteme, relation occurrence, or authority needed for the decision. Replacing the word leaves that substitution unresolved. Recover the exact object and direct relation for the named receiving use.
Anti-patterns and repairs
Worked examples
Each example begins with a receiving use and stops after the smallest sufficient recovery. The examples require no generic description record.
Description of a system-role kind
A method author needs readers to recognize the local kind currently named ChangeAuthoritySystemRole before checking any assignment. The kind is recovered through its system-candidate domain, work-facing membership condition, member/non-member boundary, and continuity rule; OperationsReview provenance locates that definition but does not identify the kind. ChangeAuthoritySystemRoleKindDescription is a C.2.1 episteme whose exact EntityOfConcern is that independently admitted kind, whose ClaimGraph describes the distinction in readable terms, and whose effective scheme is the selected operations-review reference scheme.
For the named operations-review use, record the exact operations viewpoint only if it changes which work-facing claims the review reads or checks; otherwise name the use and omit viewpoint selection. The ClaimGraph may cite credential criteria, a mandate window, separation-of-duty constraints, capability expectations, and a direct SystemRoleAssignmentStateRelation when those neighbors are current. The system-role-kind description contains none of the assignment, checklist, graph, criteria, or relation occurrence. The description admits no holder and creates no U.SystemRoleAssignment.
The receiving use needs recognizability, not specification force, so the practitioner stops with the description. A specification use opens only if exact checkable system-role-kind claims and their checking harness are named.
Method description
A team wants to teach BacklogRefinement, an independently admitted U.Method. BacklogRefinementMethodDescription is one C.2.1 episteme about that exact method. A.3.2 admits the same episteme as U.MethodDescription only when its claims make a substantive statement about the method as a way of doing—for example its applicability, preconditions, effects, bounds, enactment concern, or internal composition.
The organized claims expressed by a practice card may constitute that episteme. Its filled expression X can follow reusable form F and be borne by sheet or file C; any representation correspondence is established separately.
Classify a calendar session, chat thread, or ticket update from its actual facts as a Work occurrence, assertion or record, publication-side object, or another object whose kind and applicable relation are already known; medium and label do not decide. An assertion or work record whose claim depends on the exact method-description edition may cite that edition through the exact premise, typed reference, or A.6.1 operation-argument binding required by the receiving claim.
A.15.1 separately states the exact relation by which an actual dated Work occurrence enacts the admitted Method; the method-description episteme is not a participant of that relation. Bibliographic metadata, approval, or a method label alone grants neither method-description membership nor specification force.
Architecture description and view
An architecture review asks how PaymentService is organized for the operations concern. The architecture-description episteme has that exact holon as its EntityOfConcern; its claims identify the selected structure and state whether C.30's ArchitectureRelation obtains. The named review use selects the exact operations viewpoint because that concern changes which claims it reads and checks. If it did not change the reading, checking, or permitted conclusion, the use would remain named and the viewpoint selection would be omitted.
The episteme is a U.View only if the E.17.0 conformance relation to an exact viewpoint obtains. Distinguish asserted structure in G, a graph expression X representing that structure, and reusable formation rules F. Use an adequate mathematical result with its interpretation and subject-side grounds. C.29 applies when a separate choice, transfer or reliance question about that mathematical representation remains. An ADR or dashboard creates no permission, assurance, or work relevance without the corresponding direct claim.
If work uses the description, state the exact premise, reference, decision-use, or operation-argument relation through which the performed work consumes it.
Specification use
An integration team needs to decide whether a service-interface description is fit to drive a conformance test. The exact interface is the EntityOfConcern of PaymentInterfaceDescription; its ClaimGraph states message, ordering, error, and tolerance claims under the effective interface scheme. Name the current integration-test use. Record an exact integration viewpoint only if it changes which interface claims are read or checked or what the team may conclude from the test; otherwise omit viewpoint selection.
The team may call the episteme PaymentInterfaceSpec for this use only after the relevant claims are checkable and the exact conformance harness or validation relation is named. If viewpoint selection affects reliance, the named describing use and its exact selected viewpoint are preserved or explicitly updated.
Formal notation or an approval signature can help interpret or constrain a neighboring claim, but neither substitutes for those conditions. A changed test result changes the result or reliance claim; it does not reidentify the interface or the episteme.
Publication form and carrier
The organized claims expressed by a completed pump-inspection card constitute E when its exact claim content, inspected pump T and effective maintenance scheme R satisfy C.2.1. A particular expression X can follow reusable card-layout rules F and be recoverable from tablet file C. E.24.PUB then tests availability of E through X and C to the declared maintenance audience for its stated use.
The card episteme, layout, file, and availability occurrence retain different identities. Filling, uploading, or opening the card is dated work. Availability establishes neither that a technician read it nor that its claims are true or relied upon.
Episteme about an episteme
A reviewer writes an assessment of one exact DRR edition. The DRR episteme is the EntityOfConcern of the assessment episteme; the assessment has its own ClaimGraph and effective scheme. A PDF of either may be a carrier, a publication occurrence may make an edition available, and an evidence path may support the review assertion. None of those neighbors requires a meta-description kind or context recursion.
Same content, different use
One unchanged equipment-description episteme is first read under a maintenance viewpoint and later under a training viewpoint. Its ClaimGraph, EntityOfConcern, and effective scheme remain fixed, so C.2.1 identifies the same episteme. The two named describing uses select different viewpoints. The second selection neither creates another episteme nor proves conformance to either viewpoint.
If the training use adds another publication occurrence with another form or carrier, or relies on another evidence path, only those neighboring objects and relations change. If the training edition changes a claim or its effective interpretation scheme, C.2.1 instead identifies another episteme; retained wording or a shared file does not preserve identity.
Minimal dashboard repair
A project note says, “The architecture dashboard approves the deployment role.” The operations team needs to know which architecture description it can discuss and whether deployment has been authorized. Suppose the dashboard legend identifies the named description, its claim content, its subject and the scheme in which the team can read it. The team can then recover the following:
PaymentServiceArchitectureDescriptionis the C.2.1 episteme about the exactPaymentServiceholon; read its claims to establish which structure is described and what, if anything, the description asserts about C.30'sArchitectureRelation;- the receiving use is the operations discussion; record the exact operations viewpoint only if it changes what that discussion reads or checks or may conclude, and otherwise omit viewpoint selection;
- the dashboard may be a publication form, carrier, representation, or view only under the recognition rule for that exact use;
- no checkable-claims-plus-harness basis has been named, so specification force is not admitted;
- no gate verdict, permission relation, acting system, system-role assignment, or performed deployment work has been established.
If the operations audience can obtain the selected edition for its discussion from a dashboard expression on the identified display or file, and that same expression follows the dashboard form, state that this publication occurrence makes the edition available for that use. “Approves” and “deployment role” remain non-assertable until their direct governors and case facts are named. The practitioner stops there instead of replacing the original sentence with another overloaded noun.
Consequences
Rationale
The account distinguishes the subject being described from the claim-bearing use of a C.2.1 episteme. Each identity is recovered independently; C.2.1 permits the same episteme to serve both uses in a self-description. Specification is a checkable use of that episteme. Viewpoint selection, view membership, scope, model-use structure, grounding, evidence, assurance, edition, publication, carrier, representation, and work have different reasons to obtain and different identity rules.
Making those neighbors fields of a description tuple would erase those rules and make formality, publication, approval, or a shared context label look constitutive. Requiring all of them for every description would also make ordinary use needlessly heavy. For an identity question the constitution can suffice. For another decision, recover the applicable facts and show how they support it together. Where they concern an edition, establish which edition each fact concerns. Stop when those grounds answer the recipient's question.
SoTA-echoing and source use
Reopen this source-use synthesis when a cited standard changes the practical distinction, or when the current FPF constitution, viewpoint, publication, specification-use, Bridge, or representation interface changes enough that one of the routed decisions above would be stated differently. A newer source matters here only when it changes the working decision.
Relations
Builds on:
- A.7 - Strict Distinction. Supplies the general discipline for keeping an independently governed entity distinct from epistemic and presentation-side objects around it.
- C.2.1 - Episteme Identity, Constitution, Grounding, and Edition. Supplies the exact ClaimGraph, EntityOfConcern, effective ReferenceScheme constitution and the neighboring grounding and edition relations.
- E.10 - Unified Lexical Rules for FPF. Supplies subject-first recovery and the rule that a word, field, position, or representation does not create the governed object.
Coordinates with:
- E.17.0, E.17, and E.24.PUB. Use E.17.0 for a named describing use's viewpoint selection, viewpoint membership, and view membership; use E.17 and E.24.PUB for publication occurrence, publication form, and carrier bearing. None of these changes C.2.1 identity.
- A.2.6 and A.1.1. Govern claim scope and bounded model-use structure only when the receiving use depends on them.
- A.10, B.3, and G.11. Govern evidence provenance, assurance reliance, and currentness for exact objects and relations.
- A.6.3.RT / A.6.3.RT.OE govern expression construction and correspondence; C.29 supplies the mathematical-lens branch. A.6.2, A.6.3, A.6.4, and F.9 govern episteme morphing, source-to-receiving construction, retargeting, and cross-scheme Bridge semantics without label-only sameness.
- A.3.2, F.4, and F.5. Define method-description membership, system-role-kind-description content, and naming after the exact object and local sense are recovered.
- A.15.1 and direct receiving-use patterns. Govern performed work and the exact premise, reference, decision-use, or operation-argument relations through which work uses an episteme.
Repair moves
Use these repairs on live prose; retain old spellings only as quoted source-side trigger wording:
- Start with the exact receiving work, decision, comparison, inquiry, preservation, teaching, or publication use and its next unresolved question or action.
- Replace
DescribedEntity*,EntityOfInterest,EoI,EoIClass, and generic “object under description” wording with the exact EntityOfConcern and its independently governed identity. - Replace local episteme-slot, subject-field, tuple, card, or context-record constitution with the exact C.2.1 ClaimGraph, EntityOfConcern, and effective ReferenceScheme test.
- Replace peer-layer I-D-S wording with EntityOfConcern, description episteme, and admitted specification use; specification is not a third peer kind.
- Replace a positive source-side
DescriptionContextwith the named describing use and the exact viewpoint it selects when that selection changes the reading. Keep selection outside episteme identity, conformance, and view membership; do not recreate the rejected tuple under another name. - Replace “the role contains a characteristic space, state relation, or checklist” with a precise claim: the system-role-kind-description episteme characterizes one exact local system-role kind using claims that cite those independently governed objects or relations.
- Recover the particular expression, selected reusable form, independently identified carrier, bearing relation and publication occurrence that the current use needs.
- Replace
...Specnames lacking checkable claims and a named harness or validation relation with...Description. Preserve or update the selected viewpoint only when the relying describing use depends on it. - Route permission, evidence, assurance, gate, decision, promise, commitment, work, publication, view, Bridge, retargeting, currentness, and representation claims to their exact direct governors.
- Replace “role of this description, source, standard, evidence, or publication” with the exact typed use relation. Use one exact occurrence of a directly declared
U.SystemRoleAssignmentspecies only for an independently admittedU.Systemassigned to one exact local system-role kind; an acting holon is eligible only after that exact entity has independently passedU.Systemadmission for the claim. - Delete mandatory context recursion for descriptions of epistemes; recover the described episteme as EntityOfConcern under C.2.1, including its permitted self-description case.
- Stop when the constitution and any further relations the question requires support the receiving action or expose the particular missing ground.
Conformance checklist
Phrasebook
Didactic memory
Use the short memory use, claims, entity, scheme, grounds for this action:
- Use. What exact work, decision, inquiry, comparison, preservation, teaching, or publication use needs the description?
- Claims. What exact ClaimGraph is being used?
- Entity. What exact independently identified EntityOfConcern are those claims about?
- Scheme. What effective ReferenceScheme makes the claims readable about that entity?
- Grounds. Which further relation would change what the reader can conclude? If several are needed, how do they connect for this use?
- Stop. Use the constitution alone when it suffices. Otherwise give the relations on which the answer depends, or the ground still missing.
E.10.D2:End
First-Practical Entry and Pattern-Use Discoverability Discipline
Type: Pattern-language governance pattern (E) Status: Stable Normativity: Normative for FPF public entry, discoverability, and the publication units that carry them.
Problem frame
Use this when
Use E.11 when a pattern title, README scenario, architectural-reference explanation, ToC cue, retrieval cue, lexical query row, expanded case, or pattern-local recognition passage could change which FPF pattern a working reader should inspect first.
A reader may arrive with a project question rather than a PatternID: for example, how to compare alternatives, revise an architecture or examine a causal claim. E.11 gives that reader a recognizable entry without turning entry material into a second pattern body or universal method sequence.
First useful result. The reader can name the working situation, the first useful result or blocker, one direct pattern or small plausible set to inspect, and the ordinary stop or wrong-turn return. That is enough for ordinary entry; no card form, comparison account, or project-local value is required.
Primary EntityOfConcern. One public entry or discoverability publication unit: pattern title, README first-entry guidance, architectural-reference explanation, ToC query material, retrieval cue, expanded entry-disambiguation case, or a pattern-local Problem frame.
Who publishes and who reads. An FPF author or maintainer publishes or refreshes the public guidance. A practitioner, manager, or assisting agent reads it and opens the direct pattern; that reader is not thereby performing E.11 publication work.
Practical gain. The entry connects a project question to a direct pattern and its first useful result. The direct pattern retains the governing explanation; each navigation unit carries its own entry function.
Not this pattern when. After one direct pattern is selected, use E.11.PUA to follow its Solution to the smallest useful result or a stop that states the missing basis. Use E.11.PUR to judge applicability, recommendation, coordination, or ordering among candidate pattern uses, and to stop on an earlier result that still answers the concern. Use the direct pattern for the actual result, plan, work, evidence, decision, authorization, or publication claim.
Problem
Pattern libraries are difficult to enter from a working situation. A reader may see a long table of contents, search by a familiar word, or choose the first appealing pattern title. That choice can be premature because nearby entries may lead to different first results and different stop conditions.
Attempts to help can create a second problem. Public guidance becomes a numbered method, a shadow pattern body, or a form that asks the reader to fabricate project-local values before the direct pattern has been inspected. The discovery aid then competes with the patterns it should expose.
Forces
Solution - Give Each Entry Publication Unit One Job
Start from the work the practitioner brings to the entry. Write the short route from that question to the first useful result or blocker, the direct pattern or small plausible set, and the stop or return. If the entry helps with one part, say what the practitioner can now do about the original question.
If that prose is truthful and sufficient, stop expanding the entry. Add an expansion, exact result basis, or durable comparison only when ambiguity or a named receiving reliance needs it. The entry returns to the full architectural reference and pattern bodies when their explanation is needed.
Use this distribution:
The framework Readme is the single editable public entry set. If another publication form needs the same guidance, project it from that Readme rather than maintaining a second version. Put any unique cue in the publication unit whose job matches it, then remove the duplicate row or index.
Use E.11.PFP when one public FPF, DPF, or LPF edition needs the shared reader-facing publication form: a compact product-declared opening, separate declared product title and Readme H1, Readme and architectural reference represented in the product's established ToC grammar before one logical pattern index, Readme entry fields, a front-only development-metadata boundary, language boundary, and deterministic source-hazard plus rendered-structure checks.
E.11 still tells the reader how to state the practical question, obtain a first useful result, use the direct pattern, and stop or return. Do not copy the form grammar here or treat a form-valid carrier as a usable framework.
Use E.11.DSG for the full architectural reference of a framework, Suite or FPF Library and its useful question-led entries. The account explains the whole's architecture and reasons; an entry starts with a recognizable situation and returns to the full Methods, available results, evidence or a statement of what is missing.
For a Suite, distinguish constituent subdisciplines from outside suppliers. Embed the framework account in an all-in-one publication; a Suite or Library account can have its own carrier at that assembly level. These are applications of the same explanatory function, not additional frameworks or necessarily independent product series.
A known sufficient contribution remains directly usable; an absent or stale account limits its explanatory route. E.11.PFP governs framework publication form.
Pattern count is only a diagnostic. A one-pattern edition asks whether the result is instead a seed, candidate, or contribution to an existing framework; a larger count still does not establish a pattern language. Use E.4 and E.4.PFAD to decide framework architecture, E.4.DPF.DA or E.2.DA for the applicable package or whole-FPF adequacy, and E.21 for pattern quality.
When the reader's obstacle is finding material under unfamiliar source wording or incomplete access, point to F.1:4.4: its ordinary result is inspectable candidate material and consequential search/access limits. F.1:4.2 justifies a source cut only when the receiving question needs one. An already identified sufficient source or direct pattern can end finding.
When discoverability has become use of one selected pattern, continue with E.11.PUA. When the live question is which applicable pattern use to recommend, how several uses relate, or whether an earlier result already answers the concern, continue with E.11.PUR. Neither continuation turns a public entry order into a universal workflow.
For an FPF-grounded domain or local practice framework, README, architectural reference, ToC, practical entries, an all-in-one carrier, a skill pack, retrieval, or a callable access service may expose the entry. That publication or access use neither decides framework architecture nor supplies authority, and the carrier is not the pattern body merely because a reader reaches it first.
Use E.4 to identify the framework family and member. Only when a downstream-used framework-architecture question is live, record its selected answer in one E.9 DRR using the E.4.PFAD profile; use E.4.PFR separately when a named relation or edition maintenance use needs its representation.
Before a query exists, distinguish an unexamined way of working from a known method that goes unnoticed. B.5.PI connects an ordinary work occasion with inquiry; B.5.EA helps retain and express a sensed but unformulated distinction. A recoverable early cue need not yet become a query. Once there is a question, the entry discipline here helps reach the contribution that could answer it.
A readable, searchable entry can still go unused because it is not encountered when its method would help. Use A.15.11 to arrange that encounter and make the relation to the work apparent. Once the method is under consideration, continue with selection or application; the publication entry remains governed here.
Public first-entry scenario and optional expansion
A public entry may be ordinary prose. It is sufficient when these values remain recoverable:
The semantic keys in E.11:4.5 identify situations, not steps. A reader may inspect any finite plausible set and stop as soon as one direct pattern is worth opening or no remaining entry can change that starting choice.
When an entry describes a reusable composite Method, make its complete problem–move–result explanation obtainable in a normal public pattern or other appropriate developed owner. E.4.CM develops a missing composition and chooses that owner. The card can then be a short presentation of the Method; its status as a publication unit does not reduce the described Method to navigation. An entry that only helps select among Methods can remain navigation.
Keep three reader-facing jobs distinct. A compact locator points to a direct pattern when retrieval is enough and carries no mandatory mantra.
An ordinary practical entry makes the five values above recoverable when one direct pattern, with at most a plainly conditioned next use, can answer the difficulty, provide a first result or blocker, and say when to stop or return.
A Practical-Use Card is a selected Readme example for a recurring complex difficulty whose useful answer normally spans several direct pattern contributions, checks, and returns. Its visible mantra keeps that longer dependency in attention during repeated or interrupted use.
The card is the publication unit. Its practical-use guidance is what it tells the reader. Neither is the shared form, a pattern body, a Method, performed Work, a project result, authority, or a CGUS demonstration.
When a Card or mantra is useful because it keeps material dependencies among several independently reusable recurring problem–move–result contributions in attention, return to the candidate-recognition move in E.4.DPF:4 before treating the Card as only an access front. Its form is a cue for that comparison, not proof of framework scale or product membership. When no live framework question remains, continue choosing or maintaining the entry form under the tests below.
Start with the reader’s question about their own work. The displayed entries illustrate how the pattern language can help. An entry about a familiar activity can still answer a different question. Compare the result its method would supply with what this reader needs to decide, explain or do. Establish whose work the question concerns. Where the answer depends on composition, which parts and which whole does it concern? Where it depends on the model, which differences could change the answer or its grounds? Use C.39.QR:4.1 when a change in the question is hard to see. If the entry supplies just one needed contribution, take that result back to the original question. A general argument within one example can already settle it. Give a direct locator when the reader needs a known section. During open exploration, an example may help the reader form a question for the first time. When both uses matter to discoverability, show at least one ordinary example of cheap direct help and a few cards that demonstrate extended cross-pattern use. Return unmatched questions to the index, a guide, search, or a small plausible set of direct patterns. Apply F.19:4.4 when a catalogue or coverage-boundary reading needs an explicit correction. Do not turn every useful topic, pattern, or reader-entry set into a public example merely to prove breadth.
The following comparison selects a Practical-Use Card for an entry; it does not select which subject reasoning the language preserves. A long mantra and its developed account can explain the language's architecture even when an ordinary entry is sufficient. Use E.11.PFP:4.7.2 to publish that connected explanation and its overview without forcing it into a card.
Before selecting a card, compare the proposed entry with the same truthful content but no mantra. If a cold reader can still recognize the situation, choose the direct pattern and any plainly conditioned next use, recover the first result or blocker, and return after interruption or repetition just as reliably, keep a locator or ordinary entry.
Decide what readers need to recover, with what aids and after what delay. Seeing the answer on the first attempt can teach it for the second. Use comparable unfamiliar tasks, or compare readers whose prior preparation and assignment to conditions allow the mnemonic's effect to be distinguished. Faster initial learning, a correct first choice and successful return after an interruption answer different questions. One reader can show the route that reader took. A claim about repeated use across the intended audience needs grounds covering those readers and tasks.
Select a card only when, without the repeatable formula, the reader is materially more likely to lose a choice-changing question, intermediate result, check, branch, or return, and repeating the mantra restores that reasoning.
Immediate recognition, repeated exposure, fluent recitation, pattern count, heading depth, an existing label, a quota, or a wish to avoid validation is not evidence.
If the mantra crowds out the first result or return, or adds no recall advantage over the same content without it, use the ordinary entry or locator instead. Check declared cards and plausible non-card entries under the same test; the number of cards is an outcome, not a target.
Keep four claims separate. A product selects a card unit for a reader use. The unit publishes practical-use guidance. The product gives every selectable example one stable semantic key and assigns that key one ordinary-entry or card form in one product-wide declaration. The visible card applies the shared form from E.11.PFP.
A key, heading, form, or mantra establishes none of the other claims by itself.
A local mantra may keep one bounded result or one direct pattern contribution in attention. It belongs in the direct pattern, an ordinary entry, or other teaching material when useful; it does not by itself select a Practical-Use Card.
A card mantra keeps the reasoning from the recognizable difficulty to a more distant intended result in attention across several pattern contributions. Include only the intermediate questions, results, checks, branches, and return conditions that change how the reader continues. Phrase length does not decide either use.
The mantra remains Plain, repeatable action or judgement wording. It does not replace a direct pattern's Solution, create Work, or establish a CGUS structure.
An optional same-key expansion explains only the branch choice or result support that the compact card cannot omit truthfully; an ordinary walkthrough remains an explanation, and a demonstrative slice still requires independent A.22.CGUS admission. E.11.PFP defines the required visible field and heading grammar rather than this section.
When an entry must show how one pattern use may lead to another, name the starting cue, direct pattern or plausible set, first result or blocker, each condition that makes another pattern use current, and the stop or return. Do not imply that those references prescribe a workflow. For recurring multi-pattern use that needs mnemonic continuity, use a Plain mantra as above.
Only after A.22.CGUS independently admits the represented conditional structure, name its loci and bindings or potential continuation candidates when they change which continuation is available. An entry, card, mantra, or readable continuation creates neither the selected U.Structure nor its CGUS membership.
Use this internal explicitness ladder only when it helps decide where the explanation belongs; do not persist a score for every entry:
Use the ladder to choose where the explanation belongs. Once an entry gives the reader what its form promises, add detail only when ambiguity or a receiving use calls for it.
Authors may use these schemas when short prose alone cannot truthfully convey a card's result promise, boundary or later reuse. The schemas contain no project values or reader-project instance and are not a public form.
Use demonstrativeSliceRef only when the example independently passes A.22.CGUS admission for its declared illustrative use. Otherwise use an ordinary walkthrough; no rationale record is required merely to say that an explanation is not a CGUS slice.
Cold-reader recognition and grounded public value
Read the public entry as someone in its intended practice who has the stated prerequisites and has never used this route. For an entry to an engineering decision this might be a practising engineer or an assisting agent. The heading and opening sentence should let the recipient recognize their own question. The next sentence should identify an attainable result or blocker and the distinction that changes the next action. Introduce PatternIDs, FPF kind names and assurance fields where the reader needs them.
A public value claim is grounded when the reader can recover the project need, first useful result or blocker, why one direct pattern can help, and the ordinary boundary. Add the exact potential-result kind, identity or obtaining basis, result-relative object, or conditional receiver only when omitting it would change the truth, the starting choice, the stop, or a named later reliance. The entry may stay readable prose; the reader never has to fill a card before opening the direct pattern.
Keep the public set representative of FPF's range. Wording and description repair remain visible but do not dominate architecture, problem shaping, work, comparison, evidence, timing, causal use, mathematics, quality, improvement, framework authoring, system recognition, or system delimitation.
Recover the direct object before a PatternID is known
Some readers arrive before a practical-use key is recognizable: a familiar relation, project, process, case, context, or problem phrase is already blocking the work, but its direct object is not yet clear. Give such readers an ordinary-language recovery entry before asking them to compare PatternIDs. Choose among these independent entries as alternatives. They impose no sequence, require no form and add no card set.
Keep four moments distinct. Recognize why the ordinary situation matches this entry. Select the direct pattern whose Solution tells the practitioner how to obtain the expected first object. Use only the branch needed now. Return the smallest usable result or a named blocker. A direct result exists, or a relation obtains, only when the applicable pattern's conditions are satisfied; the entry cue creates neither.
Keep the entry short enough to consult in the work. Give the situation, the distinction the practitioner needs, what the direct pattern can yield and where to go on a wrong turn. A first result can complete a lookup. If it supplies a premise for a wider explanation or design, show how to use that premise and where the remaining work belongs.
-
An obtaining relation must be referred to, and perhaps distinguished from a repeated episode. First name the exact participants and the readable direct relation, then open the pattern whose content defines or tests that relation. A current assertion can stop there when later work only needs to know whether the relation obtains.
If history, comparison, another relation, or a declared operation application must distinguish this occurrence from another of the same kind, use
A.6.RELand the relevant direct pattern's same-versus-new-occurrence rule before naming or referencing it.The smallest result is the readable direct assertion or, only when consumed, one recoverably individuated occurrence; a missing participant, predicate, current fact, identity rule, or relation rule blocks that result. Stop as soon as the named receiving use can use that result.
If no current direct relation can state the needed claim after exact recovery, require
A.6.RCD; a row, edge, identifier, report, or mention makes no occurrence obtain. -
Project, process, or case wording no longer reveals the subject of the decision. Open
A.15.6and recover the subject before using the management label. An actual project is qualifying compositeU.Work; a process concern may be a reusableU.Method, a structure selected underA.22, or aTransformationFlowStructure; a case follows one affected referent or claim through the change history needed for closure and keeps the relevant downstream use outside that closure.The smallest result names that subject, the pattern used to identify or constrain it, and the claim the current decision may make—or the missing identity, relation, closure basis, or information.
When a source says target system, recover whether it means the product System of an extended enterprise under A.1.STM:4.1. Use system-of-interest for that role in the resulting English account; the phrase alone establishes neither identity nor the role.
Keep plans, decisions, work-to-system relations, system-role classifications, assignments, participation, responsibility, and authority separate. If role still hides the claim, use
E.10.ROLE. Stop when the recovered subject and claim answer the decision; useA.1.SCRonly if systemhood still changes the answer. -
A claimed bounded context may be only a label, boundary picture, team, or subsystem. Open
A.1.1with the engineering decision, one exact model edition, and its exact use locus. Recover the smallest direct applicability, assigned-Work use, or fixed-content coherence relation first and stop there when it answers the decision.Select a
BoundedModelUseStructureunderA.22only when the joint organization itself changes the decision and all four discriminators are exact: independently identified constituents, selected obtaining relation occurrences, applied constraints, and one named selection-use frame. The smallest result is therefore one direct relation or that optional selected structure; a missing constituent, occurrence, constraint, or use frame blocks selection of that structure.Context Mappingremains aU.Method; any cross-context structure needs its own A.22 selection; and a scheme, scope, viewpoint, conforming view, representation, or diagram remains a different object. Stop at the direct relation or selected organization.Use
E.17.0only when the actual question is whether an exact episteme conforms to a viewpoint and is thereby a view. A bounded-context phrase creates no holon, subsystem, team, structure, relation, viewpoint, view, or representation. -
Problem-side material may describe a concern without identifying an actual Problem. Open
C.22.PFRonly when the claim may concern one obtainingProblematicForRelation: an exact actual-condition occurrence and exact problem-criterion-applicability occurrence whose selected input is actually adverse. Keep that occurrence distinct from the predicate, applicability occurrence, assessment or evaluation, assertion and reliance,ProblemCard, forecast or modal concern, and current-solvability or continuation claim.The smallest result is an ordinary actual-problem sentence naming condition and value, criterion, entity and use, and applicability window, or an honest non-PFR classification or blocker when the condition, applicability, adverse input, or required PFR rule is missing. Stop as soon as the later use can distinguish actuality from problem-side claim material.
Use
C.22.2when the useful object is a reviewable problem-side card or formulation rather than the world-side relation. OneProblemCardmay describe no actual PFR; selecting or discovering a method changes only the current solvability or continuation claim, not PFR participants, obtaining, identity, or the adverse condition. -
The present work succeeds, but another useful continuation may be possible. The framework Readme's
WORK-OPPORTUNITYentry connects bounded exploration, needed method contributions and the worth of advice. Begin with the available material or action; a changed constraint is not required. SelectC.40,C.39orC.11.DUAby the first missing result, then stop at that result or its exact blocker. Continuing the present work without a new recommendation is a normal outcome.
Public helper epistemes
These helper epistemes are optional authoring or named-reliance support. Do not open them when the short public entry and direct pattern already make the result and boundary truthful. A pattern reference locates the FPF pattern episteme whose content is needed; classify it as a U.MethodDescription only when the A.3.2 criterion passes and the current use depends on that classification.
An expanded public template asserts no project result and contains no project value. It names only the exact positions needed to keep its promise or blocker truthful: the potential result and how it would be identified, the direct pattern whose content defines or constrains it, and any identity, obtaining, relative-basis, continuation, or receiving-use distinction that changes the branch.
Method, plan, dated Work, transformation, evaluation, decision, and receiving-use identities remain absent unless the current promise or later reliance depends on them.
The result-relative basis template has exactly one category. A direct-relation template asks for predicate, participants, applicability, obtaining, occurrence identity, and direct governor.
An A.6.1 template asks for operation, application, argument or result binding, and direct governor.
An A.6.RCD local-claim template asks for one C.2.1 claim episteme with polarity, substrate or constructor, base predicates and their direct patterns, participants, case facts, and any support or warrant required by the later receiving use. The base patterns define the predicates. Use case facts or constituting history to support the local claim at its stated polarity. Stating the claim creates no separate relation occurrence.
Result identity or currentness and result-relative basis are different public questions; they coincide only when the potential result is the same direct relation occurrence used to close the later application.
conditionalNextQuestionPatternRef is present only when the public branch itself promises a continuation or names a downstream reliance. A result template without such a continuation leaves it absent. A public stop, missingGovernor, or missingInformation boundary has no receiver. return, wrongTurnRecovery, and strongerNeighbor name a receiver only when that continuation is part of the branch.
The optional obstacle names a recognizable obstacle only when one matters. Practical use may begin from an object to inspect, a result to evaluate, or an existing Method to improve without first inventing a Problem.
Candidate-use templates and basis completeness
This section applies only when an optional exact expansion has been opened because the short public entry cannot carry a truthful promise, blocker, or named reliance on its own.
PatternSolutionSectionRef is an edition-pinned reference to the cited pattern's Solution. A broad result family or pattern title is insufficient.
Exactly one of expectedResultTemplateRef and resultPromiseBlockerRef is present in an expanded candidate branch. A result promise is admissible only when its potential-result kind, identification question, direct pattern, identity-or-obtaining basis, relative object and category-correct basis, minimum usable result, and any continuation that applies to this branch are stateable.
A blocker states the missing rule or information and carries no fulfilled result template. Optional omissions cannot masquerade as a weak passing promise.
The completeness condition inherits C.2.1 constitution. Its EntityOfConcern is the reusable candidate-basis position declared by the template; its ClaimGraph states the admitted filler kind and positive completeness condition; its ReferenceScheme explains how later current project fillers satisfy that position. It contains no project value.
Ordinary walkthrough
An ordinary walkthrough may remain readable prose. Use the following optional helper only when exact result, boundary, or continuation references are needed to keep that explanation truthful:
Where an exact row is used, it carries one result template or public blocker. It may contain a short repeatable formulation of the direct pattern's Solution. Call it a CGUS demonstrative slice only when A.22.CGUS independently admits the represented conditional structure; an ordinary walkthrough needs no record explaining why it is not such a slice.
Practical-use carry-through check
Read each published entry first in the form the public will see. A passing ordinary entry exposes the recognizable situation, practical question, first useful result or blocker, one direct pattern or small plausible set, and the stop or wrong-turn return. Read the question the prospective user brings beside the outcome of following the entry. An entry can provide a complete answer, a useful part with a way back to the larger task, or a reason to look elsewhere. Check whether the reader can tell which applies when the topic still sounds familiar but the requested work has changed. This check creates no project instance, applicability verdict, result entity, relation occurrence, receiving use, or separate positive record.
For a selected card, also compare the same truthful entry without its mantra. The card passes only when repeating the mantra materially improves repeated or extended use under the test in E.11:4.1.
On a first reading, ask the reader to put the formula in their own words, name the first useful result or blocker, follow Start with to the direct pattern and any conditioned next use, and recover the optional expansion by its key. Check later return under the conditions specified for the card. The direct pattern remains authoritative.
A short slogan that loses a choice-changing distinction fails, and a form-valid card that provides no mnemonic gain returns to ordinary-entry or locator form.
When an entry needs the optional exact expansion because a promise, ambiguity, or named reliance cannot otherwise remain truthful, use this conceptual view over the already published values:
The view is not a form to complete or a durable check object. Inspect only positions the expansion uses.
Use an ordinary walkthrough or an admitted demonstrative slice when it explains a distinction the branch needs. Add another case only when its changed condition reveals a consequential choice or limit that the first case leaves unclear. The demonstrative form must satisfy A.22.CGUS; the ordinary form needs no non-admission rationale.
State a principal blocked overread only when the public wording otherwise invites a consequential false project claim.
For each expanded candidate-use template, exactly one result promise or exact public blocker is present. A promise identifies the direct pattern and Solution, potential-result kind, local identification question, the identity or obtaining basis and result-relative basis that make the promise true, the minimum usable result, and a receiver only when that continuation is current.
A blocker states the missing rule or information and carries no fulfilled result template. A broad family, generic result relation, omitted value disguised as a weak promise, fabricated project occurrence, or PatternID list without selection conditions does not pass.
Stable practical-use keys and selected forms
Give every selectable public entry one stable semantic key so the reader can return to the same situation after wording, grouping, or presentation changes. The product maintains one declaration that assigns each key exactly one ordinary-entry or card form. An optional expansion repeats its enclosing card key only to remain findable; it is not another selectable occurrence. This rule creates no universal entry kind or second key registry.
E.4.FPF carries the current FPF example-key and form declaration plus its language-appropriate compactness measure and two maxima. A DPF or LPF carries its own product declaration under E.4.DPF. E.11 therefore does not maintain another FPF key list or treat displayed examples as product coverage.
E.11 records one F.13-form historical read path: splits(SYSTEM-IN-CONTEXT -> {SYSTEM-RECOGNITION, SYSTEM-DELIMITATION, WORDING, ARCHITECTURE}).
The old card had no single surviving public-guidance identity: system recognition, system delimitation, lexical recovery, and architecture have different referents, relations or evaluations, receiving uses, first results, and direct governors. Older writing remains readable through this one read path; current entry use names only the four resulting keys.
A.1.STM is a conditional continuation with a dedicated readable README guide, not a fifth resulting key. The split creates no U-kind, relation kind, record kind, result kind, or generic Context claim.
The FPF Readme carries a selected set of current public examples and their optional expansions. The First Principles Framework Architectural Reference explains why FPF's distinctions and Methods work together. ToC locates pattern families and questions outside the examples. Full patterns carry Methods, conditions, costs, consequences, and result semantics. None is a second entry store or a claim that the examples bound FPF use.
Architectural Reference, local recognition, and first-entry terminology
The First Principles Framework Architectural Reference explains the language as a whole, including the connections and grounds behind the README entries. Use E.11.DSG and E.11.PFP:4.7 for that complete account. It uses plain engineering language before FPF vocabulary and explains the cross-cutting ideas once rather than copying the scenario set.
Its coverage includes transdisciplinary use without collapsing local meaning; local closure in an open world; holons, systems, epistemes, and architecture as structure; EntityOfConcern and description/publication/view separation; thinking-through-writing; epiplexity; first-principles-to-work; mathematical lenses and FormalSubstrate distinctions; ontology-first wording repair; evidence/assurance/gate/decision/work separation; characteristic spaces, quality, NQD/OEE, and improvement; novelty, diversity, and SoTA; and didactic primacy.
A strict FPF term that carries the explanation receives an immediate plain gloss. Pattern IDs are addresses for stricter treatment, not the main explanatory language.
A pattern's own Problem frame is the local high-precision recognition section. It makes recoverable the primary EntityOfConcern, working problem, failure if missed, first admissible action, practical result, and ordinary non-use boundary. Add candidate-pattern comparison only when a real discoverability ambiguity exists; otherwise keep cross-pattern comparison in README, ToC, Relations, or an expanded case.
Keep these terms stable:
ToC and lexical-query phrases remain finding aids, not alternate names, semantic equivalences, or authority relations. A projection that needs to answer a substantive claim must return to the direct pattern or the pattern for that claim; do not strengthen the projection.
Bounded comparison
When more than one selectable entry remains plausible, compare four things: recognizable-situation fit, difference among first results or exact public blockers, direct pattern, and stop or return condition. Keep the comparison in conversation for ordinary bounded use. Open the most promising direct pattern before constructing a project candidate.
In the optional forms below, claimGraph carries the episteme's complete asserted content, including the reasons and qualifications that change what it says. The scheme designation resolves under C.2.1:4.2.6 to the rules actually governing that account. It can select another episteme's effective scheme or exact content defining those rules; a reference to a document alone does not choose between them.
Preserve an unresolved interpretation rather than filling the field with a nearby name. These forms organize addressable writing when needed; ordinary conversation and partial notes remain sufficient under the entry conditions above.
Materialize the rationale only when a named later use needs addressable comparison history; then it has one public-guidance subject and no fabricated project result:
Stop inspection when one entry has enough recognition and first-result advantage to justify direct pattern inspection, when no remaining entry can change the starting choice, or when the inspection budget opens an explicit return.
Materialize comparison history only when a named receiving use relies on it:
Guidance, practical question, compared result templates or blockers, first-result differences, named reliance, stop, and return remain ClaimGraph content or separate references under their direct patterns; none replaces the C.2.1 identity.
Each comparison cites at least one result template or exact blocker from the guidance it evaluates. claimScopeRef or modelUseStructureRef is present only when the named scope or model-use structure changes the reliance being recorded.
Several plausible entries alone do not make this record current. The named reliance may be a later review, replay, audit, automation, or another use that needs addressable comparison history. Retain only the rows that use needs.
Compare the information the reader actually receives
For a title or cue revision, name the reading operation and compare the material that operation exposes. First discovery, browsing supported working questions, selecting among neighbors and re-finding a known method can need different cues. Understanding an opened body and composing its method with others require the content and relations inside the patterns.
Keep these comparisons distinct:
For a title-only comparison, withhold the fuller cue and the author's intended answer. Keep the task and allowed preparation comparable, include tempting neighboring patterns and an ordinary non-use case, then open the selected body to inspect the promised result and boundary.
Preserve an actual first response before giving explanations that could teach the intended choice.
A desk comparison can justify a visible repair; use an actual reading when unresolved interpretations can change the selection or when claiming that a reader recovered the intended question.
For AI retrieval, inspect the real input: the file or index fields, included heading and parent context, chunk boundary and returned passage. A renamed heading can affect only a path that carries its text or resulting representation.
Adding context to a fragment is another intervention. A heading scan, a full-text search and a contextual embedding query therefore need their own declared conditions; a result from one does not establish the others.
Check both discovery from a working question and re-finding by a useful technical term. An attractive first click is insufficient if the body returns the reader from a wrong use or the familiar method becomes harder to find. Keep the useful shortlist, correction or no-use return; ordinary comparison needs no extra catalogue or score.
A local AI reading supports that observed response, not a human-effect estimate. Keep any new body content separate from a title or presentation contribution, and use an appropriate description-comparison method when that attribution matters.
Resolve a first-pattern ambiguity with a worked comparison
When a short entry and the candidate's Problem frame still leave two different starting points plausible, ask which fact would change the first result you need. A shared topic word is weak discrimination: “contract” may concern recognition or atomic claims; “improvement” may concern a measured rate or causal support. Recover the current object, proposed use and available facts before choosing by that word.
Use the worked comparisons in E.11:5.7–5.14 only when their working question resembles yours. Each shows a discriminating fact, the direct Solution worth opening, and an ordinary stop or return. They are examples, not a complete pattern catalogue or a reading sequence. A sufficient compact cue can bypass them.
Compare the plausible entries against the fact: what would each help you obtain first, and what would make that result premature or irrelevant? Obtain the fact from the task material or the inspected source. If it is unavailable, name the missing fact and the choice it prevents; an attractive example supplies no missing project evidence.
When the comparison points to a promising direct pattern, open its Solution and check the inputs and boundary stated in that pattern. If that body changes your understanding of the question, revise only the affected comparison and keep earlier useful reading.
Stop entry comparison when the direct body gives a justified starting point, no remaining plausible entry can change the choice, or a concrete missing basis prevents it. Continue with E.11.PUA to use one selected pattern. If inspected candidate uses still require applicability, recommendation, coordination or ordering, use E.11.PUR; presenting an example first has settled none of those judgements.
For publication, expand an ambiguity only when compact guidance cannot resolve it truthfully. A serious consequence of a wrong choice, repeated misclassification, difficult retrieval or a materially new comparison can justify a worked case. Show the choice-changing fact and its consequence in ordinary language, retain the direct-pattern boundary, and make the relevant example findable from the working question. Readers use that explanation without becoming its authors or completing a comparison form.
Replay and currentness
Replay one public entry first from its recognizable situation, practical question, first useful result or blocker, direct pattern or plausible set, boundary, and readable walkthrough. Consult the exact helper fields only when the entry uses them for truth, disambiguation, or named reliance. The guidance remains current only while its situation and question still point to the same use.
Recheck the smallest affected entry slice when its recognizable situation, question, first result or blocker, direct Solution, selection condition, stop, return, or true consumer changes, or when use evidence shows a recurrent wrong turn. Recheck exact result and basis fields only when the changed entry uses them. Use G.11 for edition, telemetry, currentness-window, and decay orchestration; E.11 supplies the entry-specific values and change conditions that orchestration inspects.
Archetypal Grounding
Architecture or working document?
A team says, "Our diagram no longer explains the system." ARCHITECTURE and WORKING-DOCUMENTS both look plausible. The first card can return an architecture question, candidate set, or selected-structure result. The second can return the smallest description-use, representation, publication, or other working-document result for a named reader use.
The team compares those first results and sees that the selected structure itself is unsettled. It starts with ARCHITECTURE. No comparison account is needed because the comparison is local and reversible.
A later safety review needs comparison history
The receiving safety review relies on an addressable rationale for why a team compared the ordinary MATHEMATICAL-MODELING entry, OPTION-COMPARISON, and SYSTEM-DELIMITATION with the ordinary SYSTEM-RECOGNITION entry before a hazardous test. New measurements will later reopen the choice. The four examples ask different first questions: what a model can support, which comparison or readiness result is needed, which parts or crossings matter, and whether the exact entity must be treated as a System at all.
That named reliance admits a PracticalUseEntryComparisonAccount@Context with four comparison rows, the stop boundary, and the return condition. The account does not authorize the test or replace evidence, assurance, gate, choice, or WorkPlan relations.
A card leads to a physical result without promising it
WORKING-DOCUMENTS can lead to a usable machining work instruction. Its public guidance first asks what a named reader must decide, do, check, or rely on and returns the smallest truthful document-side result or blocker. Use the direct pattern to determine whether the useful result is a MethodDescription, WorkPlan, claim, permission, publication, or another document-side result and what relation makes it useful for the later project work.
The card does not promise a machined component. When actual machining later becomes current, use A.15.1 to identify the exact performed Work occurrence; use A.15 as well only when the broader System-Role–Method–Work Alignment question is current. The instruction or plan is neither that dated U.Work nor proof that it occurred.
Repair the smallest card slice after a direct result changes
Suppose a new A.6.3.RT edition restores a progressive representation result: an ordinary target representation plus source-comparison note first, exact endpoints only when a named receiver makes their identity material, and a historical transition occurrence only when actual transition Work and all required participants are current.
Repair only the affected WORKING-DOCUMENTS branch so its first result and escalation triggers match the direct pattern. The card heading and general question remain unchanged when readers still recognize the same situation.
If the repair changes the cross-pattern path that the mantra must help the reader recover, or its first result or return, compare the same truthful entry with and without the mantra after the relevant delay or interruption. Use equally informative task inputs and allow for what readers learned in any earlier attempt. Reuse a still-applicable comparison for unchanged mnemonic content and use. Keep the card only when the comparison supports a material mnemonic gain. If the comparison supports equivalence for the intended use, use an ordinary entry or locator; if the mantra hides the result or return, repair or drop it.
A better first-click rate can make discovery worse
Suppose retrieval ranks WORKING-DOCUMENTS first whenever a diagram is mentioned and the first-click rate rises. Follow-up comparison shows that more readers now open document-use patterns while the architecture subject or selected structure is still unsettled, so first-result mismatch and wrong-turn returns also rise.
The visible navigation measure improved while the intended value worsened. Keep first-click rate as telemetry, apply E.13 to the substitution, and judge the guidance by recoverable situation fit, first-result fit, and wrong-turn cost rather than by the click measure alone.
Discharge a duplicate first-entry row by function
Suppose a compact row combines architecture and diagrams, evidence, dashboard use, and alternative comparison. Do not keep it as another public example merely to display topic coverage. Put architecture design or review in ARCHITECTURE; put description, view, dashboard, and rendering use in WORKING-DOCUMENTS and the direct E.17/C.30.AD patterns; put costly evidence or commitment questions in OPTION-COMPARISON and their direct A.10/B.3/A.21 patterns; put useful search phrases in the ToC or retrieval.
Use the worked comparisons in E.11:5.7–5.14 only if those cues still leave the first choice unresolved. Delete the duplicate after every useful function has a matching home.
Responsibility, method, plan and actual work are being mixed
A team says, “We keep mixing responsibility, method, plan and what actually happened.” First ask whether the difficulty is an alignment distinction or only a vocabulary mismatch. If someone is treating a plan as a completed run, agreeing on words alone will not resolve it.
Open the Solution of A.15 for the System–Role–Method–Work split. A.15.2 becomes relevant when the missing subject is a plan for future work; A.15.3 when a value is being planned for an existing declaration member. Inspect A.1.1 if the model-use boundary affects those claims, and B.5.1 if the question concerns coordination of Explore, Shape, Evidence and Operate states. These are conditioned neighboring questions, not a first-use chain.
If the alignment is already understood and only shared terms are missing, F.11 can distinguish method and work vocabulary, F.9 can resolve a cross-context sense Bridge, or F.17 can stabilize a first term sheet. Stop when the direct alignment body is opened or enough shared vocabulary lets the work proceed. Return to alignment if the term sheet still leaves a plan, role or run confused. Neither F.17 nor a rationale record under E.9 is a universal first stop.
A meaningful phrase is not yet a claim
A reader encounters a phrase that matters but cannot yet say what it asserts. The choice-changing fact is whether it remains a cue to preserve or already has a stable claim structure. Treating the phrase as an endpoint claim would lose the uncertainty the next operation needs.
While it remains a cue, open A.16 to identify the next language-state move and A.16.1 for early preservation. Open A.16.2 when an existing publication overstates closure, stability or framing and must be reopened, backed off, respecified or retired. When a stabilized cue needs an open explanatory question, B.5.2.0 supplies the prompt form with scope and motivating grounds. Once alternatives are stable and their route plurality or selection must be published, B.4.1 supplies that different result. A U.Episteme publication requiring explicit language-state facets makes C.2.LS current.
If the material already contains an atomic boundary claim, open A.6.B or A.6.C for that structure. If it is being forced prematurely into a relational claim, quality endpoint or action invitation, defer A.6.P, C.16.Q, A.6.A or C.25 until the corresponding subject and basis exist. Stop at a preserved cue, an opened admissible next language-state move, or a reclassified boundary-claim question. A cue is not yet a finished requirement, work record, quality claim or invitation to act; reopen when the material actually acquires that structure.
An API or contract sentence mixes several kinds of claim
A reader says, “The API or contract-language description says X.” First distinguish recognizing the encountered description from unpacking what its sentence claims. A description can mix an admissibility gate, duty, evidence claim, action invitation and access note; its familiar label does not settle any of them.
If the unresolved question is “what description is this?”, open A.6.RSIG. When recognition is sufficient and the sentence mixes boundary claims, open the Solution of A.6.B or A.6.C to recover atomic L/A/D/E-classified claims under A.6. Use A.6.P if relation wording hides participants or predicate, C.16.Q if quality wording hides its characteristic and scale, and A.6.A if an action invitation is the actual subject.
An already accepted engineering account may instead need reader-facing publication under E.17; E.17.0 answers view recognition, and E.24.PUB answers publication occurrence, form or carrier. Stop when the direct boundary-claim body is opened or the Claim Register or atomic claim set is ready for the next governing use. Return to description recognition if its defining episteme is still unclear. One API sentence supplies neither a complete Contract Bundle nor proof of downstream effect, performed action, quality or evidence.
A shortlist, a pool policy or one winner?
A team says, “We need a shortlist, not one winner.” Ask which first result is missing: an admissible comparison, a pool of candidates, policy over that pool, a selected-set declaration, publication of the result, or one local decision. The fact that the current use concerns a governed set changes the choice; a one-winner doctrine is premature.
If comparison or aggregation is not yet admissible, inspect G.0, A.19.CN and the relevant A.17–A.19 bodies; A.19.CPM or A.19.SelectorMechanism becomes current when the comparator or selector structure itself is unsettled.
For generation and archive/front stewardship while forming the pool, open C.18. For policy over still-live candidates, open C.19.
Use G.5-6 DeclareSetResult when already identified members and a current choice, pool treatment or other inclusion basis need a selector-facing set declaration. G.5-3 Select applies when selection among its admitted Method or generator candidates is itself needed. E.24.PUB separately makes a declared result available.
If the actual question is one local choice, open C.11. If the action or option is already fixed and the missing result is a tool or service call plan or checkpoint return, use C.24, including for a call that performs an already chosen comparison. Stop at the correct direct body or a sufficiently discriminated candidate set; neither a shortlist nor the order of these examples prescribes a sequence. If several inspected pattern uses still need a recommendation or coordination, take that question to E.11.PUR.
A reusable generator or SoTA scaffold is needed
A reader needs a reusable search, harvest or portfolio scaffold rather than one recommendation. The discriminating fact is reuse across searches, portfolios or updates. A compact cue is usually enough; an expanded comparison is useful only if the reader repeatedly mistakes scaffold construction for a local selector, comparison or publication.
Open G.1 to author a reusable CG-Frame generator and selector kit, or G.2 to harvest and synthesize SoTA for a CG-Frame. Inspect G.0 when the comparison frame is missing. Use A.0 when the missing result is the starter vocabulary for generative search. B.5.2.1 and C.17–C.19 are relevant when creative search, novelty or explore/exploit policy already changes that construction.
G.5 answers a selected-set declaration and E.24.PUB its availability; neither supplies the missing reusable scaffold. G.10 or G.11 becomes current when shipping or refresh is the actual next question. Stop when the scaffold/generator body is opened, or when the question has been correctly reclassified as a set-result or publication use. Reopen if the supposedly reusable account works only for the single current recommendation. A generator or SoTA concern creates no automatic tool-implementation or immediate-publication obligation.
Explain or rewrite the same subject for another audience
A team needs to explain the same EntityOfConcern to another audience. First check whether the subject remains the same and only wording, representation or reading posture changes. Creating a second subject or rule track would change the work rather than explain it.
Open A.6.3.CR for same-entity retextualization, A.6.3.RT for a representation transition, E.17.EFP for explanation-facing rendering, or E.17.ID.CR for bounded comparative reading. Distinguish the result you need before choosing: new wording, a target representation, an explanation or a comparison. A pressured local wording repair can make E.17.AUD.LHR current; stability of a PublicationUnit can make E.17.AUD.OOTD current.
Stop when the same-entity rewrite body is opened or an explanation-facing rendering is stabilized with usable source pins. If comparison exposes a real change in the subject or claim, return to its direct semantic question instead of hiding it as presentation.
A faster rate is predicted after a capacity change
A team says, “Adding review capacity for two sprints will double backlog reduction rate.” Inspect what the claim will be used to support. Ordinary prose, a snapshot, a measured rate and an intervention-sensitive temporal claim need different first results. Speed or rhythm wording alone does not select C.27.
Keep a snapshot as Dyn0 and a measured trend or rate as Dyn1. Open C.16 when measurement construction or comparability is missing. Use C.27 when the temporal claim changes admissible action because effort, window, resistance or cost, basis and reopening conditions matter. Its local Dyn2TemporalClaimAdequacyCard, or a boundary-crossing Dyn2TemporalClaimProfile, states that temporal-use question; it is not benchmark or causal proof.
Return a different question to its direct owner: A.3.3 for a reusable transition law or formal dynamics model; B.1.4 or B.1.6 when contextual/temporal or work-resource aggregation is the missing subject; G.9 for benchmark parity; C.25 for quality, first asking whether one Characteristic and Scale suffice and using a bundle only when differently typed contributors jointly determine the claim; C.26.3 for viability diagnosis and selected boundary regulation.
C.24 addresses call planning for an already fixed action. A relied-on causal effect belongs to C.28, not to the fact that the rate changed.
Use C.26 only if an exact contextual-model obstruction still changes inference or action after the applicable ordinary subject assertion and predicate are in place. Stop at ordinary prose, Dyn0, Dyn1 with the needed measurement body, the relevant Dyn2 account, or a correctly named neighbor use. Reopen when the effort, window, resistance/cost, basis or intended action changes. Faster is not automatically better. A velocity target alone establishes no improvement or service promise; a dynamic benchmark needs a comparison basis before it supports superiority.
A decision relies on a causal or counterfactual claim
A reader encounters “this policy would have prevented harm”, “the intervention caused the improvement”, “this fairness result is causal”, or “these simulated counterfactuals prove the decision”.
Ask whether publication, choice, deployment, assurance, audit, benchmarking or method dispatch relies on causal support. Causal-looking wording alone is insufficient; the relied-on claim is the discriminating fact.
Open C.28 when the required support concerns an effect, intervention success, counterfactual comparison, causal fairness, policy optimality, causal evidence, off-policy/causal-RL evaluation or causal method superiority. Its Solution recovers the causal-use question, causality-ladder rung, claim kind, estimand, identification, counterfactual sampling realizability and support components, then distinguishes supported from unsupported use.
A cheap downgrade sentence can be enough; a CausalUseTriageRecord, CausalUseEvidenceDesignRecord or CausalUseSupportResult is used only at the scope needed by the receiving claim.
If only a measured value is at issue, use C.16; if rate, trend or temporal adequacy is at issue, use C.27. If the question is method, work or work-plan structure, use A.15 and A.3.2; a sampling procedure, intervention recipe or target-trial phrase does not supply causal support by itself. Split a boundary duty or agreement with A.6. Use C.26 for residual QL modeling only when an exact contextual-model obstruction still changes inference or action after the ordinary subject assertion and predicate are in place.
A.10 governs the evidence/provenance paths relied on. B.3, D.5, G.5 and G.9 use a C.28 support result when the downstream question is respectively assurance, ethical/fairness audit, method dispatch or benchmark parity.
Keep C.11, C.19 or C.24 as the direct local-choice, live-pool-policy or fixed-action call-planning use when that is the actual work; the support result does not decide it. Stop at the sufficient downgrade, scoped causal record/support result, or named neighboring use that cites its limits.
Return to causal support if the estimand, causal rung, identification, realizability or relied-on use changes. A randomized procedure is not automatically counterfactual support; simulation is not realized counterfactual data; a fairness metric is not causal fairness; and methods on different causal rungs or estimands are not made comparable by one benchmark. Causal support itself creates no duty, promise, commitment, release gate or admissibility predicate.
Bias-Annotation
- Title-match bias. A familiar word selects a pattern before its Problem and first result are inspected. Compare situations and result differences, then open the direct pattern. When the needed fact is unavailable, return that gap rather than choosing from an example.
- Public-instance bias. A README example is filled with project values. Keep public templates context-free; project candidates belong to
E.11.PUA. - Numbered-entry bias. Entry order is read as Method order. Use semantic keys and condition-specific continuations.
- Record-first bias. Comparison emits a comparison account by default. Materialize one only for a named receiving reliance.
- Card-as-authority bias. A public card is treated as an applicability verdict, recommendation, decision, or authorization. Use
E.11.PURor the direct pattern whose content defines, constrains, or tests that claim.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits. FPF gains a human-readable entry from working questions to direct patterns without losing the result and boundary support that matters. A few ordinary examples show that one pattern may answer a bounded difficulty; selected cards show that the language can sustain longer reasoning across pattern contributions.
Readers can stop cheaply, inspect a small plausible set, and recover from wrong turns, while the unmatched-question return keeps discovery available beyond the displayed situations. README, architectural reference, ToC, local recognition, expanded cases, and retrieval keep distinct jobs, so useful entry value does not become a duplicate canon.
Costs. Maintainers must keep each public entry aligned with the direct pattern's current result and boundary, and must repair its true consumers when those meanings change. A recurrent ambiguity or named later reliance may justify an exact expansion, worked case, or addressable comparison history; those deeper objects then need currentness care. Ordinary entries pay none of that record burden when readable prose is sufficient.
Rationale
Discovery is a bounded decision under limited attention, not a one-time lookup. A recognizable situation and first-result difference let the reader choose what to inspect before learning the corpus topology. A recoverable return is more useful than pretending the first cue is always right.
Public guidance remains weaker than the direct pattern. It helps a reader choose what to inspect; it does not decide applicability, authorize work, identify a project result, or make a relation obtain. Precision is progressive: keep the public explanation simple while it remains truthful, and open exact result identity, basis, continuation, or reliance fields only when one of those distinctions changes the choice or claim.
SoTA-Echoing
Navigation question. How can a reader choose a useful first pattern without inspecting the whole corpus, and recover when the first cue was misleading? The best-known line for this use is sequential, bounded inspection with an explicit return: compare the result or blocker offered by a few plausible entries, open one direct pattern, and stop or backtrack when its boundary rules it out.
The alternative is familiar-title or ranked-result lookup with a short topical snippet. It is cheap and remains sufficient when the direct pattern is already known. For an ambiguous question, however, the same small reading budget can be spent on a situation, first-result difference, and return instead of another topical label. This choice trades some snippet brevity for a recoverable wrong-turn decision; it does not promise fewer clicks or faster task completion.
Adapt: E.11:4.1, 4.1.2, 4.4, 4.6, and 4.7 keep that bounded inspection and cheap exit; case 5.5 rejects a better first-click metric when the reader reaches the wrong use. Section 4.6.2 and cases 5.7–5.14 unfold the discriminating fact, direct result and return when a compact cue still leaves an ambiguity; they retain the cheaper sufficient-cue exit.
Jin, Bai, and Oulasvirta, Modeling Trial-and-Error Navigation With a Sequential Decision Model of Information Scent, arXiv:2603.11759v1, supplies a best-known-line candidate and counterexample to treating navigation as fully informed, one-shot selection. Its model reproduces partial inspection, premature choices, and backtracking; it does not test FPF entries or require the reader to run a navigation model.
The first-result comparison and reliance-conditioned history are FPF adaptations. Reopen this choice if a same-budget title/snippet or other entry preserves result discrimination and wrong-turn recovery with less burden, or if reader evidence shows that the added return cues distract from the first useful choice.
Mnemonic question. When is a repeatable formula worth the extra space in a cross-pattern entry? The selected line is conditional mnemonic support tested against the same truthful content without the mantra, not an acronym or repetition by default.
Cue question. When a reader's familiar wording or language hides another useful target, should the entry merely translate the same label, show more results, or explain the choice nearby? The selected line is the smallest situation-and-result cue that lets this reader distinguish the targets, with an expansion only for a distinction that cannot fit truthfully. The alternative is a concise translated title or ordinary search snippet. Keep that alternative when it already distinguishes the use; otherwise prefer a short explanation of what the reader can obtain over extra synonyms. This trades a little reading space for a visible choice, without requiring full translation, a multilingual interface, or another public scenario.
Zhu, Reinecke, and Mitra, Language Scent: Exploring Cross-Language Information Navigation, arXiv:2604.03604v2, supplies bounded evidence for considering such proximal cues: its multilingual system exposes information value and interpretation cues, and its lab study involved 16 English–Chinese speakers. It does not compare FPF names or establish that two labels have the same referent.
Adapt as a local probe: E.11:4.1.1, 4.1.2, and 4.5.1 test recognizable wording against the direct result; E11-9/10/11 keep cue, entry, and direct content distinct. Reopen if a shorter label works equally well for the actual readers, if the cue invites the wrong result, or if stronger evidence changes the transfer from multilingual navigation. No universal benefit from contextual wording is inferred.
For a cross-DPF entry, E.11.DSG supplies the four-return, full-source and known-DPF-bypass rules used in E.11:4 and E11-13. The Suite's disciplinary and framework composition follows C.20 and E.4:4.2; the entry's actual contributions can also come from external suppliers. E.8, E.17, F.17, F.18 and E.11.PUA retain their direct authoring, publication, naming and use functions. Reopen the affected entry when its supplying contribution or conditions change; use G.11 when currentness or telemetry is the actual question.
Retrieval-exposure question. Which part of a publication can help an assistant find the needed pattern? The selected line makes the actual indexed and returned context explicit. A title-only change cannot repair a passage whose retrieval representation omits that title.
Adapt: §4.6.1 separates heading, row, body-window and fragment comparisons. Anthropic's Contextual Retrieval (2024) supplies a concrete context-enrichment alternative; Conti et al., Context is Gold to find the Gold Passage (EMNLP 2025) supplies context-dependent retrieval evaluation and contextual embedding training.
These interventions differ from title revision and provide no effect estimate for it. Keep the cheaper existing cue when it suffices; reopen when the consuming system changes its fields, chunking or context, or when the changed cue repeatedly selects the wrong body.
Relations
- Builds on:
E.8for pattern recognition text,E.17.AUDfor publication-unit discipline,F.17andF.18for published terms and naming, andC.2.1for public helper epistemes. - Leads to:
E.11.PUAfor applying one selected pattern andE.11.PURfor local applicability, recommendation, and coordination. - Coordinates with:
E.11.PFPfor the shared ordinary-entry and card forms;E.4.FPFandE.4.DPFfor each product's selected non-exhaustive example keys and forms plus its compactness measure and two limits, withE.4.DPF:4governing the return from a contribution-spanning Card or mantra to candidate recognition;A.22.CGUSfor independently admitted demonstrative slices;E.18for flow-local results;G.11for currentness orchestration;E.11.DSGfor the architectural reference of the named framework, Suite or Library, its question-led entries and direct-use bypass, and returns to the constituent or external Methods, results and sources that change the answer; and each direct pattern cited by a public entry.
E.11:End
Pattern Use in a Working Situation and First Useful Result
Type: Pattern-language use pattern (E) Status: Stable Normativity: Normative within FPF pattern use from a current practical question to the first useful result or honest stop.
Problem frame
Use this when
Use this pattern when an agent has a current entity or relation of concern, a practical question, and one plausible FPF pattern, and needs to follow that pattern's Solution to reach the smallest useful result that truthfully answers the question, or to stop because the required basis is absent.
A reader may ask, “This pattern looks relevant. What do I do with it, what useful result should I expect, and when should I stop or return?” Answer that conversationally before considering any durable pattern-use record.
First useful result. One result entity, obtaining relation, honest interim result, or exact blocker that answers the current question well enough to stop or continue. The ordinary result is named in domain language; exact predicate, identity, work, and reliance fields open only when they change the claim.
Primary EntityOfConcern. One use of one selected FPF pattern for a current practical question, ending at one useful first result or honest stop. A later use is named only when a current continuation or reliance needs it.
What this buys. The user gets a direct path from a pattern to a useful subject result without confusing pattern inspection, method description, planning, performed work, and result evidence. An addressable trace remains available when another agent or tool will use the distinctions, or an audit or delayed decision will depend on them.
Not this pattern when. While several public entries remain plausible, keep comparing them under E.11; begin PUA only after one direct pattern is selected. Use E.11.PUR when applicability, recommendation, or coordination among several candidate uses is the current question. Use E.18.1 when a wider method, plan, work, interpretation, and return flow must preserve accepted problem-side distinctions. Use the A.15 family when intended or performed U.Work is itself current.
Problem
Reading a pattern does not by itself apply it. Without a usable pattern-use method, readers stop at recognition, create a meta-card instead of the subject result, or report a plan, note, generated answer, or support record as if the intended physical, clinical, organizational, learned, or epistemic result already existed.
The opposite failure is also common: every bounded use is burdened with a shortlist, candidate form, fit records, provenance graph, and closure dossier. The paperwork becomes the apparent result and obscures the direct Solution.
FPF adds no generic U.Result, U.WorkProduct, U.PatternApplication, or generic Use kind. First identify the result entity or obtaining relation and the direct pattern whose content defines, constrains, or tests it.
Identify an exact predicate, pattern episteme, ClaimGraph, Method, plan, dated Work, Transformation, evaluation, decision, later-use object, or category-correct basis only when that distinction changes the truth, the next action, the stop, or a named reliance.
Forces
Solution
Use one selected pattern through a short result-oriented procedure. Keep the subject result in the foreground; add exact identities or addressable pattern-use records only when ambiguity or a named later reliance needs them.
Cheap first screen before formal work identity
Start with five ordinary values: the working subject, the question the recipient commissioned, the selected pattern's Solution, the first useful result or blocker, and the stop or return. For a bounded reversible use, those values are sufficient when the result serves that use and the boundary tells the practitioner what has been accomplished.
When following a pattern exposes a costly, infeasible or apparently unhelpful demand, use C.11.DUA to examine what that work can change and repair the continuation. Follow the selected method through its first useful completion, including any prescribed fields or records. Keep the result and its useful basis; retain a reason or limitation when the current decision or a later recipient needs it. An inactive inquiry adds no omission account. This route is conditional on the demand being in question.
An FPF pattern supplies action- or judgement-guiding content; a person or another capable system uses it. The ordinary instructions “use this pattern” and “apply this pattern” are harmless shorthand.
Only when the selected Solution describes a Method and that distinction changes the claim, use A.3.1 to identify the admitted U.Method and A.3.2 to test U.MethodDescription membership of the pattern episteme.
Name a System, system-role classification, assignment, plan, dated Work, result, or U.Transformation only when that object is part of the current claim. Assignment never substitutes for the acting System, Work, authority, or responsibility.
When those identities do matter, keep them separate: the pattern episteme is not the acting System or Work; a selected or project-tailored Method is not automatically a WorkPlan; intended work is not performed Work; a result, evidence for it, and a later use are different values. This conditional distinction introduces no universal workflow, causal chain, production relation, TFS, or record requirement.
Before Method selection, distinguish the missing operation. Use C.39 when a sufficiently clear local result lacks an adequate obtaining explanation; it can finish with one explained candidate or the precise unresolved contribution. Use C.40 when usable material needs variation, examination and warranted continuation, whether or not its ultimate contribution is settled; its C.40:4.3 entry adds problem construction and target-tested transfer when those are needed together. A sufficient existing Method or result exits directly, and ordinary use of a selected pattern requires neither search.
If the working question is still represented only by a pre-method-selection TaskSignature, use that signature to constrain method search; do not treat it as the task, plan, or Work occurrence. Use OEE or NQD to retain Method or architecture candidates before selection, and use G.5 to declare a selected-set result.
For publication, use E.17 for a source-backed face and return to source and E.24.PUB for the occurrence, form, carrier, audience, bounded use, and availability. Use A.3.1 to identify a selected Method and A.15 for planning and Work.
Open these distinctions only when candidate retention, selection, result declaration, publication, planning, or performed Work is current.
The ordinary seven-step use
Before making any pattern-use record, ask what the recipient needs to decide, explain or do. What can this use of the pattern establish for that work? Stop when the commissioned result is available. If this result is one premise, carry it back to the outstanding question.
-
Recognize the working situation. Name the subject or relation in ordinary domain language and ask the current practical question. State an exact kind now only when a nearby kind difference can change the pattern or result.
-
Inspect one direct pattern. Read its Problem frame, Problem, Forces, Solution, Consequences, ordinary boundary, and nearest stronger neighbor. Do not select from its title or one trigger word alone.
-
Say what useful result would answer the question. Name the entity, obtaining relation, honest interim entity, or blocker plainly enough to distinguish it from a plan, description, recommendation, work occurrence, or nearby value. Name the Method, plan, dated Work, Transformation, evaluation, decision, or later-use object relative to which it is a result only when the phrase depends on that object. Add an exact kind, predicate, pattern locator,
ClaimGraph, or category-correct basis only when ambiguity or replay makes it necessary. -
Use the pattern's
Solution. The agent follows the guidance under its stated conditions. Name a system-role classification or assignment only when that claim matters; it necessarily matters for a precise Agent or actual-Work performer claim because A.13 requires both in the performer core.If actual Work is current, first recover every precise performer's A.13 core; A.15.1 then independently admits the dated Work from the exact performance history, enacted Method, extent, and containing-System relation. Add F.6 afterward only when this use needs precise assignment-bound attribution through the same obtaining A.13 assignment; otherwise the Work-only account stops after A.15.1.
Identify a Method, authority, responsibility, or another independent value only when its own claim is current. A responsibility claim names its predicate and participants, or the A.6.RCD missing governor; routine pattern use needs no such expansion.
Use A.15.PROD when the current claim must settle whole or proper-part production Work, Work-attributed first existence, or production completion; apply only the branch that answers that claim.
-
Check what now exists or obtains. Identify the result under the direct pattern whose content defines, constrains, or tests it. A pre-existing entity may instead receive new grounding for the current question. If the expected subject result still does not exist, name the honest interim result and leave the subject expectation open. Do not turn grounding, planning, evaluation, acceptance, publication, or non-agentive change into production.
Compare the obtained result with the original receiving use, not only with the selected pattern's expected output. At an existing demonstration or handover, inspect what the result now enables and any extra adjustment needed to make it usable. If the account leaves a consequential difference unexplained, B.5.PI helps form the inquiry; the user need not already know an anomaly label or another PatternID.
A reported concern is not closed by a correct answer to a substitute question. B.5.EA helps recover what the formulation lost. Use the available episode and text before asking the participant to supply a diagnosis. An adequate direct result still finishes this use without an extra audit.
-
State the immediate continuation only as needed. Name a later use, stronger neighboring pattern, or unresolved clarification in conversation. Materialize an expectation, basis, result, flow, provenance, or boundary episteme only when a named later use needs it to remain addressable.
-
Stop or return. Stop when the smallest useful result, honest interim entity, or exact blocker answers the current question at the precision that use needs. Return when the concern, basis, expected entity, direct pattern, relation, or later-use condition changes. A stop needs no receiver.
An exact blocker can answer the current question. When reporting a subject or interim result, distinguish three cases. A new entity or relation occurrence becomes current under its own rule and basis; use the applicable A.15.PROD branch only when a production-work, Work-attributed first-constitution or completion claim is current.
A pre-existing entity remains unchanged while a grounding finding becomes adequate for this use.
Or the expected subject result remains absent while an honest interim result and return condition become explicit.
Read a reference by the contribution needed here
Read what the sentence asks you to do. A citation gives the basis of a claim or explanation already stated; an application reference asks or invites you to use another contribution. “This distinction follows X” and “use X to compare the arrangements” serve different purposes, even if they point to the same pattern. E.8:4.2.2.2 defines this writing convention. A compact list may use Source: or Apply:, but ordinary clear prose needs no such labels. An address that only names a relation participant is neither kind merely because it is an address.
At a citation, use the quoted or explained material directly when it is sufficient for the present work. Inspect the source when its assumptions, meaning, currentness or evidence need examination for your reliance. The citation neither certifies the claim nor imposes an automatic source-reading step.
At an application reference, distinguish a required dependency, a conditional next move and one possible way. Follow a branch when it is required or chosen and its conditions hold; needing the result does not make every suggested Method compulsory. Reuse an adequate result after checking that it answers the present question under the current conditions.
If the contribution is missing, obtain the relevant explanation, apply what is needed and bring its result back to the work that needs it. Reading a title or naming the pattern supplies no such result. An already understood Method need not be read again merely because it is referenced.
A reference to comparison, for example, is satisfied by an adequate comparison of the actual alternatives and conditions, not by a list of patterns to read or by choosing a familiar alternative. A direction to reuse an existing sufficient arrangement concerns that arrangement's adequacy for the receiving work; it is not permission to stop because several relevant pages have been read.
If the purpose is ambiguous, recover it from the referring passage and destination, not from the URL, document kind or absence of a label. Establish what the next operation needs and whether the current passage already supplies it. If the purpose or necessary material remains unavailable, keep the dependent result open and continue work whose basis is available. Report a consequential ambiguity as a text defect; do not turn uncertainty into either reading every citation or skipping every reference.
Stop following references when the needed contribution and its conditions are sufficient for the current use. This is not a requirement to read all related patterns, reproduce a cited theory or invent a missing connection privately. Where the connection remains unexplained, use C.39 to obtain or develop it before claiming the larger result.
Reliance profiles
In ordinaryBounded use, the subject, practical question, inspected pattern, useful result or blocker in ordinary language, and stop or return remain recoverable in conversation. State a relative object, exact kind, predicate, pattern locator, ClaimGraph, or category-correct direct basis only when it distinguishes the result from a nearby value. No candidate basis, fit record, flow-position record, provenance note, closure record, or receiver is required.
In relianceBearing use, materialize only the distinctions that the named reliance will use. Another reader may need a candidate basis and rationale. Automation may need an exact result kind, predicate, pattern locator, ClaimGraph, relative object, and category-correct basis. Delayed review may need a descriptive flow position and a separate later-use disposition.
A receiver appears only for an actual communication or admitted route relation; an ordinary return needs only its condition and optional next-pattern locator. No profile causes every support record to be materialized.
In the optional forms below, claimGraph carries the episteme's complete asserted content, including the reasons and qualifications that change what it says. The scheme designation resolves under C.2.1:4.2.6 to the rules actually governing that account. It can select another episteme's effective scheme or exact content defining those rules; a reference to a document alone does not choose between them.
Preserve an unresolved interpretation rather than filling the field with a nearby name. These forms organize addressable writing when needed; ordinary conversation and partial notes remain sufficient under the entry conditions above.
The trace is absent from ordinary conversational use. When materialized for a named reliance, C.2.1 identifies it through claim content, exact EntityOfConcern, and effective reference scheme. claimScopeRef, modelUseStructureRef, and projectWorkRef are present only when the exact neighboring relation changes the pattern use; they are not additional episteme-identity fields, and the reference alone does not make that relation obtain.
The expectation names the exact result kind, predicate, defining or constraining ClaimGraph, and pattern locator; it also names the kind of Method, plan, dated Work, Transformation, evaluation, decision, or dependent-use object relative to which the result phrase would be true, and one category-correct basis branch. It asserts neither existence nor obtaining.
For a selected candidate use, the obtained-result core positions—from obtainedResultRef through obtainedResultDirectBasisRef—are present together or absent together; a rejected candidate leaves them absent.
In the direct-relation branch, the claim graph exposes predicate, participants, applicability, obtaining, occurrence identity, and defining ClaimGraph.
In the A.6.1 branch, it exposes the operation, application, argument or result binding, and defining ClaimGraph.
In the local-claim branch, the direct relation-or-binding locator is absent, the A.6.RCD derivation-rule locator and every base-predicate ClaimGraph locator are present, and the claim graph exposes polarity, substrate or constructor, base predicates, participants, case facts, and any support or warrant required by the dependent use. The claim episteme does not obtain, and A.6.RCD replaces none of its base predicates.
A reconsideration names conditionalNextQuestionPatternLocator only when that continuation is current. A genuine stop leaves the field absent. No receiver is fabricated merely to complete the trace.
Admitted support species and rule-content locators
@Context in these legacy support-species names is a compatibility and retrieval suffix. It names no U.BoundedContext, universal situation, project container, relation, or identity field. Every support episteme follows C.2.1 identity. Claim scope, bounded model use, project work, qualification window, and other working conditions enter only through the exact neighboring object and direct relation needed by the receiving use.
The defining ClaimGraph located at PUA states the practical-question, optional compact-trace, candidate-basis, candidate-support-episteme, candidate-rationale, result-expectation, result-closure-finding, and dependent-use-disposition-finding schemas. The exact rule content at E.11 states public-card comparison rationale; E.11.PUR states fit, applicability, recommendation, coordination rationale, coordination, and ordering.
A candidate basis is an ordinary C.2.1 claim about the exact candidate when that addressable account is needed. It cites actual supporting claims, independently obtaining relations or evidence uses under their own definitions. It introduces no generic candidate-basis or additional-basis U.Relation.
PUR separately governs prospective coordination and precedence. Its result-dependent readiness claim needs the actual result and category-correct basis, including an adequate earlier result; a PUA closure finding is materialized only when named later reliance needs addressability. The prospective ordering relation itself does not require that result already to exist.
PUA findings cite the result predicate, defining or constraining ClaimGraph, pattern locator, and one category-correct direct basis. In the local-claim branch they keep the A.6.RCD derivation-rule locator distinct from every base-predicate ClaimGraph locator. They introduce no result or actual-use relation kind.
Question, boundary, and expectation
The expectation never proves that the result entity exists, that a relation or binding obtains, or that a local claim is true. It first identifies the result kind, predicate, defining or constraining ClaimGraph, and pattern locator. It then names which kind of exact Method, plan, dated Work, Transformation, evaluation, decision, or dependent-use object a real closure must identify, and which direct basis would make the readable result phrase true relative to that object.
The basis description is branch-specific: relation occurrence; A.6.1 operation-application binding; or A.6.RCD local C.2.1 claim with polarity, substrate or constructor, base predicates and their ClaimGraph locators, participants, case facts, and any support or warrant required by the dependent use. The last branch is not an obtaining basis, and the derivation-rule locator is not a substitute for any base predicate.
The flow position is a descriptive PUA position. intendedUseClaimRef, intendedReceivingGovernedObjectKindRef, intendedReceivingUseDescriptionRef, and dependentUseReconsiderationBoundaryRef are present together only when an actual continuation or named later reliance is current; otherwise all four are absent. A genuine stop needs no receiver.
In a boundary record, return, wrongTurnRecovery, strongerNeighbor, and receivingPatternContinuation name conditions and optional next-pattern locators, not receivers; stop, missingGovernor, and missingInformation invent none. Receiving-position kind and ref are both present or both absent.
candidateAdmission means that Problem frame, Forces, Solution conditions, expected result, category-correct basis template, and ordinary boundary are recoverable enough for further inspection; it is neither an applicability finding nor a selection.
Candidate basis under named reliance
Construct a durable candidate only after inspecting the direct pattern's Problem frame, Problem, Forces, Solution, Consequences and ordinary boundary. When a named reliance needs its basis addressable, state one ordinary C.2.1 claim whose EntityOfConcern is that exact CandidatePatternUse episteme. Say which direct Solution content and case facts support using the pattern for this practical question and expected result, with their applicability and unresolved or contradicted premises. Ordinary conversational use need not materialize this account.
An optional representation of the basis claim is:
The claim states how each cited object supports the candidate's question and expectation. It keeps each object's kind and the asserted use recoverable. A README template can supply a reusable starting point, while case facts come from the actual subject, practical question, effective scheme and action-changing working conditions.
Reference respelling that still designates the same source changes no source referent. Added evidence can change support without changing that source. A richer display may preserve the episteme. Changing its claim content, EntityOfConcern or effective ReferenceScheme identifies another candidate or basis episteme under C.2.1. A direct basis relation remains a separate option only when a receiver needs its occurrence and its participants, predicate and identity are independently settled under A.6.REL.
The public template is absent when the candidate was formed by direct pattern inspection without a README template. directSolutionSectionRef is the Solution section of directPatternRef. A project-tailored MethodDescription is a separate U.MethodDescription under A.3.2. If dated Work first constitutes that episteme and the inception claim matters, state the exact A.15.PROD assertion; any derivation or reuse relation to the direct pattern episteme remains separate. Applicability, recommendation, and coordination remain exact E.11.PUR assertions.
Rationale subjects stay distinct
Candidate rationale has one candidate subject. The ClaimGraph located at E.11.PUR defines the coordination-rationale schema over a declared candidate set. The ClaimGraph located at E.11 defines the public-card comparison-rationale schema over one public guidance episteme before a project candidate is constructed. Keep those rationale subjects distinct.
Joint-success candidate and its premises
A candidate proposes using A.9 to compute the probability that both named components succeed, with marginals 0.9 and 0.8. The expected product 0.72 is supported by A.9's joint-success case only when the independence premise and the model's applicability to these events hold.
PUR can use those claims in its existing fit/applicability judgement without requiring a fused basis relation. Spelling an A.9 reference differently while resolving the same source does not change the support's referent.
Changing the candidate's expected event from “both” to “at least one” changes its claim content; the product expectation cannot be retained as though only the reference display changed. Under the independent alternative-success model in A.9, the corresponding value would be 0.98, requiring that different candidate content and result basis. Neither basis claim authorizes Work, delivers the result, or supplies a result-closure finding.
Actual-result closure and receiving-use disposition
PUA introduces no actual-result relation and no universal actual-use relation. Keep two questions separate: what establishes, under the applicable identity or predicate rule, that the candidate result entity exists or the relation occurrence obtains; and what category-correct basis makes the readable result phrase true relative to the current Method, plan, dated Work, Transformation, evaluation, decision, or later-use object. One relation occurrence may answer both questions only when the result itself is that occurrence.
When a named later use needs addressable closure, materialize a C.2.1 finding that states the result assertion, locates its defining or constraining rule content through the applicable ClaimGraph and pattern locator, and records the category-correct basis:
The three flow-position values are descriptive PUA positions, not kinds, relations, or occurrence identities. The finding's claim graph names the result entity, its exact predicate, defining or constraining ClaimGraph, pattern locator, the object relative to which the result wording is true, and exactly one direct-basis branch.
For a direct relation occurrence it names predicate, participants, applicability, obtaining, occurrence identity, and defining ClaimGraph.
For an A.6.1 binding it names operation, application, argument or result binding, and its defining ClaimGraph.
For an A.6.RCD local C.2.1 claim, the relation-or-binding locator is absent; the claim ref, polarity, substrate or constructor, base predicates, their ClaimGraph locators, participants, case facts, and any support or warrant required by the dependent use are recoverable, with the derivation-rule locator named separately. Evaluate the claim’s support separately from whether its base relations obtain.
If the result itself is a relation occurrence, entityOfConcernRef and resultDirectBasisRef may designate that same occurrence. The closure finding reports those facts; it creates none of them.
When the closure asserts production-work participation, Work-attributed entity inception, or production completion, use the applicable A.15.PROD branch. An inception claim requires independently identified dated Work, actual changes and the applicable identity rule. A completion claim keeps subject-state satisfaction separate from the governor that closes the production Work.
A relation occurrence may first obtain through its direct predicate; an evaluation or decision becomes current under its exact subject assertion and defining ClaimGraph; a non-agentive change needs no production claim. Completion, evaluation, acceptance, publication, continuation, and later use remain separate.
If the claimed result existed already, use 4.6 instead. If no direct basis is recoverable, retain the independently identified entity and return the exact missingGovernor or missingInformation boundary rather than minting a closure relation.
Path slice and DesignRunTag are both present only when the exact result-bearing position and its one TFS are already recoverable under E.18; otherwise both are absent. These fields are provenance cues, not identifiers for another TFS, a network, or a cross-flow relation.
Record receiving-use disposition separately, and only for a named reliance:
In the realized state, the later object, kind, direct-basis kind, and basis ref are present; the intended positions are absent. The same branch rule separates a direct relation or A.6.1 pattern locator from the A.6.RCD derivation locator and the direct pattern locators for the local claim's base predicates. The claim graph exposes the exact participants and facts.
In intendedNotYetRealized, the intended claim, object kind, description, and condition are present together and all realized positions are absent; an intention is not an obtaining-use relation. A stop without a later use has no disposition finding.
When ordinary language says that a result from one TFS is used as an input, tool, context, or constraint in another, treat those words only as cues. Name the exact result-bearing position and exact receiving position—one FlowPositionRef for each—plus the direct relation occurrence connecting their participants and its applicable predicate, and keep the result's kind unchanged.
With no direct relation kind or predicate, return missing-governor; with a predicate but undecided facts, leave the relation open and name the grounding boundary; with a false predicate, assert no occurrence; with an obtaining occurrence but a missing endpoint binding, return missing-endpoint-binding and name that binding.
Use E.18 for each TFS-local position and local DesignRunTag; use E.18.NET only when independently identified TFS values must be treated together as a network. No input, tool, context, constraint, result, or adjacency label supplies the direct relation.
When the thing being called a result is U.Work, identify that dated occurrence under A.15.1. Planning, setup, authorization, triggering, or enabling work does not produce that Work. Call the occurrence a result of the selected use in a reliance-bearing closure only when the exact category-correct basis for that reading is present; otherwise keep the Work and the pattern-use description separate.
Pre-existing and still-absent subject results
When the expected entity existed before the current use, the current use may establish a C.2.1 grounding finding about that unchanged entity:
Each GroundingBasisPair preserves one relation occurrence and the exact pattern content defining or constraining it. The finding's ClaimGraph names the grounded proposition and covered subject claim. For a C.2.1 episteme a pair may cite an exact EpistemeEmpiricalGroundingRelation; for another subject it uses the direct measurement, observation, evidence-use, diagnostic, or subject predicate that actually grounds that proposition. Inspection, a record, or evidence proximity does not ground the entity by itself. If no direct grounding basis is recoverable, return its exact blocker. The pre-existing entity is not newly produced.
If the current use calls the grounding finding its result, add a separate PatternUseResultClosureFinding@Context whose EntityOfConcern is that finding. Its direct basis must connect the finding to the current method, plan, Work, transformation, evaluation, decision, or receiving-use object through a relation occurrence, A.6.1 binding, or category-correct local claim. The occurrence that grounds the pre-existing subject does not by itself make the grounding finding a result of the current use. Cite A.15.PROD only when exact dated Work and its actual changes first constituted the finding episteme.
In reliance-bearing use, when the expected subject result still does not exist, close the current use only on an interim PatternUseResultClosureFinding@Context. Identify that interim entity under its own kind and rule, and record the category-correct basis that makes it the current use's result relative to the current object. Keep the subject-result expectation open. A machining plan does not become a machined component; a treatment recommendation does not become a changed clinical state; an assessment plan does not become learned capability.
Reliance-bearing final-practice test
Use this test when the declared teaching, rehearsal, or evaluation use is to establish that a participant can select a pattern, preserve the kind and direct basis of its result, and leave another participant a replayable continuation. This exercise is relianceBearing because the evaluator relies on the selected basis, expectation, grounding state, and continuation.
Inspect what the exercise supplied. When a tutor names the pattern or the form gives its decisive question, the participant can still demonstrate application and preparation of a continuation. Credit independent selection when the choice was open and the participant made it. Earlier attributable work can supply that evidence. To entrust the complete job to the participant, establish that they can take their own choice through application to a handover the recipient can use. For instance, a decision log can show which pattern they selected and why, while the matching work and handover show what they did with that choice. A prior selection in one job and execution after a tutor supplied the question in another leave that connection open.
Its row count is a test condition, not a general rule for pattern use. The test does not require or assert a wider CGUS.
Each practice continuation description states an action or proposed use, the expected result and its kind, the full PatternID and pattern name, and the condition under which that continuation is current. Its entityOfConcernRef resolves to the same selected CandidatePatternUse@Context as the test result.
For full credit on this exercise, the participant must select the pattern, preserve the kind and direct basis of its result, and carry that selection through application to the required handover. The handover must include three to five continuations, at least one with continuationDisposition=branch or return, the final continuable work position, and one consequential unknown with its minimum clarification pattern and expected clarification-result kind. A prompted application can earn credit for the work actually done. A prior complete performance can qualify the participant for a subsequent assignment. To report that a successor actually resumed the work, establish what that successor did. A client who commissioned only a usable result can accept it under the terms of the order.
The selected basis claim concerns that same candidate; the candidate names the same question and expectation as the test result, and the expectation and test result name the same descriptive flow position.
The practice descriptions remain ordinary PUA epistemes whether or not a wider CGUS qualifies. When demonstrativeSliceRef is present, that post-qualification slice shows the existing descriptions in its declared display order; it does not create wrapper rows, replace or retype the descriptions, or change the candidate, subject result, or continuable work position.
SelectedFirstResultGroundingStateValue is newlyCurrentSubjectResult | preExistingWithGrounding | expectedSubjectResultAbsent. Exactly one state branch is filled:
- For
newlyCurrentSubjectResult, fillnewlyCurrentSubjectResultClosureFindingRefand leave the other state positions absent. The closure separates the rule under which the result exists or the relation obtains from the basis that makes it this use's result. A relation occurrence may first obtain through its direct predicate; an evaluation or decision becomes current under its own rule; an actual non-agentive change remains under A.3.4. Cite the applicable A.15.PROD branch for a current production-work, inception or completion claim. A Work-attributed first-existence claim requires the exact dated Work, actual changes and applicable identity rule; completion remains a separate claim. - For
preExistingWithGrounding, fill bothpreExistingResultGroundingFindingRefandpreExistingGroundingResultClosureFindingRefand leave the other state positions absent. The grounding finding names the already-existing entity, its paired grounding relation occurrences, and the exact pattern content that defines or constrains each relation. Its separate closure uses another category-correct basis to make that finding the exercise's result; a subject-grounding occurrence alone does not. Cite A.15.PROD only if exact dated Work first constituted the finding episteme. The exercise does not produce the pre-existing entity. - For
expectedSubjectResultAbsent, fillexpectedSubjectResultAbsentInterimResultClosureFindingRefand leave the other state positions absent. The interim entity keeps its own kind and rule; the closure records the relative object and category-correct basis required by this reliance. It may support later work but does not satisfy the selected subject-result expectation.
Fill selectedFirstResultReceivingUseDispositionFindingRef only when the declared test relies on an addressable realized or intended receiving use. It must point to the selected state-specific closure, including the grounding-finding closure in the pre-existing branch.
The continuable-work description says what project work can proceed from this state-specific result. The test fails when it merely retells a card, expands into a whole-project plan, treats a public template as a recommendation, claims performed work without an A.15.1-grounded U.Work, infers a physical, clinical, organizational, or learned change from its description, or asserts a CGUS only because the practice contains several rows.
Replay and currentness
For immediate ordinaryBounded use, recover from the conversation the working subject and question, the direct pattern inspected, the useful result, honest interim entity, or blocker, and the stop or return. Recover a relative-object kind, exact predicate, pattern locator, ClaimGraph, or category-correct direct basis only when it changes the truth, distinguishes a nearby value, or is needed by a named reliance. Do not reconstruct a candidate dossier, flow position, or receiver merely to replay a cheap local use.
When a named later use relies on fuller replay, recover the exact EntityOfConcern, effective reference scheme, practical question, selected direct pattern and edition-pinned Solution, expected result kind and pattern locator, descriptive flow position, relative object, category-correct direct basis, grounded actual or honest interim entity, any separately current later-use disposition, and stop or return boundary from the support epistemes materialized for that reliance. Add claim scope, project work, model-use structure, qualification window, receiver, or another working condition only through its exact neighboring relation when that relation changes the replayed use.
Recheck the smallest affected claim or relation when the concern, candidate basis, direct Solution, expected result, result grounding, flow position, receiving-use condition, or boundary changes. Reopen pattern selection only when that change alters candidate fit; a new measurement of the same result does not by itself select another pattern. G.11 governs edition, telemetry, currentness-window, and decay orchestration; PUA supplies the use-specific values and change conditions that orchestration inspects.
Archetypal Grounding
Episteme result: a usable problem card
A team has a vague recurring pump-failure concern and asks whether it can be articulated well enough to guide later method selection. In cheap ordinary use it can say, "Use C.22.2 to make a usable problem card," then state the bounded concern, affected entity, obstacle, stakes, evidence state, and honest next use. Those contents can leave the exact C.22.2 ProblemCard episteme and selectedPatternApplicationFlowResult position recoverable without stating or recording either one. Name them explicitly only when a nearby kind confusion or named reliance requires replay; C.22.2 and C.2.1 still govern the episteme.
If the card did not exist before this exercise and the team claims that the drafting episode first constituted it, identify the dated drafting U.Work, the card's C.2.1 identity rule, the actual changes, and the local A.15.PROD entity-identity-inception claim. That claim establishes the card's Work-attributed inception, not by itself that the card is this PUA use's result.
A reliance-bearing closure separately names the category-correct basis that makes the card the result relative to the current pattern use or another named relative object.
Without the inception basis, do not say that pattern application produced the card; state only the card content and leave its inception provenance open. Any later P2W participation uses its exact direct relation or local claim. The team need not materialize PUA closure records during a cheap conversational use.
Evaluation specification without a ProblemCard
An architecture team already has a bounded comparison question and needs no accepted C.22.2 ProblemCard episteme. It applies A.19.ECS and states one exact EvaluationCharacteristicSpaceSpec with declared coordinates, scales, comparators, and evidence rules; A.19.ECS and C.2.1 govern that specification episteme. The optional problemCardRef remains absent.
The application-flow label is only a readable PUA position. If the team claims that exact planning or specification Work first constituted the episteme, cite its local A.15.PROD inception claim for that subject fact. A reliance-bearing PUA closure separately identifies the category-correct basis that makes the specification a result relative to the current pattern use or another named relative object. If a later comparison actually uses the specification, cite the exact direct relation, A.6.1 binding, or local relation-bearing claim defined or constrained by that comparison pattern. Without the corresponding basis, keep the specification, its PUA closure, and the later comparison separate.
A selection result can support later planning
E.11.PUR governs the identity and content of one PatternUseRecommendation@Context; A.15.2 separately governs one U.WorkPlan. patternSelectionFlowResult and selectedPatternApplicationFlowResult are descriptive positions that keep these two entities apart. If either episteme is claimed to have been first constituted by dated Work, cite its own local A.15.PROD inception claim rather than saying that selection or application generically produced it.
If the recommendation participates in later planning, state how its content is used and retain the applicable predicate, A.6.1 binding, or supported local claim that establishes this participation. Identify exact source and receiving flow positions only when the receiving question needs the selected E.18 organization; an asserted crossing between independently identified TFS values retains the direct relation and endpoint bindings required by E.18.NET. If a required basis is missing, keep the recommendation and plan distinct and state the corresponding unresolved claim or binding. Neither the recommendation nor the plan becomes the machined component expected from downstream subject work.
Build-the-builder recognition case. An executable compiler edition occupies one exact result-bearing position (FlowPositionRef) in a compiler-build TFS and participates at one exact compiler-use position (FlowPositionRef) in a separately identified program-compilation TFS through a direct compiler-use relation occurrence. Its applicable predicate identifies that occurrence; the compiler edition keeps its kind. Use E.18 when either TFS-local position is unresolved; use E.18.NET when the question is how the separately identified build and compilation TFS values form a network, including a recursive one.
With no compiler-use kind or predicate, return missing-governor; with undecided case facts, keep the relation open; with a false predicate, assert no compiler-use occurrence; with an obtaining occurrence but a missing endpoint binding, return missing-endpoint-binding and name that binding. None of these branches permits calling the compiler edition the second flow's input by label alone.
AI-assisted ordinary use returns the subject result
An engineer asks an AI assistant to apply an already selected A.19.ECS pattern to a pump-comparison question. The needed result is one exact EvaluationCharacteristicSpaceSpec with admitted coordinates, scales, comparators, and evidence rules. No later use asks for a durable pattern-selection trace.
The assistant returns the specification content in ordinary language and keeps the concern, pattern, and stop condition recoverable in the conversation. The text is the required specification only when it satisfies the A.19.ECS and C.2.1 identity rules. Successful ordinary use creates no candidate, fit, applicability, rationale, expectation, or closure record merely because AI helped.
If the use also claims first constitution, recover every precise performer's A.13 core and let A.15.1 independently admit the dated Work; add F.6 afterward only when the closure also consumes exact assignment-bound attribution through the same obtaining assignment, then state the A.15.PROD inception claim. A short attribution projection may omit only an assignment identifier unused by its receiving claim; every fact consumed by the A.13/A.15.1/F.6 basis remains recoverable. A Work-only account stops after A.13 and A.15.1. Work is not a responsibility bearer.
If the available basis cannot support the specification, name the unresolved coordinate, scale, comparator, or evidence-rule position, use A.19.ECS to resolve it, and leave the completed-specification expectation open. Materialize that return as PatternUseBoundaryCondition@Context only when a named reliance needs an addressable boundary; do not emit a complete meta-record stack.
Physical result: work is still future
A machining team applies a planning pattern for a dimensionally accepted component and states one exact U.WorkPlan under A.15.2. If it claims that planning Work first constituted that plan episteme, cite the local A.15.PROD inception claim. The metal blank remains unchanged.
The plan is the result at the selectedPatternApplicationFlowResult position and keeps its A.15.2 identity. The component remains an open downstreamSubjectWorkFlowResult expectation until dated machining Work occurs. The team may continue with the A.15 work patterns; it cannot fill the component position with the plan, simulation, inspection checklist, or generated prose.
After machining Work occurs, identify the A.3.4 transformation and the applicable work-to-change predicate or admitted local claim. If the Work first makes a new component satisfy its identity rule, cite the separate A.15.PROD inception claim; if a completion criterion is satisfied, cite the separate completion claim. Evaluation, dimensional evidence, and acceptance remain separate.
A reliance-bearing PUA closure must additionally name the category-correct basis that makes the component or changed state the result relative to the machining Work or another named relative object; neither the flow position nor result wording supplies it.
Clinical result: a state and its note stay separate
A clinician uses a direct decision pattern to state one treatment-plan episteme. The decision pattern and C.2.1 govern the plan; a claim that dated decision Work first constituted it requires its local A.15.PROD inception basis. The plan may describe an intended receiving use, but it does not establish that treatment occurred or that the patient's state changed.
After treatment Work occurs, identify the clinical change through A.3.4 and the applicable treatment-work-to-change predicate or admitted local claim. The clinical state and the case-note episteme can both be current, but they keep different identities and relations. The note may support grounding and later reliance; it does not become the patient's state.
If the clinically relevant state existed before the current pattern use, state what the examination, measurement, diagnostic, or evidence-use relation establishes about it for the present question. Materialize a PreExistingResultGroundingFinding@Context only when a named reliance needs that addressable account. The examination does not produce that state or supply an unknown earlier treatment history. Missing direct grounding returns its exact blocker.
Learned capability and assessment remain separate
For a precise teaching-performer claim, first recover every performer's A.13 core and let A.15.1 independently admit the dated teaching Work; add F.6 only when exact assignment-bound attribution through the same obtaining assignment is current. Use the applicable educational pattern for the educational claim.
A later assessment episteme may support a claim that the learner demonstrated a bounded capability or skill only when an applicable educational, assessment, evidence-use, or subject predicate establishes that claim. The capability and the assessment episteme remain distinct.
A completed lesson, assessment plan, or filled record cannot occupy the learned-capability position by itself; if no applicable capability predicate is available, return missing-governor rather than a generic learning result.
Pre-existing result: inspection does not reproduce it
A maintenance engineer inspects an installed pump that predates the current pattern use. Current measurements may ground the pump for a compatibility question only through the exact measurement, observation, diagnostic, or evidence-use relations defined by the applicable patterns; the historical production claim lies outside the basis.
State the present grounding and its direct basis; use PreExistingResultGroundingFinding@Context and its exact GroundingBasisPair values when a named reliance needs that addressable account. An inspection note or record may describe the pump and support evidence use, but it cannot replace the physical pump or occupy the expected subject-result position. Keep producing-work provenance absent. The current inspection neither manufactures the pump nor proves how it was manufactured. If the direct grounding relation cannot be recovered, return that blocker instead of treating inspection proximity as grounding.
Repair a plan-as-component closure locally
A machining rehearsal selected the correct planning pattern and stated a valid U.WorkPlan, but its closure named the plan as a downstreamSubjectWorkFlowResult and treated the component expectation as satisfied. The concern, candidate basis, direct pattern, and WorkPlan remain sound.
Repair the expectation and closure finding: place the U.WorkPlan in the descriptive selectedPatternApplicationFlowResult position, cite its exact A.15.2 identity and any current A.15.PROD inception claim for Work-attributed first constitution, and separately cite the category-correct basis that makes this plan the result relative to the current pattern use.
Remove the unsupported component closure and any realized receiving-use finding that depended on it, and keep the component as an open downstream expectation. The next current use enters the A.15 work family. No new candidate selection or reconstruction of the WorkPlan is needed.
Complete trace, absent result
An automated report raises pattern-use trace completeness to 100 percent by filling every candidate, rationale, expectation, and boundary position. Operators begin treating the green report as completion, while the direct basis for the claimed result and the intended receiving use remain absent more often.
The trace measure improved while subject progress worsened. Keep completeness as a trace-quality measure, apply E.13 to the substitution, and judge PUA success by the useful result or honest interim entity, the direct pattern and basis needed to make that claim true, and any separately current later-use disposition. Empty result-basis positions are not repaired by adding more support records.
Bias-Annotation
- Recognition-only bias. A matching title or trigger word is treated as use. Repair by inspecting the full direct pattern and naming the result or blocker its
Solutioncan support. - Record-as-result bias. A candidate form, trace, note, dashboard, or assessment record replaces the subject result. Restore the exact entity or relation occurrence and, when needed, the direct pattern and category-correct basis that make it this use's result.
- Plan-as-work bias. Intended work or a generated plan is reported as performed work. For every precise performer, recover A.13 first and let A.15.1 independently ground the dated occurrence; add F.6 only for a current exact assignment-bound attribution.
- Flow-collapse bias. A selection result, selected-pattern result, and downstream-work result are merged because each is called “result”. Restore the distinct results and the basis for each; name an E.18 crossing only when that crossing is current.
- Maximum-trace bias. Every use emits every schema. Return to the named reliance and materialize only distinctions it will use.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits. A cold reader can use one pattern and reach a useful result without learning a meta-workflow. Ordinary use remains light. High-reliance use can preserve the exact result assertion, the pattern content defining or constraining it, the relative object when relevant, the category-correct basis, and any separately current later-use disposition.
Costs. A claimed result needs enough direct basis to make the claim truthful; reliance-bearing use may add addressable findings and exact locators. Cross-flow participation requires both TFS-local positions and the direct relation only when that crossing is current. Ordinary bounded use pays none of this trace cost merely for completeness.
Rationale
FPF patterns supply action- or judgement-guiding content for recurring working situations. Some selected Solution sections describe methods; others define, constrain, test, or guide a judgement without doing so.
Establish formal U.MethodDescription membership under A.3.2 only when the episteme's EntityOfConcern is an admitted Method, at least one claim concerns that Method as a way of doing, and that distinction changes the claim.
The missing middle is neither discovery nor recommendation: use one selected conditional Solution to identify the first result with its own identity and basis that answers the current question, or stop when its basis is missing.
Separating ordinary semantic checking from conditional record materialization protects both usability and rigor. A conversation can be sufficient for a bounded reversible question. Later use by another agent, an audit, an automated use, or an expensive decision can demand addressable support. The same ontology serves both profiles; only the reliance changes the recording granularity.
The first-result boundary prevents proxy completion. A plan, note, simulation, or assessment may be valuable and may be the result at the current pattern-use position. It cannot stand for a later physical, clinical, organizational, or learned change. Stop where the current basis supports value, then continue under the pattern that defines, constrains, or tests the next result or relation.
SoTA-Echoing
Conditional pattern use is the lineage anchor, not a claim that traditional pattern-language practice already supplies PUA's result and flow ontology. Recipe following, PatternID matching, and universal trace production remain the common comparators.
The 2026 navigation study is a current preprint anchor rather than settled consensus. Reopen its adaptation when peer review, replication, or use evidence changes the observed value of inspection, memory, or backtracking. Reopen the ordinaryBounded and relianceBearing split when real later uses repeatedly lose needed distinctions or pay trace cost without later use. G.11 orchestrates that evidence and currentness; PUA changes the profile boundary or exact support relations.
Relations
- Builds on:
E.11for public entry,E.8for action-guiding pattern form,E.18for each TFS-local position and localDesignRunTag,E.18.NETfor a current network,A.6.P.WMRandA.6.RCDwhen a relation rule or direct claim cannot be recovered, A.13 for every precise performer core, A.15.1 for independent dated-Work admission, F.6 only for current assignment-bound attribution, A.15.PROD for the separately current production-work, Work-attributed entity-inception or production-completion claim, A.15 for wider planning and work coordination, C.2.1 for support epistemes. - Coordinates with:
E.11.PURfor applicability, recommendation, and coordination;E.18.1for accepted problem-to-work carry-through;E.22andE.23for evaluation and repeated improvement;G.11for currentness; and each direct pattern that defines, constrains, or tests the selected result. - Use next when current:
E.11when no direct pattern is selected,E.11.PURwhen recommendation or ordering among several uses is current, and the direct pattern for any result or work claim beyond PUA's boundary.
E.11.PUA:End
Pattern-Use Applicability, Recommendation, and Coordination
Type: Pattern-language use pattern (E) Status: Stable Normativity: Normative for deciding applicability, recommendation, and coordination among candidate FPF pattern uses, with addressable support only when a later use relies on it.
Problem frame
Use this when
Use E.11.PUR after one or more candidate pattern uses have been inspected and an agent needs to decide whether each use fits, which use to recommend, how several uses should be coordinated, or whether an earlier result already answers the current concern.
The candidates may remain conversational in ordinary bounded use; addressable CandidatePatternUse@Context values are required only when a named later reliance needs them.
Primary EntityOfConcern. One current applicability, recommendation, or coordination judgement over already inspected candidate pattern uses. When that judgement must remain addressable, it may be represented by PatternUseApplicabilityFinding@Context, PatternUseRecommendation@Context, or PatternUseCoordination@Context; a PatternUseOrderingRelation@Context exists only inside the coordination it qualifies.
The @Context suffix on these compatibility support names is retrieval wording only. It names no bounded-context entity, generic situation, project container, relation participant, or identity field; every episteme follows C.2.1 identity, and an ordering relation follows its own participant, condition, obtaining, and occurrence rules.
What this buys. A team distinguishes applicability from recommendation and presentation order from workflow order. A result that still answers the concern can end a repeated pattern use. A project can preserve exact reasons for a consequential recommendation without burdening ordinary bounded use with five separate forms or copying one result under another stage name.
Not this pattern when. Use E.11 while public entries are still being compared. Use E.11.PUA to use one selected pattern and obtain its first result. Use A.15 for work planning or performed work, A.21 for a gate decision, and the direct decision or authorization pattern when those claims are current.
In this pattern, next move is Plain shorthand for the currently recommended pattern use or conditional continuation. It is not a shared Move identity, U.Method, U.WorkPlan, performed U.Work, or actual U.Transformation.
Problem
Several claims are often compressed into “use this pattern next.” A pattern can fit the Problem frame but fail its Solution conditions. It can be applicable yet not be worth recommending: another use may offer a more useful first result, or continuing the present work may serve the concern better once the full burden of a new use is counted.
Several candidate uses can belong together without forming a sequence, and displaying a sequence creates no WorkPlan, performed work, Transformation, or transformation-flow structure.
When these distinctions are missing, familiar PatternIDs become proxies for value. Teams recommend the pattern they know, copy one result description into several order relations, and treat a diagram or teaching order as execution order.
Forces
Solution
Evaluate candidate uses against five fit aspects. An ordinary reversible judgement may remain conversational: keep the aspects in one compact rationale, state the aggregate applicability when the current question needs it, then compare the expected receiving value and full burden of the serious continuations, including continuing without a new pattern use.
Before repeating a recommended use, apply the result-reuse branch in 4.2.1. Materialize separate findings or a recommendation episteme only when a named later use needs addressable support.
When coordinating several candidates, recover the result their use together must support. Specify order where one use requires something from another. Also identify any common condition or compatibility on which the result depends, including when either use can come first.
Fit and applicability
In the optional forms below, claimGraph carries the episteme's complete asserted content, including the reasons and qualifications that change what it says. The scheme designation resolves under C.2.1:4.2.6 to the rules governing that account. It can select another episteme's effective scheme or exact content defining those rules; a reference to a document alone does not choose between them.
Preserve an unresolved interpretation rather than filling the field with a nearby name. These forms organize addressable writing when needed; ordinary conversation and partial notes remain sufficient under the entry conditions above.
The five criteria refer to one candidate. When its basis must be addressable, consume PUA §4.4's ordinary C.2.1 claim about that candidate and its exact supporting claims, relations or evidence uses; no candidate-basis U.Relation is required. A contradicted product-law premise and missing dependence information remain distinct inputs to these fit tests.
In ordinary conversation, inspect all five without materializing five findings. State the aggregate applicability when it answers the current question; if the later comparison supports a recommendation, carry that result into the recommendation.
PatternUseApplicabilityFinding@Context is the reliance-bearing support episteme: when it exists, its five findings cover each criterion exactly once.
Use mutually exclusive branches: any misfit yields inapplicable, including when another aspect has insufficientBasis; with no misfit, one or more insufficientBasis values yield insufficientBasis and a missing-basis boundary; otherwise all five are fit and the result is applicable.
A known misfit settles this candidate's applicability under the inspected conditions. Do not obtain more information merely to fill the remaining aspects when it cannot change that answer. If changing the failed condition is a serious continuation, appraise that change separately and reconsider the affected fit under 4.6. A different candidate may still be applicable and worth recommending.
problemFrame compares the candidate pattern's Problem frame with the current concern; it does not assert that an actual Problem obtains.
When an actual Problem is relied on, cite one current C.22.PFR ProblematicForRelation occurrence with its exact actual-condition and criterion-applicability participants and adverse-episode identity. A ProblemCard, fit finding, assessment, or recommendation may support a claim about that occurrence but neither creates nor splits it.
Recommendation
Before recommending a new pattern use, compare its full receiving value and burden with the other serious continuations, including continuing the present work without it. Use C.11.DUA when the worth of the advice, its presentation or its interaction with other proposed changes is unclear.
Applicability can coexist with no worthwhile recommendation now. In that case, finish without selecting a pattern use or materializing a positive recommendation; give a short reason only when the current question or later reliance needs it.
Continuing without optional advice does not remove an existing obligation.
When the comparison supports a recommendation, state which candidate is applicable, why its expected first result warrants its full burden for the current concern, and where to stop or return. If the judgement is local, reversible, and has no named later reliance, that readable statement is sufficient.
When the recommendation must remain addressable, use the schema below. ordinaryCompact keeps one compact rationale and no five-finding dossier; relianceBearing adds the current applicability finding only because a named later use needs independent replay.
A recommendation selects one applicable candidate whose expected receiving value warrants its full burden compared with the other serious continuations, including continuing without a new use. When a receiving use is current, the rationale explains how the recommendation supports it. A conversational judgement needs no record.
In an addressable ordinaryCompact recommendation, the applicability result and compact rationale are carried directly and applicabilityFindingRef is absent.
In relianceBearing, the same recommendation also cites one current applicability finding whose five fit findings can be replayed independently. The profile changes support cardinality, not the recommendation kind or authority.
When an addressable recommendation is materialized, expectedResultExpectationRef points to its exact E.11.PUA expectation. It identifies the expected result and only the pattern, relative-object, or category-correct basis distinctions that expectation uses; it does not assert that the result exists or that any relation, A.6.1 binding, or local claim is current.
A recommendation does not authorize work, establish a gate, prove evidence sufficiency, create the expected result, or supply its later closure.
When a stronger neighboring pattern better addresses the current question, name it and state the return condition. Populate strongerNeighborPatternRef only when the exact pattern identity matters to an addressable recommendation. The reference does not establish formal U.MethodDescription membership; such membership requires its own A.3.2 basis. Familiarity with the current candidate is not a recommendation reason.
Reuse an earlier result when it still answers the concern
After identifying an applicable candidate use, ask whether an earlier result episteme already answers the present concern. Use the pattern that defines or tests that result to compare:
- the result episteme's
EntityOfConcernand edition; - the question answered and declared use;
- the source and dependency conditions on which the answer relies; and
- its qualification and currentness boundary.
When those values still match, cite and use the earlier result.
Use A.10 when a named claim or bounded action relies on that result and the source-to-use account is still implicit. Add a dated U.Work occurrence only when that Work is itself a current claim. Use G.11 when currentness or refresh changes the use.
When the earlier result matches, stop without repeating the same pattern use or copying the result under another stage name.
When one value changed, reopen the smallest affected result question under the pattern that defines or tests that result. Repeat the complete pattern use only when the unaffected reach cannot be established. If the applicable pattern supplies no basis for comparing the earlier result with the present concern, stop at insufficient result-comparison basis.
The candidate pattern use, the earlier result episteme, a later reliance relation, a currentness assertion, and later Work remain separate. This branch introduces no generic result-reuse relation.
Coordination without forced order
For a question about sequence, state the candidates, whether they are unordered, partially ordered or totally ordered, the reason for any precedence, and the stop boundary. For a question about their joint use, follow each required contribution into the same decision or work and check that the contributions can serve together under its conditions. Two uses in the right order can concern different versions of the object. The order answers the sequencing question. Version compatibility can decide whether the resulting conclusion is usable. Retain the rationale and any established ordering relations as addressable epistemes when a later use needs them.
unordered has no ordering relations. partialOrder and totalOrder use explicit pairwise relations. A total order is the bounded PatternUseSequence@Context specialization under its named receiving use; it is not a universal route or project WorkPlan.
Treat the declared pairwise precedence as strict: its transitive closure must be irreflexive. A totalOrder additionally compares every pair of distinct members.
A cycle or unresolved required pair blocks that ordering claim; preserve the known dependencies and state the unresolved coordination question.
Pairwise precedence
A pairwise precedence relation is an applicable directed constraint between two exact candidate uses for the governed coordination question. It can obtain before either is performed. Establish its branch predicate:
- prerequisiteResult: the dependent use's direct rule requires exact content, and the coordination has a stated basis for selecting this prerequisite use to supply it. Cite that rule, content and selection basis together with the prerequisite candidate's exact expectation. Matching output kinds alone is insufficient. The result need not yet exist for the dependency to obtain.
- methodPrecondition: performing the dependent use first would invalidate a condition that the prerequisite use needs, and the proposed order preserves that condition. Name both uses and the condition; unary candidate fit alone supplies no pair.
- sharedConstraintResolution: an applicable schedule or priority rule selects this direction to resolve the actual shared constraint. Shared-resource exclusion alone establishes conflict, not A-before-B. A rule selecting B-before-A defeats the proposed A-before-B claim.
The candidates are distinct members of the coordination. The relation obtains while the applicable branch predicate holds for those candidates, the governed question and semantic condition. A known false branch condition establishes non-obtaining; a missing rule or needed case fact leaves it unresolved. An unmet prerequisite result is not a false dependency condition: it concerns readiness below.
Identity and extent. Keep the same dependency for the same exact candidates, governed coordination use, branch and meaning of its precedence condition, including any constitutive schedule/priority rule. A different rendering, witness or descriptive coordination episteme alone changes none of those values.
A changed candidate or meaning-changing condition/rule requires a new relation claim and identity assessment; the record reference is not itself the identity law.
Within a temporally qualified use, the occurrence lasts while that branch predicate holds; a gap where it ceases to hold ends the occurrence, and a later re-established occurrence has a new extent. Omit temporal machinery for an atemporal constraint.
Readiness is a separate claim. For prerequisite-result continuation, identify the actual result and category-correct basis that satisfy the required content and receiving conditions.
When a named later reliance needs addressable closure, cite a current E.11.PUA result-closure finding for the prerequisite use or an adequate earlier result reused under §4.2.1. State readiness/applicability for this continuation as its own ordinary C.2.1 claim.
A known unmet condition means not ready; missing information leaves readiness unresolved. An expectation, ordering edge or closure record's presence alone supplies no achieved result.
The dependency is not an instruction to repeat a use whose adequate earlier result is already available. Apply the result-reuse exit before proposing execution.
A changed closure can change readiness without changing the prospective dependency. No ordering or readiness claim authorizes Work. Page, seminar, identifier and display order do not establish a dependency.
Procedure
-
Recover each candidate's current concern, direct pattern, Solution, expectation, and ordinary boundary.
-
Keep a local reversible applicability, recommendation, or coordination judgement conversational when no named later reliance needs it. When a recommendation must remain addressable, choose
ordinaryCompactunless that reliance needs the fit aspects separately addressable; userelianceBearingonly for that reliance. -
Inspect all five fit aspects. In ordinary use, keep them in one compact rationale. Under
relianceBearing, materialize five separate findings and one applicability finding. -
Establish the aggregate applicability under
4.1. A knownmisfitends the applicability inquiry for that candidate under the inspected conditions.If the aggregate is
insufficientBasis, obtain missing information only when an attainable answer can change a worthwhile continuation; otherwise return the missing-basis boundary.State the aggregate separately when the current question needs it. If step 5 supports a recommendation, include it there; when a reliance-bearing applicability finding exists, the two result values agree.
-
Compare an applicable candidate with the other serious continuations, including continuing the present work without a new pattern use. Recommend it only when its expected receiving value warrants its full burden; use C.11.DUA when that judgement is unclear. Finish without a positive recommendation when no candidate warrants one. The expectation is not an achieved result.
-
Before repeating a recommended use, compare any earlier result through
4.2.1. Reuse a matching result or reopen only the affected result question. -
Establish the ordering the receiving work needs under §4.4. When the commission concerns a combined result, use the rationale to explain how the contributions fit that result, including any joint condition the ordering leaves open.
For prerequisiteResult, name exact required content, the selected supplier-use basis and its expectation.
Assess readiness separately from precedence. When a continuation depends on a prerequisite result, check whether an actual result and its basis satisfy the receiving conditions. Reuse an adequate earlier result before requiring another use.
-
Stop at the applicability answer or missing-basis boundary, recommendation, matching earlier result, coordination result, or conclusion that no new use is worth recommending now.
The last outcome selects no candidate and requires no refusal document; it leaves existing obligations in force.
A Plain next move, when one is useful, names only the recommended pattern use or conditional continuation. Continue to PUA, P2W, planning, gate, decision, or work only when that next claim becomes current.
Replay and currentness
Replay an ordinary conversational or addressable compact recommendation from the current concern, inspected candidate pattern and Solution, aggregate applicability, compact rationale over all five aspects, serious continuations considered, expected result and full burden, any current receiving use, and recommendation boundary.
Replay a reliance-bearing recommendation from those same positions plus the current applicability finding and its five fit findings.
Replay coordination from its inspected candidate uses, question, ordering mode, any pairwise precedence and bases, stop boundary, and, for each prerequisiteResult relation, the exact expectation, required content and selected supplier-use basis.
Replay readiness separately from precedence, using the actual result, its category-correct basis, receiving conditions and any earlier-result reuse; recover an addressable closure finding only when that reliance needs it.
When a later use needs to replay a conclusion without a recommendation, recover the concern, the serious continuations, the value and burden that mattered, and the condition for reconsideration. No selected-candidate reference or five-finding dossier is required for that conclusion.
Replay a result-reuse stop from the earlier result episteme and edition, the question and declared use, relied source and dependency conditions, qualification and currentness boundary, and any separately current A.10 reliance or G.11 assertion.
Recheck the smallest affected finding, result question, or relation when a candidate Solution, result expectation, result entity or edition, relative object, direct basis or defining ClaimGraph, relied source or dependency condition, qualification or currentness boundary, fit basis, value or burden, alternative under consideration, dependent use, coordination member, precedence basis, condition, or boundary changes.
A changed candidate fit reopens its applicability and any recommendation that relied on it. A changed earlier result condition reopens only the affected result question and later uses unless the candidate or present concern also changed.
A changed prerequisite expectation or semantic precedence condition reopens the affected dependency and dependent use. A changed closure reopens readiness and only dependencies whose semantic basis also changed. A changed witness or rendering alone creates no new dependency.
Recheck a constitutive schedule or priority rule when it changes.
Separate G.11 assertions state edition, telemetry, currentness-window, and decay facts; PUR supplies the judgement-specific values and change conditions.
Archetypal Grounding
Applicable but not recommended
A team considering a high-cost pump test has candidate uses of C.28 causal triage and A.21 gate discipline. Both may be applicable. The immediate uncertainty is whether a causal model output may support intervention, so C.28 offers the more useful first result. That uncertainty and the recommendation are epistemic; neither asserts an actual C.22.PFR Problem.
Recommend C.28 without claiming that the test is authorized. The later gate use remains a separate candidate whose applicability can be reconsidered after the causal-use result exists.
Because this local recommendation is reversible and no named later use relies on it, the team states the applicability result and one compact rationale over all five aspects in the working conversation; it materializes no recommendation episteme or support profile.
If a later gate review needs to replay each aspect independently, that review may create current fit findings and a current applicability finding from the then-current basis.
Do not backdate the later findings. If the original readable rationale was retained, it remains the earlier recommendation's historical basis.
An earlier result still answers the concern
The same team already has a C.28 causal-use result for the same pump model, intervention question, declared use, sources, assumptions, and qualification window. The team uses C.28 to compare that result episteme and edition with the present concern. Every comparison value still matches, so the team cites the result and stops instead of performing the C.28 use again.
If the team uses the result as a premise in a project discussion and the source-to-use account is still implicit, it uses A.10 to make the source → result → claim connection explicit for that bounded use.
If later gate Work relies on that result, retain its required evidence and authority grounds; use A.10 when the source-to-use account remains implicit or the gate's replay needs require that account, naming the result and bounded use.
If new operating conditions change the causal-use assumptions or qualification window, the team reopens that affected C.28 question rather than treating the old result as current or restarting every coordinated pattern use.
No new use is worth recommending
A team considers an optional check that fits its concern and conditions. The most it could save is ten minutes; performing it takes an hour. The present work already meets the needed result, and no obligation requires this check.
The team establishes applicability, compares the hour with the possible saving, and continues without selecting a new use. It needs neither a recommendation episteme with a fictitious candidate nor a refusal document.
If asked why, it states the cost comparison in the conversation.
If new information makes that check capable of preventing a consequential loss, the team reopens this comparison. If the check becomes obligatory, the earlier optional-advice conclusion does not discharge the obligation. A less burdensome check with a worthwhile expected result can support the ordinary positive recommendation instead.
Unordered complementary uses
A clinical team needs both a terminology repair and an evidence-basis review before revising a protocol. Neither result is a prerequisite for the other in the current context.
State an unordered coordination with respect to the two results. Neither is a prerequisite for the other. Available working hours may still determine which pattern can be used first. With a flexible calendar one practitioner can choose either sequence. To schedule the uses concurrently, check that the people, permissions and resources for both are available in the same interval. The local judgement needs no coordination episteme or ordering relation.
If the later protocol revision becomes a named reliance that needs the coordination replayable, materialize one PatternUseCoordination@Context with orderingMode=unordered and no ordering relations. Coordinating these two uses establishes neither a lifecycle nor a WorkPlan.
Result-based precedence
A design team's comparison rule requires the exact evaluation coordinates for candidates X and Y and their intended review question. The coordination selects its A.19.ECS candidate to supply that EvaluationCharacteristicSpaceSpec under a stated criterion for this task. The known rule, required content and supplier selection establish prerequisite-result precedence even while the spec is absent. Readiness is false when absence is known, or unresolved when its existence/applicability is unknown.
When the actual spec and its basis meet those receiving conditions, readiness can be supported. A named later reliance that needs addressable closure also requires the current PUA finding.
If an earlier ECS result already meets them under §4.2.1, reuse that result without repeating ECS. A same-kind spec for another question is insufficient.
If the comparison rule no longer requires this content, the dependency fails; if the rule itself is missing, the dependency is unresolved. Neither result changes candidate inspection or supplies Work authorization.
When a method precondition establishes precedence
An already-current material-kind classification is a solutionConditions fit fact for one machining candidate; it supplies no second candidate and no ordering pair.
Candidate A inspects the intact surface of one specimen and candidate B destructively tests that specimen. If B destroys the surface condition A needs and A-before-B preserves both applicable uses, the methodPrecondition predicate supports that direction.
If B leaves A's required condition unaffected, this ground supplies no precedence. Two individually fitting candidates alone do not establish an edge.
Shared resources and changes to support
Two candidate uses require the same exclusive instrument. Exclusion alone leaves their coordination conflict unresolved.
An applicable selected schedule S1 assigning A before B supplies the directional sharedConstraintResolution basis; an applicable S2 selecting B before A defeats that A-before-B claim.
An unknown priority rule leaves the direction unresolved. Adding both opposite edges would violate the strict-order cycle condition.
If a coordination description copied an old ECS expectation, replace the copy with the current exact expectation and required-content/supplier basis. Reassess the dependency only where that semantic content changed.
A new closure or witness can instead change readiness without changing the ordering relation.
A new rendering of S1 is descriptive; an actual change from S1 to S2 is constitutive and requires the changed order to be assessed. Candidate fit and unrelated dependencies remain usable.
A higher recommendation score can reduce useful fit
An assistant ranks candidate pattern uses by historical recommendation acceptance. The familiar A.21 gate candidate receives a higher score and is recommended first more often for causal-use uncertainty. Recommendation acceptance rises, but wrong-turn returns also rise because the needed C.28 causal-use result is still absent.
The score improved while first-result fit and receiving-use value worsened. Keep the historical score as telemetry, apply E.13 to the substitution, and base the recommendation on current applicability, expected result, any current receiving use, and the other candidates still under consideration. A higher score is not another fit finding.
Bias-Annotation
- Applicability-as-recommendation bias. A fitting pattern is automatically selected. Compare its expected receiving value and full burden with the serious continuations, including no new use; the best of several unhelpful candidates need not be recommended.
- Favorite-pattern proxy bias. Familiar PatternID substitutes for current value. State the concern, expected result, and any current receiving use in the rationale.
- Five-form bias. Every ordinary use creates five findings. Keep them in one compact rationale unless their separate identity is relied on.
- Sequence bias. Presentation order becomes precedence. Repair by naming the pairwise basis.
- Result-copy or expectation-as-result bias. A prerequisite result kind is duplicated in ordering fields, or its expectation is treated as achieved. Reuse the prerequisite candidate's exact expectation. Assess readiness from the actual result and basis; cite a current E.11.PUA closure finding when named later reliance needs it. The finding reports those facts rather than creating them.
- Stage-name repetition bias. The same result question is answered again because a later review or phase uses another label. Compare the earlier result through its direct pattern, reuse it when the relevant values still match, and reopen only the changed result question.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits. A team can explain why a pattern fits, why another is recommended, how several uses relate, and when an earlier answer remains usable without creating a false workflow. Ordinary reversible judgement remains light; reliance-bearing recommendations remain replayable. Prospective precedence retains its exact required content and supplier basis; readiness is judged from the actual result and its basis, with earlier-result reuse and addressable PUA closure when needed.
Costs. A consequential or delayed-use recommendation needs an explicit rationale and may need five addressable fit findings. Partial orders need justified pairwise relations. Ordinary local judgement pays no record cost merely for symmetry, and candidates that answer different questions are not forced into a scalar ranking.
Rationale
Applicability, recommendation, and coordination answer different questions. Applicability asks whether a candidate's conditions hold. Recommendation asks which applicable use best serves the current concern. Coordination asks how several candidate uses belong together. Keeping the questions separate prevents a familiar label or score from becoming an unexamined decision.
A set of candidate pattern uses can be unordered, partially ordered, or totally ordered. Only an applicable directed constraint justifies an edge.
A prerequisite-result edge can be established from the required content and selected supplier basis before the result exists. The actual result and its category-correct basis are required for current readiness, including through a reusable earlier result. A PUA closure finding makes that basis addressable when a named later reliance needs it.
The alternative enabled-only edge would leave the known planning dependency unstated. This preserves graph structure without turning every explanation into a chain or minting a generic result relation.
Result reuse answers another question: whether an already obtained answer still serves the present concern. The direct result pattern supplies that comparison; A.10 supplies any later reliance account; and G.11 supplies currentness when it matters. Keeping those contributions separate avoids both duplicate work and a generic reuse relation that would hide why the result remains applicable.
SoTA-Echoing
Que et al. is the current decision-bearing recommender source in this narrow use; Nunes and Jannach supplies lineage. The 2026 navigation preprint supplies bounded reconsideration, while current FPF NQD, OEE, A.19, C.2.1, A.10, and G.11 supply the transdisciplinary candidate, comparison, result-identity, reliance, and currentness basis.
Reopen the score-proxy adaptation when stronger evaluation evidence shows that the relied-on score tracks current expected-result and receiving-use fit without the identified exposure or rationale loss.
Reopen the wrong-turn adaptation when peer review, replication, or use evidence changes the value of reconsideration.
Use G.11 to manage source and telemetry currentness, then revise the affected PUR fit, rationale, recommendation, or return relation.
Relations
- Builds on:
E.11.PUAfor candidate uses, expectations, rationales, and boundaries;A.6.5for slot discipline; andE.18for coupled-flow relations when results cross flows. - Coordinates with:
E.11for public discovery;C.22.PFRfor an actual Problem;A.19,A.19.ECS, andA.19.CPMfor characteristic-space construction and comparison;E.18.1for P2W; the pattern that defines or tests an earlier result for result reuse;A.10for later reliance on that result;G.11for currentness; and the direct pattern that defines, constrains, or tests any stronger plan, work, transformation, gate, evidence, decision, authorization, result, or basis claim. - Leads to:
E.11.PUAfor using the recommended pattern, or to the exact neighboring pattern when the stronger claim becomes current.
E.11.PUR:End
Framework Publication Form Profile for Markdown
Type: Specialization of E.11 Status: Stable Normativity: Normative unless marked informative.
Problem frame
Use this pattern when one FPF, DPF, or LPF edition needs a public Markdown form that a cold reader can enter and a small deterministic checker can recognize. The framework's pattern set, product boundary, and edition values must already be selected for the publication being assembled or checked.
The first useful result is a publication arrangement in which readers can find both the whole-framework explanation and each directly useful pattern. Name the edition source and public units; locate the twelve substantive answers under section 4.7 and each missing, reordered, duplicated, unresolved, or mismatched form element. A passing form check does not accept the edition, prove framework adequacy, identify a carrier, or establish a publication occurrence.
Do not use this pattern to decide whether one pattern set is a framework, whether a catalogue or guide is another product, or whether an edition is current or available. Use the E.4 family for the product and framework boundary, E.24.PUB for publication occurrence and carrier relations, and the applicable decision, quality, and currentness patterns for those claims.
Problem
FPF-family publications can expose the same useful material through different headings, edition labels, index layouts, and Readme cards. A familiar reader can compensate. A cold reader or parser cannot reliably tell which edition is present, which index is authoritative, whether several Part tables form one index, or whether a support table is a rival front door.
Even a navigable collection can leave readers unable to explain what its Methods accomplish together, why they are arranged that way, which assumptions and sources support several patterns, or how a narrower profile changes use. Those answers can be scattered through development notes and disappear when the pattern bodies are published.
A rigid carrier template can put authorship, credits, dates, status, dependencies, build details, and maintainer records ahead of the reader's question whether or not those facts change the reader's choice. It can also force a catalogue, inquiry programme, guide, or other adjacent product to pretend that it is a framework edition. The common form must be exact where shared recognition matters, put the practitioner's question first, and explicitly limit its scope to framework editions.
Forces
Solution
Apply common reader-facing publication-form rules to construct or inspect a particular expression of one FPF, DPF, or LPF edition. This profile states requirements for the selected reusable rules and their realization in the expression. Keep the expression, its reusable form, the independently identified presentation carrier whose actual state yields it, the expressed edition, and the publication occurrence distinct under E.24.PUB.
Preserve the compact product opening
For an all-in-one Markdown publication, preserve the product-declared compact opening and use this H1 route:
# <product-declared publication title>;# Table of Contents;- the exact product-declared Readme H1;
- the exact product-declared Architectural Reference H1;
- the pattern bodies or pattern collection in the order selected by that edition; and
- reference and maintenance material under headings declared by the product pattern.
The title and Readme H1 are separate product declarations. A checker receives both exact strings; it does not derive the Readme H1 by concatenating Readme to a longer carrier title.
The common profile does not insert a metadata block, edition record, warning, or other lines into a compact predecessor opening merely to make products look alike. A product-specific builder may pin a compact front shape, including the line at which the ToC begins, when that shape protects an established reader entry.
Between the title and ToC, retain only the shortest public cues already justified by product use. An exact edition designation or locator belongs there only when its possible values change the reader's next use, reliance, return, language, dependency, or access choice.
When such a cue is present, project it from one product-owned edition or relation record; do not maintain a second editable copy. Add authorship, credit, date, dependency, language, access, or a product-declared maintenance status, support window, or currentness window only under the same next-working-move test.
A date is a cue, not edition identity, and a visible status or window is not evidence of acceptance, currentness, maintenance, availability, access, or authorization.
When readers copy, adapt, translate or redistribute a framework publication, its copyright-license notice tells them what permission the rights holder grants. Identify the covered material, rights holder and chosen license, and make the license text reachable from a standalone copy. Project a compact notice from the product's chosen licensing terms; retain applicable attribution and mark any third-party or differently licensed material.
The author of an independently created DPF or LPF chooses its license. Where copyright permission is needed, reuse of another framework's protected expression is governed by that material's terms; applicable copyright exceptions remain available. Adopting its methods or publication form does not impose that framework's license on the new work.
Reader front matter extends from the opening title through the Readme and architectural reference up to the first pattern-body collection H1. It must not contain campaign keys; candidate, review, or result identifiers; local disk or repository paths; source or candidate digests; Git commits or blobs; generated comments; build commands; machine warnings; or "do not edit" instructions. Detailed edition, provenance, rebuildability, and maintenance records remain adjacent maintainer evidence or product-declared reference-tail material unless a separately selected public use justifies a reader-facing projection.
Put public units into the established Table of Contents
Immediately after the single # Table of Contents H1, continue the product's established ToC grammar. Represent the exact Readme and architectural-reference unit before the logical pattern index using the same kind of labelled segment and rows already used for non-pattern units in that product.
When an established ToC already represents the whole-account unit and pattern groups, add Readme there; do not invent a generic Publication route, a second mini-menu, or a new table shape.
A non-pattern publication unit receives no fabricated PatternID. Its product-declared entry remains mechanically recognizable and, when the carrier supports links, resolves to the exact unit. The unit's public title follows E.4's <name of the explained whole> Architectural Reference rule; its declared section key is a separate address choice.
Place the one authoritative logical pattern index after those public-unit entries. It may be one table or several ordered, uniquely labelled Part or placement segments. Every authoritative segment uses:
Across all segments, every published pattern body has exactly one row, every current PatternRef row resolves to exactly one complete body, and no PatternID or reserved locator appears twice.
Explicit PlannedCatalogEntry rows reserve future locators under E.8 H-10; they are outside the current-pattern row/body bijection and supply no available pattern or current dependency. A Part label groups rows for navigation; it is not a pattern row, a semantic parent, or another index.
PatternID, title, Part, and § position remain separate even when one row displays them together. PatternID supplies the stable public address within the named framework; the title explains the pattern; Part and § show where the current edition places it. Within each Part, the ToC rows and pattern bodies follow the same order. That order need not ascend by PatternID, and moving or retitling a pattern does not by itself change its PatternID.
For a portable reference, apply E.8 H-10 to retain the complete PatternID or SectionRef and enough context to recover the publication set and document. Add a qualifier only where that identification is not already unambiguous.
A publication may declare a short DocID beside its full document title. State the publication set in which the declaration is unambiguous. The declaration identifies the whole document, not its embedded Readme, Architectural Reference or pattern collection.
A framework's existing reference code may be declared as that DocID without changing any PatternID; a standalone Architectural Reference may declare its own document qualifier. A document without a DocID remains referable by its full title within the named set.
Keep the document, the framework it publishes, the selected edition, its publication units and its carrier distinct. Several units can share one document; one framework can have several edition documents or carriers. A reference to an exact edition names that edition separately. A current-carrier catalogue derives its document identities from their declarations and full titles; it does not confer framework membership, pattern identity or authority.
Keep document identity recoverable from cross-document link labels and their accompanying context; add a qualifier when the complete address does not already supply it. The hyperlink may still provide a convenient carrier. An extracted passage retains the complete local address and any identifying context it needs.
For a DPF, E.4.DPF:4.0.3 supplies the reference-code, local-locator and continuity decisions; this profile makes the selected distinctions visible.
Use E.8:4.2.2.2 to distinguish a source citation from an invitation to apply another contribution. Keep that purpose and any use-changing condition clear in ordinary prose. If an extract loses the heading that supplied this meaning, retain it in the extract; preserving the address alone does not preserve the intended reading. No repeated labels are needed when the wording is already clear.
Reserve Support index — <lookup job> for a secondary pattern lookup. Its exact header is:
Ordinary relation, source-return, maintenance, and reference tables may cite PatternIDs under truthful headings and other complete headers. Do not infer that they are indexes from their cell values. Reject a second # Table of Contents, a Pattern Index heading for the same job, an authoritative header outside the authoritative ToC region, or a support heading and header that do not occur together. Public-unit entries are navigation inside the one ToC, not another pattern catalogue.
Keep one practical-entry set and two visible forms
Start the Readme body with ## Practical entries. The product maintains one declaration for every selectable example and assigns each key exactly one public form: ordinary practical entry or Practical-Use Card. Each declared key occurs once, at H3 for an ordinary entry or at H4 for a card. A compact locator may precede or follow these examples, but it is a finding aid rather than another editable entry set.
The Readme explains what its selected examples help the reader do. In the opening, invite the reader to start from the work and question they bring. Apply E.11:4.1 when showing how to use an entry, including one that sounds familiar but answers only part of the question. Keep the index, direct patterns or another finding aid available. Use F.19:4.4 when the reader also needs help interpreting the scope of the catalogue.
Tell readers that they can ask an assisting agent for an explanation or comments in ordinary language. For example, a reader commissioning a method could ask, “I need to decide who will design this way of working. Explain how the method helps me make that decision and which choices the worked case has already made.” Adapt the wording to the reader’s own question and work under F.19.
The selected examples should make two uses visible without implying that every question belongs to either displayed case:
- an ordinary entry shows how one direct pattern or one bounded direct route can answer a comparatively simple difficulty without a mantra; and
- a Practical-Use Card shows a recurring complex difficulty whose useful answer spans several direct pattern contributions and whose long dependency is easier to retain with a mantra.
The First useful result or blocker value gives the result of the action the entry supports, or a condition preventing that action. When several actions could be meant, name the relevant action in the value.
Use this ordinary-entry form:
When the product selects at least one card, place all selected cards under one group:
### Practical-Use Cards is a group inside the one Practical entries set, not another front door or selectable key. Omit the group when the product selects no cards.
The card's canonical structural field is Mantra; its value begins directly with the repeatable wording and needs no Local/Long prefix. A local reminder for one direct pattern or bounded result may remain in that pattern, an ordinary entry, or other teaching material, but it does not by itself select the richer card form. E.11 owns this content decision.
Every selectable ordinary entry and card shares one product-wide semantic-key namespace. The product declaration assigns each key one form, so a key cannot occur as both H3 and H4. An H5 expansion repeats its enclosing card key only to attach optional explanation; it is not another selectable occurrence.
A card has zero or one expansion. Its compact portion ends after Stop or return; the expansion ends at the next H4-or-higher heading or the end of the group. The expansion may explain branch choices, examples, or exact result support, but cannot contain another H4 card or H5 expansion.
The product-language application declares a deterministic measure of compactness and two maxima, one for the mantra and one for the complete compact card. Choose a count that makes sense for the language. For example, the English FPF profile allows 80 whitespace-separated tokens for the mantra and 220 for the compact card. At the same count, one version may assume a prerequisite its reader has never learned.
Use the limits to keep the short form compact. Judge the reader's effort or ability to recover a choice under the intended conditions. If a choice depends on a distinction the short form cannot carry, put it in the direct pattern or, when needed for the first choice, the same-key expansion.
Applying this grammar does not select a card, prove that the examples cover the product, or show that every cited pattern is needed in a particular case. E.11 defines the direct-entry/card comparison, cross-pattern mnemonic-gain test, and non-exhaustive discoverability purpose. The product-specific E.4 pattern declares its selected example keys, forms, compactness measure, and two limits. A validator consumes those values and checks structure; it does not decide content value.
Use First useful result or blocker for this field in new entries. In existing publications, First useful result or honest blocker identifies the same field, with the same value and position; including both labels in one entry repeats the field.
This profile keeps structural field keys in canonical English. A translation may translate surrounding prose and values and may add a human-readable gloss, but it does not silently replace or reorder the field keys. A translated structural-key profile needs a separately selected recovery and checking rule. Test translated and low-tool publications with actual readers and navigation tools rather than treating English parser success as accessibility evidence.
Keep support units and adjacent products distinct
A Readme, architectural reference, ToC, pattern-body collection, relation or edition note, and refresh route may be publication units of one framework product when they serve its declared use and edition boundary. Place each unit where its readers can find and use it. A separate file is a carrier choice.
Use E.4:4.1 to decide an adjacent product boundary from the useful result and its intended use. People may need to use, cite, or change a result as an identified whole, whether it appears inside a collection or through a separate carrier. State its identity, edition or current state, users, use, content boundary, and access route to the extent needed for that use. An established maintainer or maintenance commitment is not a prerequisite; record maintenance and refresh conditions when they are established and change reliance or use.
Examples include a source registry, MethodDescription collection, decision-support publication, inquiry evidence package, practitioner guide, pedagogical companion, catalogue, tool reference, access service, or inquiry programme. Their labels do not decide the boundary. Where an independent boundary is useful, point to the exact product edition or state. An annex may carry a declared snapshot or projection with a return to that product. Where ordinary framework use needs the material within the same boundary, include it as a named support publication unit.
One outer presentation carrier may expose several products. Each keeps its own identity, edition or state, form, access, and any separately established status or maintenance relation. Apply this profile to the FPF, DPF, or LPF constituents. A catalogue, evidence package, guide, service, programme, or other non-framework product uses the form selected for its own kind.
DRRs, build manifests, quality runs, digests, logs, and campaign state are process or maintainer evidence by default. A selected public use can justify a separate reader product or a public explanation drawn from that evidence. Publish durable use-changing reasons in the whole account or the relevant pattern; keep the dated decision, review, and release history in their records.
Check syntax and product truth at the right boundary
The deterministic part of the form check handles recoverable syntax and projection agreement:
- absence of raw HTML from the complete Markdown publication under E.8:4.1.1, while retaining literal code, mathematical notation and Markdown autolinks;
- the product-declared title and Readme H1, the compact opening, and absence of prohibited development or machine material from reader front matter;
- the required H1 sequence plus the product-declared body and reference tail;
- product-declared Readme and architectural-reference entries in the established ToC grammar, before the logical pattern index, with no generic rival mini-menu;
- one correctly addressed, content-free architectural-reference end heading at the boundary of that unit;
- authoritative index segments, aggregate current-pattern row/body bijection, distinct future-facing
PlannedCatalogEntryrows, duplicates, and reserved support-index grammar; - the Readme's one practical-entry set; the product's declaration of example keys and forms; exactly one H3 ordinary entry or H4 card per declared key; five ordered ordinary-entry fields; a non-empty card-group explanation; six ordered card fields; the shared compactness measure and mantra/card limits; and zero or one same-key H5 expansion with the declared boundary; and
- equality and source agreement of every optional public cue that is projected.
Check Markdown grouping with a focused source-hazard scan and a parser-backed render of the same candidate. Inspect the rendered heading outline and the block, list, table, code and link structure together. Reuse matching unchanged results. A clean mechanical result supports but does not replace the reader-visible judgement.
The product-specific check compares every visible cue with the exact edition or relation record from which it was projected and checks the product-specific body, reference tail, and any pinned compact-front shape. A syntax-valid but unresolved value fails there. A field absent from the public opening is not a form defect unless a selected reader use and product-specific rule require it.
A reader also follows the twelve content questions in section 4.7 to their public answers. A link or heading proves only that a target exists; reading establishes whether it answers the question at the claimed scale and preserves direct use. Reuse unchanged answers and current matching results. This content reading belongs to the selected authoring or publication question; the common form adds no separate release procedure.
Neither check decides framework scale from pattern count. Report pattern_count = 1 as a diagnostic. Use E.4, E.4.PFAD, E.4.DPF.DA, E.11, E.21, and the applicable subject patterns to judge whether the result is a usable pattern language for its declared field and first use.
Return the form result without overclaiming
Return the exact framework edition, edition-record source, carriers checked, form units found, public-cue agreement, logical-index result, practical-entry declaration and form result, product-specific tail checked, the public answers or unresolved content questions under section 4.7, and every mismatch or unresolved ref. Say separately whether the edition, carrier, publication occurrence, availability, currentness, or framework adequacy has an applicable result. Do not infer those claims or the truthfulness of card selection from form conformance.
Explain the Methods at each selected scale
Develop the practical way and its explanation for the receiving use through C.39:4.3–4.6. E.4.CM applies that general work to connected framework Methods. The publication functions below organize its public answers and add framework-specific form and placement; they are not a universal template for developing or describing every Method.
Use all twelve substantive E.8 functions as authoring questions for the whole framework and each substantive profile selected by its architecture. Answer them at the scope the publication promises. A framework may describe one composite Method through many patterns, several related Methods, or a repertoire used in different combinations. Where an exact MethodDescription claim matters, use A.3.1 to identify the admitted Method and A.3.2 to test the description episteme's MethodDescription membership. The number of patterns, mantras, files, or description media does not settle that identity.
The whole account connects the answers that individual pattern bodies supply. Write the shared answer once and give exact returns to inherited content. Determine those answers from the language's promised work before committing to an illustrative route.
E.8:4.2.1.3 governs the choice and continuation of examples: a connected account need not follow one project, and a successful route through one case does not establish the account's completeness. Keep explanations of significant alternatives and limits even when they require leaving that case.
At a narrower scope, state what changes in the situation, contribution, combination, evidence, result, or boundary. When an answer is missing, say which question remains open, which promised use it limits, and what remains usable.
Distinguish a shorter account of the same Method, an account of a constituent method and a more abstract model that admits additional ways to do the work. A profile may make any of these moves. Show what the practitioner can still decide and which relationships or distinctions the promised use requires.
Develop a missing instruction in the underlying account through E.8:4.2.1.1 before choosing how to present it. Then follow the promised use through the resulting publication, including any return to a supplying explanation. The reader must still obtain the input meanings, necessary operation, intermediate result and action-changing conditions that make the continuation possible.
A short entry may rely on its reachable developed destination; it must keep any condition needed before taking its first action. A full explanation elsewhere does not close a gap if the reader cannot find or use it under the declared access conditions.
Restore missing content at its source or repair its presentation and return, then revalidate the affected connected reading under F.19. A sound file structure or resolved link alone does not establish that the instruction survived.
When a condition governs a whole combination of Methods, give it one public statement at the scope where that combination is used. State the quantities, assumptions, or other conditions needed to apply it, and return to that statement from affected profiles and patterns. Pairwise relations remain useful, but a condition on the whole set may require a different decision. A change to that condition reopens the combinations that rely on it; unchanged local contributions remain available.
Also explain the relevant Method-composition structure and how its constituents are enacted in ongoing Work. A selected action can constitute part of a larger work that is being performed at that same moment. Give a representative example in which a condition of the whole changes the constituent action; show where a missing intermediate capability or incompatible resource demand prevents the combination. B.1.5.EW recovers that connection and B.1.5.RS examines a constituent replacement.
A DPF can describe only some of the needed constituent–whole connections: state the preparation, support and coordinating performance needed beyond its own descriptions. Reuse clear existing explanations and stop decomposition when further detail cannot change the use. Keep dependencies on earlier results where later work relies on them.
Publish the connected account as <framework name> Architectural Reference, the framework-level application of E.4's architectural-reference function. An all-in-one publication embeds the unit in its front matter; a multi-file framework gives it a discoverable place in that framework's assembly. Readme explains how to enter it. E.11.DSG governs the account and its useful question-led entries; this profile governs the framework's form.
Use Reference as this unit's address key, identifying its explanatory function. The qualified title, unit key, file and edition identity remain distinct; a preface with a different introductory function does not replace the architectural account.
At a wider language-assembly level, the corresponding Suite or FPF Library Architectural Reference explains the connections added there and returns to the framework's account. A separate source host, downloadable copy or site page is not another assembly level. Preserve enough rationale, source synthesis, alternatives, and worked detail for the intended reader to understand and adapt the language without its development intake or DRR. An exact inherited answer can satisfy a question; merely naming a pattern or listing source titles cannot supply missing explanation.
These are twelve content functions, not twelve compulsory new H1 sections. Each full pattern body retains its address and E.8 form. The outer Solution explains how to use the contributions together and returns to the bodies in the existing collection. It does not nest them under a new parent heading or require reading the whole account before a direct pattern use. If the public account uses the canonical function labels, prefer Architectural Rationale; Rationale remains the accepted alias defined by E.8.
A reader can start from the general treatment of Methods, enter the Method-development family, follow a specialization for constructing pattern languages, select a literature-based extension profile, and use a pattern for recovering one candidate Method. These steps can answer different questions. A publication can organize as many such entries as its readers need.
Explain the relation at each move: a profile is a bounded-use projection; specialization carries inherited content plus a delta; composition connects Method contributions; reuse makes one contribution available in several settings; a chapter or Part groups publication text. A substantive narrowing must pass E.8:4.1.3's same-situation usefulness comparison.
When two entries compare models of the same subject for the same inquiry, use B.5.MR. A stricter model has at least one admissible completion, preserves everything the earlier model required of its completions, and excludes at least one earlier possibility for that inquiry. Selecting another member of a family, restricting the audience of a profile or starting a different task can be useful while leaving that modelling question unchanged.
A pattern can participate in several profiles, and a profile can draw on several contributions. Describe this structure through the relations that hold.
For a mathematical view, select the elements and relation being modeled and use the applicable mathematics. Use C.29 only when a separate question about choosing, transferring, or relying on that mathematical representation remains.
A partial order needs its order properties; a lattice additionally needs the required bounds for every relevant pair. Multiple parentage or overlapping membership alone establishes neither. Keep that mathematical representation distinct from the Methods and their subject-side relations.
Make an architectural-reference section recognizable on its own
A reader entering through search, a quotation, a link or a retrieved excerpt may see a heading without its parents. Every heading that introduces architectural-reference content therefore carries the framework's public reference code, the publication-unit key Reference, and its complete ordinal section path:
For example, ## STR.Reference:1 - Problem frame - Direction and commitment under changing conditions identifies the first section of a Strategy architectural-reference unit whose key is Reference. #### FPF.Reference:16.8.2 - Production and situated use are different contributions identifies a nested section in the FPF Core architectural-reference unit.
Use the framework's declared public code; name the framework as well when quoting outside a context that identifies it. The enclosing H1 gives the qualified Architectural Reference title and has a product-declared ToC entry. Changing a former section key requires updating its maintained current returns; changing the title alone does not preserve a former URL fragment.
Close the complete architectural-reference unit with one content-free H2 heading, ## <FrameworkCode>.Reference:End, after its last paragraph, list or other content and before the next publication unit. For example, the Strategy architectural-reference unit ends with ## STR.Reference:End.
This visible boundary shows where the unit ends when it is copied or retrieved separately. It closes the whole unit, not each numbered subsection; it adds neither a content section nor a pattern-index entry. A heading shown inside a fenced example does not close the unit.
Number sibling sections in reading order, starting at 1, and carry the complete parent path into nested headings. Each nesting level adds one heading level and one ordinal. These ordinals locate sections in this publication unit; their titles state the content functions. An account can combine several E.8 functions in one section or explain one function across several sections. Keep that useful arrangement instead of adding twelve empty sections to match numbers. The rule introduces no limit on useful conceptual scales; physical Markdown heading depth remains a carrier constraint.
STR.Reference names a publication unit. Its section addresses are SectionRef uses under E.8, not declarations of additional patterns: the pattern index continues to contain the individually declared pattern bodies. Apply the same self-identifying construction to a profile account or other support unit when it needs its own section addresses, using its product-declared unit key. A profile explained inside the architectural reference retains that unit's section path; this text position does not decide the profile's semantic relations.
The prefix lets a reader distinguish a whole-language Problem frame from the Problem frame of one pattern before choosing what to read or cite. Visible names and numbers also survive copying and printing, where a hidden anchor cannot help.
The visible address and title are separated by an ASCII hyphen with one space on each side. Build each clickable fragment from the complete rendered heading according to the target Markdown carrier's rules, including punctuation removal and duplicate handling. When a heading changes, update its direct links in the publication, source templates and public consumers together. Check that the link resolves to the intended heading, then read that target for the answer the link promises. Keep the visible address usable for search and non-clickable copies. E.8:4.1.1 excludes raw HTML from this form, including compatibility anchors.
Make the language's long reasoning visible
Use a long mantra when readers need to keep a substantial connected account in attention across several questions, explanations or pattern contributions. Recover that account before compressing it. Identify what the reader is trying to understand or obtain, the questions that must be related, and why their answers belong together.
A result dependency is one possible connection; shared assumptions, jointly necessary conditions, alternative constructions, conflicting demands and returns can be equally important.
Recover missing reasons through B.5.RA and develop missing constructive joins through E.4.CM. The presence of one operational step per source topic does not preserve the explanation.
Explain how to assemble and revise the answers. Which partial answers can be used now, what remains unknown, and how do local answers constrain the whole? Which changes reopen which other answers? How can a reader choose the next inquiry or construction without mistaking the teaching order for a work sequence? Use the actual subject knowledge to answer these questions. A drawing of connected boxes or a sequence of PatternIDs cannot provide an absent inference, construction or applicability condition.
Keep the full explanation at the scale that owns its subject. E.4.CM:4.5 distinguishes completing an existing pattern, developing a separately useful recurring Method as a pattern, and explaining the material relations of several Methods in an Architectural Reference. The title or mnemonic form settles none of those choices.
A framework can have several long mantras for different questions; a repertoire need not acquire one artificial total sequence.
Where an existing explanation is sufficient, preserve it and repair its name or access instead of writing a shorter substitute. Where a source explanation enables readers to understand or adapt the contribution, retaining its thesis alone is insufficient.
Give the contribution a recognizable name, a stable existing kind of address, and both a compact reminder and a reachable developed account. Use F.18 when selecting a durable public name.
A mantra that recalls a separately applicable reasoning Method returns to that Method's ordinary pattern address and pattern-index entry. A reminder of the language's architecture can return to its Reference section. Establish that difference through E.4.CM:4.5 before choosing the address; completeness of a Reference explanation alone does not make it an adequate sole Method owner. No new markup or universal mantra record is required.
The compact expression may use questions, relations or actions. Choose it from the explanation rather than restricting the explanation to what fits one memorable sentence. Preserve the subject and scope across the two forms.
Make the account visible at each assembly level where it helps explain the whole. A Framework Reference explains the framework's connected reasoning and returns to its full pattern bodies. A Suite or Library Reference explains the connections introduced at that wider level and returns to the constituent accounts, without copying all their explanations.
In a README, show the most consequential reasoning connection in ordinary language, name the long mantra when one is useful, and give a direct return to its full account. This overview can be published without selecting a Practical-Use Card. E.11's comparison of entry forms decides whether a card helps a particular first-entry use; it does not decide whether the underlying subject reasoning is worth publishing.
Treat a compact order as a teaching unfolding of the connected reasoning. Explain material branches and returns, and use A.22.CGUS for the actual continuation and structure questions. A phrase order does not determine Work order or qualify a CGUS by itself. Different long mantras can constrain one another through a shared subject or premise: for example, a system construction changes the means of obtaining it, while a capacity limit can make that construction or its promised use worth reconsidering. State that specific connection; neither co-location in a book nor a universal instruction to read every mantra establishes it.
Check what the reader can reconstruct from the publication. Can the reader explain the needed grounds, recognize locally plausible but jointly incompatible answers, and determine which answers a consequential change reopens? Use materially different circumstances when they expose a hidden restriction.
These are questions about recoverable explanation and application; fluent recitation, one completed example and a mnemonic advantage do not answer them. Conversely, a source-grounded explanation can be useful without a measured recall advantage. Claims of improved learning, memory or field performance require their own evidence.
Archetypal Grounding
DPF with non-ascending pattern addresses. A Systems Engineering DPF edition orders SYSE.1, SYSE.16, SYSE.17, and SYSE.2 because that sequence helps readers. Its ToC rows and H2 bodies follow the same order. The § column reports each current position; it is not part of the PatternID. A later move changes the rows and bodies together without renumbering a continuing pattern. A citation outside the carrier says Systems Engineering DPF, SYSE.16; one intended to recover the earlier body also names the edition.
DPF, all-in-one and low-tool. A horticulture DPF is distributed as one Markdown file and a printed copy. Both open with the public framework name and Edition: Horticulture DPF 2.1; the Markdown line links to a public edition page and the print line gives the same public address. The ToC, practical entries, Horticulture Principle Framework Architectural Reference, four pattern bodies, coverage account, and refresh note follow. Authorship, source provenance, and change history remain reachable after the bodies. Readers can identify and return to the edition without crossing build records before their first working question.
FPF, split carriers. One website exposes an FPF edition through a front page and a separately downloadable Readme. The front page already identifies the edition, so its embedded Readme begins with practical entries. The standalone Readme repeats only the short edition line because it can circulate alone. Both return to the same public edition record; neither mints another edition or editable status copy.
LPF with a choice-relevant cue. An LPF supports two public language editions whose maintenance windows differ. The product-specific rule shows one short language-and-support cue after Edition because it changes which edition a practitioner should use. It does not copy the maintainer, build digest, source path, or complete dependency record into the opening.
Adjacent product. A separately maintained horticulture source registry has its own current state, users, selection rule, access route, and refresh commitment. The DPF points to that state; copying a snapshot into an annex does not create a second authoritative registry. One combined website may expose both, but the registry retains its catalogue form and receives no invented framework fields.
Whole Method Engineering account. An engineer extending a pattern language from a handbook needs both source recovery and a decision about what the new language will let practitioners do. The Method Engineering Principle Framework Architectural Reference explains the overall working question and returns to the source-recovery and comparison patterns. Architectural Rationale compares retaining a profile inside the language with giving it its own usable identity; the shared sources and trade-offs remain public. If the handbook cannot be inspected, the account names the source-dependent claim that remains unresolved and leaves unaffected direct uses available. A catalogue of source titles would not supply that missing basis.
Condition of a whole combination. Three proposed activities each need seven observer-hours that cannot be shared between activities. The available total is eighteen observer-hours. Any pair needs fourteen and fits; all three need twenty-one and exceed the total by three. The whole account states the resource condition once. The reader can select a feasible subset or obtain at least three additional observer-hours before committing to all three. Separate successful pairwise checks leave this whole-combination decision open.
Inquiry and direct use. Four pump inspections show three repeated orders and one reversal after a vibration cue. The public entry returns directly to A.3.1.MR, which recovers candidate Methods from Work evidence. Its worked case compares a fixed order with an exception against a cue-responsive order. In a fifth inspection, observe whether the technician changes order when the vibration cue is present. That observation can distinguish the accounts; until then, both remain candidates. The reader has a next investigative action while the larger inquiry remains open.
Shared profile. A Russian-language profile supports both a guide and a slidement. It inherits F.19 for recovering a sentence's claim, participants, and useful action, and adds Russian agreement checks. For example, English “the team are checking the result” can take plural agreement in British usage. Russian «группа проверяет результат» takes singular agreement when «группа» is the subject here. The named team performs the work in both versions. The translator can correct the verb without changing who acts. Read the repaired sentence under F.19; reuse that matching result when the unchanged sentence appears in both formats. The guide and slidement retain their own format and use conditions. The profile's public account explains this shared contribution and language-specific delta once, with returns from both uses.
Near miss. A relation table has rows whose first cells are PatternIDs and titles, followed by relation and source-return columns. It remains a relation table. A checker that calls it another pattern index from those cell values is guessing semantics from data shape and fails this profile.
Bias-Annotation
Scope: Limited to the public Markdown form of an FPF, DPF, or LPF edition and faithful low-tool projections of that form. It is not a universal publication template and does not prescribe the form of an adjacent guide, catalogue, service, programme, evidence package, or maintainer record.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Readers retain each product's compact familiar opening and find Readme and the architectural-reference unit in the ToC grammar already used by that product, before the one authoritative pattern index.
Inside the Readme they see one selected practical-entry set: ordinary examples show cheap direct use, while only cross-pattern cards selected by E.11's mnemonic-gain test add a visible mantra and an optional bounded expansion.
Optional public cues remain recoverable from one source when they change use, while development and rebuildability records stay out of reader front matter.
Builders gain checks that fail on missing public-unit entries, duplicate or cross-form keys, card grammar, structural, projection, and development-state drift without guessing table meaning, deciding card value or coverage, or inventing a rival navigation block.
The whole account also makes shared choices and source synthesis accessible beyond individual bodies. Authors can refine one profile without copying the entire language, and readers can see which answers remain inherited. Keeping those returns current costs work; concentrating the shared explanation reduces the number of independently editable copies.
Architectural Rationale
The shared rule fixes only the recognition points whose reuse pays across FPF, DPF, and LPF: a compact product-declared opening, Readme and the architectural-reference unit represented in the established ToC grammar, one logical pattern index, one selected practical-entry set with five-field ordinary examples and six-field selected cards, truthful product boundaries, and recognizable major units.
It leaves titles, optional public cues, the exact product-native ToC segment, example selection, the compactness measure and two limits, front line shape, and reference tails with the product-specific rule because their value depends on the reader's choice and publication language. These product-specific choices preserve direct entry while keeping deterministic checks narrow.
The twelve E.8 functions supply the content questions because a whole language, a substantive profile, and an individual pattern all need to explain working situations, Methods, evidence, use, and consequences. Reusing the functions makes omissions visible without inventing a new account for every product. Exact inherited answers and explicit deltas keep the explanation proportional to what changes.
The alternative of ad hoc architecture prose can preserve valuable reasons but offers no stable way to find a missing kind of answer. Literal recursive pattern formatting would add parent bodies and rigid heading depth that the subject's relations do not require. The selected form therefore reuses the substantive questions inside the established publication units and keeps the full patterns directly accessible.
SoTA-Echoing
Relations
- Specializes:
E.11for the common reader-facing form of one FPF, DPF, or LPF edition;E.11retains practical-use discoverability and first-result routing. - Uses:
E.8for the twelve substantive content functions, preferred Architectural Rationale title, intended-reader boundary, and usefulness of narrowing;A.3.2for an exact MethodDescription claim andC.29when choosing, transferring, or relying on a mathematical view raises a separate representation question. - Coordinates with:
E.4,E.4.FPF, andE.4.DPFfor framework/product boundary, product-specific publication units, optional reader cues, body order, and carrier assembly. - Coordinates with:
E.24.PUBfor publication occurrence, selected edition, form expression, carrier bearing, audience, bounded use, availability, and access; andE.17for bounded publication projections and source return. - Coordinates with:
E.4.PFRfor exact dependency and edition relations,G.11for currentness and refresh,E.4.DPF.DAandE.2.DAfor applicable package or whole-FPF adequacy, andE.21for pattern quality. - Does not replace: product-specific builders or validators, the edition record,
FPFEditionRebuildabilityRecord,FrameworkPackageManifest, an architecture decision, or a public product boundary.
E.11.PFP:End
Architectural Reference
Type: Specialization of E.11 (E) Status: Stable Normativity: Normative for an architectural reference and its public entries.
Problem frame
Use this when
Use this pattern when writing an architectural account for the users of a named method language: how it is assembled, why its parts and connections are selected, what they make possible and how to use them. The account can explain one framework, a DPF Suite or FPF Library.
Understanding the language's choices also lets a user question a premise or propose an improvement. Call the account <name of the explained whole> Architectural Reference. It carries the public architecture, shared explanation and useful question-led entries, with returns to the full Methods and evidence.
It is a support unit of the explained language unless independent use or change justifies a separate product boundary; this function does not by itself make the account another framework or the discipline being described.
First useful result. Give the reader a usable account of the whole's purpose, constituents, connections and limits, then a direct way into the needed explanation or Method. For a working question, give an answer or honest blocker. Distinguish an available result, a MethodDescription to use, direct-source evidence and a result still missing. Explain each necessary contribution and end that entry with an ordinary stop or return. Do not fill a missing result with another title.
Primary EntityOfConcern. One architectural-reference episteme: the public account of one named language assembly under C.2.1. Its content belongs to the explained whole and can be embedded in that whole's publication or separately carried. It can share the whole's publication and change boundary or have an independently useful product boundary under E.4:4.1. Separate files alone settle neither choice.
What this buys. The reader can understand the whole without its internal development records, select a useful route, recognize constituent and external contributions, and reach the full Methods, cases, limits and evidence needed to act or adapt the language.
Not this pattern when. Use E.4:4.2 and E.4.PFAD to decide which discipline a Suite describes and how its subdisciplines compose it. Use E.8 and the applicable Core or DPF authoring pattern to develop an unwritten Method, including a composite Method.
Use a known pattern directly when its contribution is already clear; the architectural reference is not a prerequisite for that use.
A separate curriculum or instructional Guide is a pedagogical companion. E.11.PFP governs framework publication form; it does not turn a supporting architectural reference into another framework. Describing FPF Library's connections does not claim a discipline above all its DPFs.
Problem
A list of titles does not explain the language's architecture or tell a practitioner which contributions answer a working question. A collection of short answers can omit the shared rationale, source synthesis and whole conditions needed for adaptation. Conversely, a long tutorial or metadata catalogue can hide the useful first answer, and an attractive summary can conceal a Method that has never been developed enough to reproduce.
These failures confuse different structures. A reader may treat all cited disciplines as constituents of the Suite, treat a MethodDescription as an already obtained result, or treat a cross-reference as proof of compatibility. A changed supplying Method can then leave the advice misleading even though every link still opens.
Forces
Solution
Develop the account of the named whole under section 4.1, preserving the explanations needed to understand and adapt it.
For a problem-led entry, write the practical answer first. State the recognizable situation and question, then say what each returned item is and what it contributes: an available result of its own kind and supplying product, a MethodDescription used to select or perform its Method, direct-source evidence for a named claim or decision, or a named unavailable result with its blocker and retry.
State maintenance only when it changes the answer. End the entry with the ordinary stop or return. Add identity, relation, evidence, warning, or reliance detail only when it changes the answer's truth, the reader's choice, or a named later use.
When an account cites a basis or directs readers to another contribution, use E.8:4.2.2.2 to make the purpose clear in ordinary prose. Preserve the difference among an explanation already given, an available result, material to apply, and a contribution still unavailable. This distinction does not make every relation between frameworks an application instruction or require a label beside every address.
Identify the explained whole and the account's boundary
Start from the language assembly selected under E.4 and E.4.PFAD. Identify what the whole enables and which constituents and connections it contains. For a Suite, recover its composite discipline under C.20, constituent DPFs, whole-forming connections and outside suppliers. For FPF Library, recover the relations among Core, Suites and standalone frameworks without inferring disciplinary parthood from their use or co-publication. A useful outside contribution remains outside unless its separate part relation is established.
Identify the architectural-reference episteme under C.2.1 by its claims, subject and effective ReferenceScheme. Its claims include the whole's purpose and architecture, intended readers and questions, answers and gaps, and sources and conditions needed to use them. Its scheme resolves the named whole, constituent and external frameworks, Methods, results and relation words. A title or a folder supplies a locator, not those distinctions.
Keep the architectural reference as a support unit of the explained whole when their readers, content change and publication boundary agree. Give it a separate product boundary only when readers need to use, cite or revise the account independently. If continuity across independently managed editions matters, E.4:4.2.5 supplies the product-series distinction. This publishing choice does not make the account a subdiscipline or a constituent DPF.
Establish publication, availability, currentness and maintenance only to the extent claimed. An author, locator or recent date does not establish a maintaining System or a commitment to future updates. A stale Reference can cease to supply a reliable route while the Suite and its constituent Methods remain directly usable.
Explain the architecture and its reasons at this level
Explain the working problems served by the whole, the contributions of its constituents and the connections that make their use coherent. State the shared assumptions, combination conditions and limitations that are not recoverable from one constituent alone. Include the alternatives and tradeoffs that help a reader choose or adapt the arrangement, the source synthesis and SoTA grounds behind it, and representative cases in which a whole condition changes the next action.
Use the substantive questions in E.11.PFP:4.7 for a framework account; at another assembly level retain the questions that its promised use needs. A short entry and exact constituent returns can spare a familiar reader repetition, but a locator does not supply an unwritten shared explanation.
Place the account at the assembly level of the whole it explains, then choose its carrier. In a one-file framework, embed that framework's architectural reference as a named unit. In a multi-file language, provide the account at that language's assembly level and make the constituent accounts reachable.
For example, an Engineering Suite Architectural Reference explains the Suite's composition; the FPF Library Architectural Reference explains its relations to other Suites and standalone DPFs.
Neither source-file boundaries nor generated website pages create additional explained wholes. Use the full qualified title at the unit's entrance and an unambiguous shorter mention within it; a generic heading Reference alone does not identify the subject.
Keep the account's referential index recoverable: who speaks, to whom, about which language, and for which reader action. In an Architectural Reference, its author explains the named language to its users. They need to understand how the arrangement works, what it enables, how to use its contributions and why it was chosen. Resolve wording such as 'we', 'you', 'here' and 'this framework' from that position. Identify the reading purpose, not a person's profession: a framework developer selecting a Method is a user in that passage; when the explained framework itself teaches framework development, its developers are its practitioner users.
Distinguish that explanation from instructions for constructing or revising a pattern language. This pattern and the applicable E.9, architecture and authoring patterns address that development work. A Reference can explain a constraint on valid use of its language; state whom it constrains, under which conditions and from which governing pattern it follows. A sentence about what the authors selected does not by itself impose an obligation on users or future authors.
Publish the architectural decisions realized by this language and the reasons needed to understand and use them. The architectural reference can be their public statement, not merely a summary of a private record.
Keep chronology, superseded decisions, review correspondence and release logs in the corresponding project's documentation. E.9 develops and records a substantive decision; a decision record is not historical by its kind. The public account retains the operative choice and its useful grounds.
An explanation of an already described Method can remain in its pattern or the architectural reference of the whole it explains. A separate course or Guide serves a learning purpose; it is not a substitute for the language's own explanation.
Make current decisions understandable and revisable
Organize the account around the architectural questions that matter to users of this language. A reader may need to understand a boundary before choosing a Method. The same explanation can let that reader challenge the boundary and propose a change; it need not teach the work of developing the next framework edition.
Give such a decision a recognizable title and section address. Connect related decisions: a choice of field boundary can determine which Methods are constituents, which are suppliers and what a whole-level Method still has to obtain. A list of isolated decisions cannot replace the account of that whole.
For a decision needed to understand, combine or revise the language, make these answers recoverable:
- Question and conditions. What difficulty or requirement does the choice address, and which competing forces shape it?
- Selected arrangement. What is constituted or separated, which relations hold, and where are they realized in the language? Distinguish the selected architecture from a proposed alternative or an illustrative case.
- Grounds and alternatives. Why is this arrangement useful under those conditions? Compare the serious alternatives that could change the choice, with their sources and limits. An unexplained preference or reference to an inaccessible DRR supplies no rationale.
- Consequences and revision. What can readers obtain with this arrangement, what does it cost or leave unresolved, and which Methods or connections depend on it? Explain what changed condition would justify reconsideration and what related claims would then need examination.
These are substantive questions, not four mandatory subheadings or a demand for one ADR file per decision. Use connected prose, a comparison table, a diagram or a developed case as the question requires.
Section titles should name the architectural issue or choice; E.11.PFP:4.7 supplies the framework-wide content functions, and section 4.2.1 supplies the addresses. Keep a shared premise once and return to it from the decisions that use it. The reader should not have to assemble the whole from repeated miniature forms.
A decision can define the architecture of this published language without imposing a universal rule on all disciplines or frameworks. State any normative requirement with its actual scope and return to its governing pattern. Conversely, an explanatory form does not make the language's declared composition an optional opinion. Claims about the discipline still need their C.20 grounds; publishing a decision does not create those subject-side relations.
Architectural patterns can contribute to this account or organize its explanation when they belong to the explained language and supply a result its users need. Develop a missing reusable construction as a complete E.8 pattern in Core, the relevant DPF or the Suite according to its subject.
An architectural-reference publication may carry that pattern if its full body, identity, reader use, entry and authoritative source remain explicit under E.11.PFP. Otherwise return to its existing full body.
Distinguish a domain architectural Method used by the language's practitioners from a Method for developing the language itself. Reusability alone does not place the latter in a user-facing account. Sharing a carrier does not turn ordinary explanatory sections into normative authoring Methods.
A record explaining why this language selects one architecture is not automatically a new pattern; E.9 and the applicable architecture and authoring patterns supply the development Methods.
First test the account for its user's work. From the account and its reachable suppliers, can the reader recognize when the language is useful, select and connect its contributions, and explain why a relevant architectural condition changes that use? If the reader can only repeat the arrangement or guess why it was chosen, develop the missing explanation.
An informed change proposal supplies an additional test of that understanding. A user who questions the arrangement should be able to identify the decision, the challenged premise, a meaningful alternative, its expected gain and the connections it could affect. This requires neither private authoring history nor a mandatory issue form. It does not replace the account's user-facing purpose with a Method for changing the framework.
Make the public minimum immediately useful
Make the account noticeable from the ordinary entrance to the explained language. A reader who does not know that an architectural explanation exists must be able to recognize its purpose before searching for its title. In a README or equivalent entrance, offer a prominent, ordinary-language question such as “How is this language organized and why?”, link directly to the account, and state what it explains.
Place that choice where readers first choose how to enter, alongside the route to a known Method. A link buried among filenames or offered only as an extra worked example does not communicate that purpose.
Test this entrance from the reader's starting situation, without supplying the Reference's name or destination as the answer. Can the reader notice that the explanation exists, tell what question it answers, and choose it when the arrangement is unclear? Can someone with a familiar sufficient Method still proceed directly? Successful filename search or link resolution answers a different question.
Make the facts needed to identify and use the architectural reference recoverable before the reader chooses an entry. An embedded account can use the containing publication's identifying information and ToC. A separately carried unit exposes the necessary information at its own entrance:
- the qualified architectural-reference title and a locator; identify an exact edition when the receiving use depends on it;
- intended readers and practical use, with the edition date when it changes identification or use;
- a status or warning when it changes the reader's continuation, with the grounds and as-of basis needed for any currentness claim;
- a direct entry to the account of the explained whole, with constituent and external contributions distinguished; and
- a table of contents that locates the architectural account, useful question-led entries and their supplying explanations without implying a mandatory order.
The edition date says when this edition was constituted. It is not a changing currentness claim or the date of every publication occurrence. Show the author when attribution, trust, contact, reliance, or source return changes what the reader should do. A byline does not identify the Reference maintainer, suite maintainer, publisher, or authority.
Every problem-led entry keeps this small visible core:
Add a DPF product series' or edition's state, field promise, detailed locator, applicability, evidence, availability, dependency, compatibility, warning, author, or claim-local reopen condition only when it changes the choice, truth, stop, return, or named reliance.
State a boundary shared by the entries once at Reference or section level. Do not repeat empty fields or turn ordinary architectural decisions into pattern-index entries. Use section addresses for the account. If the publication also carries declared E.8 patterns, retain their distinct entries and full bodies under E.11.PFP.
Frame each entry around a real working question and the decision or action the reader needs next. Let the answer branch, overlap, or offer several justified stopping points when the situation warrants them; do not force a false linear procedure. Keep the action-changing answer in the entry and return to detailed sources and explanations already provided in the account or its constituents instead of repeating them.
At Reference level, state whose information need is served, how the publication is presented and made available, how readers return to its sources, and where later review or retirement is decided. Future revision or continued availability requires a separate commitment.
Give each answer a stable address and a recoverable extent
Use the architectural-reference unit's declared key and give each substantive heading a stable section address, for example Engineering.Reference:2 and Engineering.Reference:2.1. These are SectionRef values for the unit, not PatternIDs or a framework identity.
Keep one qualified unit title, a contents entry for every substantive heading, and a final <ReferenceKey>:End boundary.
In an embedded account, use the containing publication's ToC rather than add another one. A structural Contents heading needs no section address. Use nested contents entries to make the answer hierarchy visible and every substantive heading reachable.
Make the heading levels express the actual containment of the answers. In a Markdown carrier, a complete addressed section includes its own text and child sections up to the next heading of the same or a higher level, or the document's end boundary. The numbers and heading hierarchy must agree; a number alone does not change the returned extent. Place examples and variations under the explanation whose answer they develop. An independently retrieved example returns to any parent or related explanation needed to interpret it.
Keep an established SectionRef when only the heading title changes. When moving a section beneath another parent requires a different hierarchical path, assign its current SectionRef and record the correspondence in the publication or access maintenance material.
Update maintained links to the current section. Preserve former returns in each access route that claims that continuity; do not reuse a former address for another answer or place a migration history in practitioner prose.
A new answer receives an unused address. Preserve returns to saved heading fragments using facilities permitted by the selected publication form.
For Markdown, follow E.8:4.1.1 and update maintained links to the corresponding native heading. Resolution of a carrier fragment does not establish that another access route can resolve the former SectionRef. If a route cannot preserve a return, identify the replacement and the affected reliance before claiming continuity.
Check the full contents and complete returned sections, including their children, in each selected presentation or access route. Successful title search establishes findability for that query; stable identity and complete extraction need their own evidence.
Keep lookup Work and the answer separate
A person, team, or assisting System may use one Reference edition while doing lookup Work. Ordinary use alone does not establish a Method, assignment, operation application, evidence, or authority. Identify those objects only when the claim needs their direct rules.
An ordinary answer may remain readable conversation. Persist one only when review, reuse, publication, or later reliance needs an addressable result.
For an answer that must be persisted, first identify the exact practical-question episteme Q. Then identify the answer episteme A under C.2.1 as <claim content = J_a, EntityOfConcern = Q, effective ReferenceScheme = R_a>.
J_a states the answer, exact Reference edition used, every returned resource or blocker, and what each does in this answer. R_a resolves those values and the use-specific relation words. This is an ordinary episteme, not a new lookup-result kind.
Say directly what each returned item does. For an available result, name the result's actual kind, supplying product and edition or current state, receiving use, and any currentness or availability condition that can change that use. State maintenance only when it changes the answer.
For a MethodDescription, name the described Method and how the reader uses the description; do not present its expected result as already obtained.
For direct-source evidence, name the supported claim or decision and the source limits. For an unavailable result, name the blocker and retry condition.
Recommendation, alternative, dependency, compatibility, and co-listing remain separate claims and create none of these stronger relations.
Test sufficiency and each item's necessity
Start from the result and range in the commission. For a request to find one usable combination, check the selected resources against that result and the sufficiency rule. If asked for the smallest combination or for all eligible alternatives, establish what makes a resource eligible. Track which eligible resources the search or a general bound covers. When the entry exposes only part of that range, a sufficient set found there can still be used. State the range within which any stronger conclusion follows.
To test one deletion, remove the item, keep the other resources and conditions fixed, and judge whether the promised result is still available. Report which deletions defeat this particular combination.
Call the combination minimal under inclusion if no proper subset suffices. Single deletions settle that question when sufficiency is monotone under adding eligible items. Otherwise a smaller subset may work even though every single deletion fails. An engagement can accept either a trio or a solo and decline a duo. To claim the smallest combination, give the size criterion and establish that no smaller in-scope combination, including an alternative outside this set, suffices.
Return the strongest supported conclusion. Established sufficiency remains usable when item necessity or minimal size is unestablished; when sufficiency itself remains open, name the missing premise and qualify the combination accordingly.
Distinguish resources included together from alternatives retained for later choice. When justified all-member inclusion needs a selector-facing JointUseSet declaration, use G.5-6 DeclareSetResult with the identified members, named use and current inclusion basis. The resources retain their own kinds; this declaration does not perform their selection.
Return to the full Methods and refresh the affected answers
The architectural reference supplies the current account of the explained whole, with exact returns to inherited explanations. For a Suite it distinguishes constituent DPFs from the external frameworks that a question happens to use. It also identifies the supplying edition, Method or result whenever that difference can change the answer. A reproducible comparison may pin an assembly and constituent-edition selection; ordinary current use needs a reliable current return, not a history of editorial decisions.
When a supplying DPF changes, compare the contribution consumed by the answer. Refresh the answer if the Method, conditions, result, evidence, applicability or availability changed materially. A new layout need not change the answer; a stable URL does not prove stable content. Use E.4.PFIP for preservation across reconstructed publications and E.4.PFR for a genuine edition dependency or compatibility claim.
When a DPF moves or a former pattern is divided, return to the full current Methods that supply the required results. Preserve a distinct result only where its explanation still exists. A list of successor titles, an empty redirect pattern or an old locator cannot stand in for lost content. Update maintained links and test their destinations and complete returned sections in the selected publication form.
If the field boundary or constituent relation changes, return that architecture question to E.4:4.2 and E.4.PFAD. Do not keep a false membership statement because an older Reference printed it. A Reference can show the newly supported arrangement without claiming that a publication edit itself transformed the discipline.
When an answer loses a required contribution, name the missing result and its consequence. Offer a direct known DPF route when sufficient, a bounded qualified alternative when supported, or an honest stop. An unavailable Reference does not prohibit direct DPF use. An unavailable constituent publication does limit the Suite's claimed usable coverage; neither fact erases the discipline.
Distinguish expression, derivative, edition, and product
Another layout, carrier, rendering, or faithful expression of the same exact claims under the same scheme presents the same Reference episteme. A translation or other derivative that changes claims or effective scheme is a distinct episteme with an exact source-to-use path under C.2.P; when meanings cross schemes, test the F.9 Bridge and bounded use separately. Title or provenance alone establishes no EpistemeEditionRelation.
A language-specific derivative stays within the same Reference product only while it uses the same intended readers and use, access rule, warning rule, later-review rule, and retirement rule. A separately established maintenance relation changes product identity only when the identity rule says so. If a language community needs a different current state or any of those rules, select another Reference product.
A comparison across Suites can belong to the FPF Library Architectural Reference. Give it an independent product identity only when its use and change boundary justify that choice under E.4:4.1.
Archetypal Grounding
Organization change and continuing operation
A manager asks how to reorganize a service while daily operation continues. Name the proposed change and who can keep accepting today's orders during each transfer of responsibility. If the sole authorized dispatcher has two incompatible mandatory duties at the same time, the proposed staffing cannot keep both covered. To promise continuity, change the timing or establish who else can take a duty. If the handovers and staffing cover the interval, the manager has a sufficient arrangement. A joint use on this project establishes no dependency between the two DPFs.
If the decision is safety-critical or legally constrained, add the exact source date, jurisdiction or applicability, authority boundary, warning, and reopen condition because those values change the answer and action. The simple and expanded answers use the same distinctions; they carry different justified detail.
External suppliers and a translated Reference
A Reference may combine results from narrative, language-practice and pedagogical DPFs for one lesson-planning question. It labels each as a constituent or an external supplier according to the actual field architecture; their co-use alone creates no Suite.
Explain how the selected resources together provide what the lesson planner needs. Keep that sufficient combination when the order asks for one that works. If the planner needs to omit resources, test the proposed omissions, using the criterion in §4.4 for a claim about every proper subset. Retain the alternatives still available.
A Spanish translation of an English Reference is a derivative episteme when its effective scheme changes. It remains in the same Reference product only while it uses the same reader and use, access rule, warning rule, later-review rule, and retirement rule. A separately established maintenance relation changes product identity only when the identity rule says so. A different Spanish use or current state may select another Reference product even when its title and cited DPFs remain recognizable.
Returns after content or publication changes
Understanding an arrangement before questioning it
A Library user needs to work without a network and considers proposing that each DPF include its own editable copy of all general Methods. The Library Architectural Reference explains two relevant decisions: general constructions have one full Core source because their conditions apply across subjects; a usable publication must also make its required texts obtainable.
Their consequence is not a requirement for network access. A self-contained distribution of identified editions can answer the access problem without creating separately maintained versions of the same Method.
A change to the general Method is a different proposal and must consider its other users. The account lets the user choose a sufficient offline arrangement and distinguish it from a proposal to change a general Method. It does not teach how to maintain the Library.
A user of a single-file semantic-integration framework wants repeatable queries and considers proposing a materialized graph for every integration. Its architectural reference explains the selected freedom of realization: a live provider can forbid copying, while another use may require repeatable local queries.
SIE.8 carries the full Method for selecting a realization under those conditions. The Reference explains why the language retains that choice and how it relates to identity, interfaces and validation.
An issue claiming that every integration must copy the sources can therefore be tested against a stated condition; an issue showing a new arrangement that meets both access and repeatability requirements can improve the Method. Reprinting SIE.8 in the Reference would add no missing architectural explanation.
In both cases the author first explains an arrangement that the user needs to apply. Understanding its reasons also makes a meaningful challenge possible. That additional result does not change the account into instructions to future framework developers. A project log or a table of current filenames supplies neither the practical explanation nor its reasons.
Bias-Annotation
- Catalogue bias. A longer list of included DPF product series looks more complete. Judge the entry by whether it returns the right contributions or blocker for the current question.
- Combination bias. Co-use looks like dependency or compatibility. State those relations only after their exact edition-level predicates pass.
- Freshness-display bias. A current-looking page or recent date looks maintained. Establish any maintenance or currentness claim from its direct grounds; retain the source return and status information needed for the receiving use.
- Precision-display bias. Repeated fields and formal identities look safer. Keep the ordinary answer first and add only detail that changes truth or action.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits. Readers can understand and adapt the language from its public architectural account or start with a short answer and reach the complete Methods and evidence. They can distinguish a Suite constituent from an outside supplier, a description from an available result, and a useful combination from a gap. A changed field architecture need not destroy valid cross-framework uses.
Costs. Someone must keep the answers connected to the contributions on which they rely whenever continued maintenance is promised. Consequential answers may need more source or applicability detail. Developing an unwritten Method costs more than adding a link, but the link alone would leave the practitioner to invent that Method.
Rationale
An architectural reference explains the structure and reasons of a method language. For a Suite, the discipline described and the account of its framework remain different subjects; for FPF Library, connections among languages need not be disciplinary part relations. These distinctions let the account explain outside contributions without absorbing their fields. Repairing or retiring the account does not by itself change the discipline's identity.
Progressive detail keeps ordinary use short without shortening the full explanation needed for adaptation. A trustworthy entry exposes its useful contribution and returns to the developed Method rather than making navigation compensate for missing content.
SoTA-Echoing
The composite discipline, Suite edition, constituent DPFs, external suppliers, Reference, answer, lookup Work, publication and availability remain governed by their direct FPF patterns. Compare their contributions to the actual reader question; identity evidence, source evidence and use evidence support different claims.
Relations
- Specializes:
E.11for the architectural reference of a named language assembly and its useful question-led entries; it does not specializeE.11.PFP. - Uses:
E.4:4.1for the architectural-reference function and product boundary;E.11.PFP:4.7for a framework account;E.4:4.2andE.4.PFADfor the Suite's composite-discipline and framework architecture;C.20for disciplinary composition;E.9for developing and recording the explained decisions;C.2.1for Reference and DPF editions; andG.5-6 DeclareSetResultwhen an already justified all-member combination needs aJointUseSetdeclaration. - Coordinates with:
E.4.PFRfor exact edition dependency and compatibility;C.2.PandF.9for derivatives and cross-scheme use;E.17,E.24.PUB, andG.11for source return, publication, availability, and currentness;E.11.PUAandE.11.PURfor actual selected-pattern use and pattern-use coordination. - Constrains: public architectural accounts of frameworks, Suites and FPF Library, their question-led entries and source returns, truthful treatment of external suppliers and gaps, and repairs after changes to Methods, language or field architecture, or publication.
E.11.DSG:End
Didactic Primacy & Cognitive Ergonomics
Status: Stable
Problem Frame
Use this pattern in either of two working situations:
- An assurance claim has no clear practical benefit. An author or reviewer can see the proof or assurance level but cannot say what it helps the intended user understand or decide. Apply the Rationale Mandate in §4.2: explain that benefit in ordinary language, or leave the claimed benefit unexplained pending a substantive answer. The assurance-level judgement retains its own B.3.3 profile.
- A reader gets stuck, completes a different task fluently, or encounters a costly demand. Compare the work the intended recipient brought with the task and help available in the passage. Use §4.3 to locate the lost choice, missing support or unnecessary demand and test a repair against the original use.
The first route asks why formal assurance helps. The second investigates the work a passage actually elicits, including a different choice made smoothly. A missing subject prerequisite or unsuitable assignment may call for a different intervention from a repair to the explanation.
Problem
Giving theoretical purity or formal completeness priority over cognitive ergonomics exposes the framework to two serious failures:
- Goodhart's Law: When a measure, such as an
AssuranceLevelwithin a declaredB.3.3profile, becomes the primary target, it can cease to indicate the understanding sought. Teams may optimize for the metric and produce assurance-bearing epistemes or publications that satisfy the visible criteria but remain conceptually shallow or pragmatically useless. - Cognitive Overload & Rejection: The framework becomes so dense, jargon-laden, and procedurally complex that its users burn out or switch to simpler, less rigorous methods. Certification procedures can consume effort without helping users understand or decide the matter at hand.
Forces
Solution
FPF makes Didactic Primacy (Pillar P-2) an architectural principle. The Rationale Mandate connects formal assurance to its cognitive benefit; the Human-Factor Loop investigates a concrete obstruction or doubtful demand.
The Principle of Didactic Primacy
The primary purpose of the FPF is to enhance the cognitive capabilities (qualified holder ability under A.2.2) of a reasoning system, team, organization, or other acting holon in service of its objectives.
Any architectural decision that increases formal rigour at the cost of clarity or usability must explicitly justify the gain in reasoning that warrants this cost. A checked bound can let an experienced engineer rule out an infeasible operating mode today. That decision with the proof available can be the entire benefit of the work. To claim that the engineer has acquired a lasting ability, compare relevant performances over time with the aids available on the intended job.
Mechanism 1: The Rationale Mandate
Every key assurance episteme or publication, such as an assurance case or proof publication, MUST contain a human-readable rationale component. Use B.3:4.2 to develop and inspect the assurance case's argument for the stated use; the rationale explains its practical and cognitive benefit.
- Nature: The
rationaleis a narrative explanation of the cognitive benefit. - Content: It MUST answer: "How does this level of formal assurance help the agent understand the problem better or make a more reliable decision?"
- Purpose: The author explains what the formal assurance helps the user understand or decide. An empty or perfunctory rationale indicates that the assurance work may be an exercise in formalism for its own sake.
Name the decision the assurance makes possible. To claim that the work now takes less effort, compare the work required at comparable correctness, including preparation or help that the job calls for.
The "So What?" Test for Managers
When your team presents a complex, formally checked episteme or publication, ask them to explain its benefit in the
rationale. For example, how does the formal work help you ship a better product, make a better investment, or avoid a critical risk? If the answer is not clear in therationale, the formal work's claimed practical benefit remains unexplained.Use this question to keep the team's formal work focused on the value it is meant to create. Any assurance-level claim needs its own applicable
B.3.3profile and result.
Mechanism 2: The Human-Factor Loop (HF-Loop)
Use this inquiry when the working result gives reason to examine what a passage or workflow asks of its intended reader. Difficulty can be visible as delay, an unsuccessful attempt, excessive work or a fluent answer to a different question. A workload report or a smooth performance can justify looking closer. Establish what the reader was actually doing before attributing a cause.
-
Recover the working question. Read the recipient's commission beside the task they actually received. Identify who had to decide or explain what, which objects and conditions mattered, and which choices the instructions or ordinary help already supplied. Follow the difference to its consequence for the intended work.
Preserve the relevant assurance and control conditions. If the suspected demand cannot change a worthwhile result, C.11.DUA can justify stopping the inquiry at that limit.
-
Inspect the initial attempt before explaining. Retain what the reader understood and did with the published passage, task inputs and ordinary permitted support. Include a successful answer to a different task when it helps locate the change of question.
Distinguish an implicit condition, a missing stated prerequisite and an unavoidable subject question.
Choose the inquirer for this question; profession, competence, assignment and amendment authority remain distinct.
-
Compare an alternative for the original use. Make a condition explicit, remove an unnecessary demand or restore the decision the reader was commissioned to make. Distinguish a change to the wording or help from a change to the work assigned.
Keep the commissioned task, ordinary prerequisites and assurance requirements constant when comparing expressions. If the previous assignment asked for a different result, record that difference and judge the repaired assignment against the original commission. Use F.19 on the connected passage.
Evaluate the attainable inquiry's burden under C.11.DUA.
-
Qualify the evidence. An expert walkthrough judges recovery from the public text; it is not an observed reader response.
Preserve the first attempt and the support available to the reader at the time. Unaided recognition requires an unaided first attempt. In a practice with ordinary help, observe what the reader can do with that help. A repeat after the reviewer's explanation shows performance under the changed conditions.
Psychometric or causal conclusions require their own selected domain methods and data.
-
Stop with the supported result. Return a justified wording/content repair, no change, or the exact evidence limit and any worthwhile next question.
The inquiry proposes; the applicable content/amendment authority decides. Use E.9 for a substantive FPF decision and its lighter wording route for ordinary editorial repair. Neither a role title nor a DRR implements or authorizes an amendment by itself.
A result-reuse passage examined at equal conditions
E.11.PUR §4.2.1 asks whether an earlier result answers the present concern, using the direct result pattern to compare its EntityOfConcern/edition, question/use, source/dependency conditions and qualification/currentness boundary.
The reuse instruction reads: “When those values still match, cite and use the earlier result.” Its following instructions retain any needed A.10 reliance and G.11 currentness account, and reopen the smallest affected result question when a value changes.
For the same task and prerequisites, compare that connected passage with this alternative sentence: “Cite and reuse the earlier result only when its subject and edition, question and use, relied sources and dependencies, and qualification/currentness still answer the present concern.” Keep the surrounding reliance and reopen rules in both versions. This makes the antecedent explicit without relaxing any assurance condition.
An expert walkthrough can recover all four comparison duties from the actual passage's preceding list; on that basis alone, no change is required. No reader observation or workload measurement is asserted by this example.
If an actual initial attempt instead misses the reuse condition, retain that response and examine whether the explicit alternative repairs that miss.
If the reader lacks an expressly required subject prerequisite, or the difficult step is deciding whether the source remains applicable, making the sentence shorter does not justify deleting that condition.
A second attempt after explaining the list can support only a qualified assisted-reading result. Stop at the demonstrated repair, no-change result or exact limit of that comparison.
Conformance Checklist
- CC-E12.1 (Rationale Mandate): Every key assurance episteme or publication, including an assurance case or proof publication, MUST contain a non-empty, human-readable
rationalethat satisfies the "So What?" test. AnyAssuranceLevelclaim separately cites its applicableB.3.3profile and shows that the target meets that profile's criteria. - CC-E12.2 (Bounded inquiry trigger): Examine a consequential mismatch between the commissioned work and the work the passage elicits, an obstruction or a doubtful work demand. Recover the task, passage, prerequisites and assurance the use requires.
- CC-E12.3 (Inquiry and stop): Apply §4.3 at a burden warranted by C.11.DUA. Preserve what the reader did before the reviewer's explanation and the help ordinarily available. Compare the candidate with the commissioned task and its intended conditions. State the repair, the grounds for retaining the passage or what the evidence leaves open. A reader's success supports conclusions about the work that reader actually did.
- CC-E12.4 (Didactic Primacy in DRRs): Any DRR proposing a change to a normative pattern MUST include a section analyzing its impact on cognitive ergonomics and didactic clarity.
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
The Rationale Mandate connects formal assurance to its cognitive purpose. The Human-Factor inquiry examines a consequential difference in the reader's work, including apparent success on an easier assignment. Its result may justify a repair, retention or further inquiry.
A purely conceptual workload warning would leave the reader without a next action; a stronger psychometric or causal method would require separately justified observations and design.
The bounded inquiry supplies an attainable examination and honest stop. It complements formal rigor with the Rationale Mandate and preserves independently justified controls. Together they connect formal assurance with what practitioners need to understand or decide for their work.
SoTA-Echoing — investigate the obstruction before prescribing simplification
Practice question. Does a reader's missed condition justify rewriting the passage, supplying a stated prerequisite, or retaining a necessary difficulty? The selected line examines one task and the initial attempt, compares a meaning-preserving alternative at the same prerequisites and control conditions, and stops with the supported result.
An expert heuristic review offers a less costly alternative. It can identify likely wording or presentation problems without recruiting a reader, but cannot establish what a reader actually recovered.
The GOV.UK Service Manual's moderated usability-testing guidance supplies the task-observation line: use believable tasks and neutral instructions that do not reveal the route to completion. Adapt that discipline to §4.3 steps 1–2 and 4: preserve the public passage, task inputs and initial attempt before explaining.
Its domain is service/interface research; it supplies neither a psychometric workload scale nor evidence that HF-Loop improves reasoning in every framework domain.
Moran and Gordon's heuristic-evaluation method, 2023 supplies the expert-review alternative and its stated limit. It scopes review to tasks and acknowledges that a heuristic violation need not be a defect in context.
Adopt that economical inquiry where expert recovery settles the wording question, including a justified no-change result. Reject promoting that judgment to observed reader success.
Use actual reading when the F.19 conditions require it, particularly when competing reconstructions or an observed miss can change the repair.
In §4.4 the expert can recover the four reuse duties from the preceding list, so the shorter explicit sentence is not automatically a better instruction.
If the intended reader's first attempt misses one duty, the task-bound comparison can distinguish a hidden antecedent from missing subject preparation.
At the same passage and task scope, expert review saves participant time; actual reading costs more but adds evidence the expert cannot supply.
Holding assurance fixed prevents an apparent improvement obtained merely by deleting the reuse condition. After an explanation, a successful second attempt remains assisted and familiar.
This comparison governs §4.3's initial-attempt, equal-condition, evidence-qualification and stopping steps, and §4.4's no-change/observed-miss branches. It selects evidence appropriate to the claim instead of a universal reader-test requirement.
Reopen when a new unassisted attempt contradicts the expert account, the task or prerequisites change, or a proposed method can answer the same consequential recovery question with lower justified burden.
Broader causal, psychometric or population-wide effectiveness claims require separately selected methods and data.
Relations
- Implements: Pillar
P-2 Didactic Primacy. - Uses: F.19 for connected meaning-preserving reading and evidence qualification; C.11.DUA for the value and full burden of demanded work or further inquiry.
- Complements:
E.13 Pragmatic Utility and Proxy-to-Value Alignmentkeeps visible measures, scores, review results, and release cues tied to intended value; this pattern focuses on the cognitive and working-reader usability of the framework. - Coordinates with:
E.9for the bounded content decisions that proposed HF-Loop improvements require. The DRR records the selected answer and rationale; the resulting pattern change carries the implementation. Ordinary editorial repair follows the lighter method identified byE.9.
E.12:End
Pragmatic Utility and Proxy-to-Value Alignment
Type: Part E FPF evaluation and repair pattern Status: Stable Normativity: Normative unless a section is marked informative
Use This When
Use this pattern when a project treats a visible measure, score, proxy, benchmark, dashboard, quality value, review result, release posture, or evidence volume as if it were the practical value or objective itself.
Look for situations such as these:
- a metric improves, but the team cannot say what intended value improved;
- a quality score, all-
5posture, assurance level, citation count, source count, or review pass becomes the target; - a proxy is used as a gate, incentive, resource-allocation signal, reputation signal, or release argument;
- a model, method, pattern, or system is formally better while users or operators are worse off, or safety, maintainability, learning, or decision quality worsens;
- an evaluation loop adds apparatus to satisfy the evaluator instead of improving the object of concern.
First useful move. Name the intended value or objective and the proxy or visible measure. State how the proxy is used now: measure, target, incentive, gate, release argument, decision driver, reputation signal, repair target, or orientation cue.
What goes wrong if missed. The team optimizes the proxy and loses the value. It can produce a better score, cleaner review proof, larger source packet, or more complete record while practical utility gets worse.
What this buys. FPF can keep measurement, evaluation, and quality loops useful without letting their visible outputs replace the value they were meant to serve.
Not this pattern when.
- If the question is whether a measurement scale is admissible, use
C.16. - If the question is ordinary pattern quality, use
E.21; useE.13only when a visible quality value is being treated as the practical value. - If the question is DRR adequacy, use
E.9.DA; useE.13only when DRR marks become a surrogate for decision usefulness. - If the question is whole-FPF Pillar adequacy, use
E.2.DA; useE.13only when Pillar values become the target. - If the question is assurance, gate passage, evidence sufficiency, or decision authority, use the governing pattern for that claim before treating the visible proxy as value.
Problem Frame
Practical work often needs visible measures. Teams use scores, dashboards, quality coordinates, tests, evidence counts, source freshness rows, release checks, and worked examples because practical value can be hard to observe and steer directly.
The danger starts when the visible measure becomes the object being optimized. A proxy can be useful as a signal and harmful as a target.
A pattern can become easier to defend while harder to use. A safety dashboard can look better while unmeasured hazards increase. A review result can look more complete while the decision it was meant to support becomes less decisive.
E.13 governs the proxy-to-value repair. It asks whether the visible measure still serves the intended value in the declared use, and what became worse when the measure improved.
Problem
Typical failures include:
- Measures replace objectives. Teams speak as if the score, metric, benchmark, assurance level, or all-
5posture is the value. - Evaluation loops become reward functions. A checking reader asks for improvement; the author adds fields, guards, source rows, proof sketches, and relation catalogues until the visible evaluation looks better.
- Unmeasured value is damaged. Usability, safety margin, maintainability, learning, domain fit, affordability, or operator action quality gets worse while the proxy improves.
- The proxy's use is unspecified. The same metric is treated as orientation cue, target, incentive, gate, and release proof without saying which use applies.
- The payoff has no grounds in the work. The account neither connects the proposal to the intended value under its stated conditions nor supports a claim that the value was realized.
Forces
Solution
Use ProxyToValueAlignment to recover the intended value, the proxy's current use, the value at risk, and the justified repair or stop. A short answer or local repair is enough when it settles the case.
When later review, reuse, or reliance needs a note, use the fields below that change the conclusion.
Record only the fields needed for the current conclusion, repair, stop, or later reliance. Use them to explain the relationship between the proxy and the intended value.
Name the Value Before the Proxy
Name the intended value, objective, or practical payoff in terms of the work it is supposed to improve. If only the proxy can be named, report that the project has a measure. Its connection to the intended value has yet to be shown.
Type the Proxy Use
A proxy can be harmless as an orientation cue and dangerous as a target. State how the proxy is being used.
Ask What Got Worse
Whenever a proxy improves under optimization pressure, ask what became worse or more fragile.
Check at least usability, affordability, safety or harm boundary, maintainability, domain fit, source preservation, decision quality, learning, and neighboring-pattern fit when they are live in the case.
Report whose decisions, protected qualities and interactions the comparison covers. An effect on each part, a dependence between parts and an effect of their organization are different claims. For example, if two shifts need one specialist at the same time, their joint outcome depends on how that time is allocated. A claim about the improvement of one shift can be settled with the other shift's outcome still open.
Require a Minimally Viable Value Slice
A minimally viable value slice gives the recipient enough grounds to judge the practical value claimed. For an expected consequence, relate the proposed change to the recipient's work and establish the conditions under which the consequence follows. A general argument, a worked comparison or an available observation may supply the needed grounds. For a claim about actual improvement, examine use and the comparison needed to attribute the change.
Identify the grounds the value slice supplies, whose outcome it concerns and which variations its reasoning covers. A consequence of a binding obligation and an observed change in someone’s work support different claims.
A short account is sufficient when the recipient can recover the connection over the range in question. If a measure concerns a proxy, explain why it bears on the intended value.
A worked illustration can explain how a proposed repair would serve that value; it does not by itself establish an observed gain. Keep any claim about actual improvement within the comparison and evidence that support it.
A wrong dispatch costs 20 minutes and an extra check costs one minute per dispatch. If the check reduces the probability of a wrong dispatch by at least 0.1 in the intended work, the expected net saving is at least one minute per dispatch, with other costs unchanged. Describing how the check might catch an error identifies the mechanism. For the saving, establish the probability difference or a sufficient bound. A decline in recorded errors can also reflect changes in staffing, reporting or the jobs accepted.
Use partial evidence when it already settles the decision. For a course of twelve people, five verified successes, three verified failures and four unknown outcomes give five to nine successes. An at-least-40% target is met and an at-least-80% target is missed. A 60% target remains open. Reports of ease from the same respondents answer a question about their experience. A rate for future courses needs a basis covering those participants and conditions.
Repair by Value Movement
When the proxy has displaced the value, choose one or more of these repairs:
- where the governing rule permits it, change the proxy use from target/gate/incentive to orientation or bounded measure. If a binding requirement must change, propose the amendment under its authority and keep the current requirement in force until that amendment takes effect;
- add a protected quality or counter-metric that names the value at risk;
- change the work or design where the proxy can improve while the intended value deteriorates, and assess what the change does to that value.
- split the claim: one measure report, one value claim, one assurance or gate claim if needed;
- suspend an unsupported claim, or state the consequence and conditions the available grounds already establish.
Archetypal Grounding
Bias-Annotation
E.13 rules out two substitutions. In proxy-for-value bias, the visible measure, score, evidence volume, review result, release posture, or dashboard state is treated as the practical value itself.
In evaluator-satisfaction bias, a team adds apparatus to satisfy an evaluation signal while the governed object, user work, safety, maintainability, or decision quality does not improve.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
- FPF can use scores and metrics without making them the object of optimization.
- Improvement loops gain a simple value-proxy stop condition.
- An expected consequence can follow from the structure of the work. A claim about achieved improvement needs evidence of the work and an appropriate comparison.
- Some attractive proxy improvements are rejected, split, or lowered.
- The cost is a small proxy-to-value check whenever a visible measure becomes a target, incentive, gate, release argument, or repair target.
Rationale
FPF needs measurement, evaluation, assurance, and release checks as instruments for pursuing practical value. E.13 keeps the visible instrument attached to the intended value and asks whether the value survives optimization pressure.
To repair a Goodhart-style failure, restore the relation among the intended value, the proxy, its use, the protected qualities, and a small case that shows the value. Adding another large audit cannot by itself restore that relation.
SoTA-Echoing
Relations
- Implements:
E.2PillarP7 Pragmatic Utility. - Complements:
E.12for cognitive ergonomics andE.14for human-facing working models. - Coordinates with:
E.8for authoring practical-payoff claims,E.19for review/admission proxy-to-value checks,E.22/E.23for improvement framing and repeated improvement loops,C.16for measurement admissibility,C.25for engineering quality-family endpoints,E.21for pattern quality,E.9.DAfor DRR adequacy,E.2.DAfor whole-FPF Pillar adequacy,B.3for assurance,A.21for gate passage,C.11for decisions, andA.10for evidence. - Used by: improvement loops, release checks, pattern reviews, dashboards, metric-driven work, AI reward or judge-score cases, and any project where visible performance may displace intended value.
E.13:End
Human-Centric Working-Model: Readable Claims and Recoverable Assurance
Status: Stable Type: Pattern
Use This When
Use this pattern when FPF text needs to stay readable as one human working model while heavier mapping, logical, constructive, or empirical assurance remains recoverable underneath it.
What goes wrong if missed. The working text can slip into local jargon and slash labels, or expose so much proof machinery that practitioners cannot use it in ordinary design, review, or management work.
What this buys. A working reader sees one small model first, while assurance readers can still recover mapping, logical, constructive, and empirical grounding without forcing that machinery back into the Working-Model vocabulary.
Ordinary route. Name the claim, its subject and what the recipient came to learn, decide or do. A short statement can close a request to identify the claim. An explanation may also need the relation that makes it follow. Someone designing a way to choose needs grounds they can use when the choices change. Supply what the commissioned work requires and add the assurance route when warranted.
When an assurance-bearing reliance question is current, place only the needed Mapping, Logical, Constructive, or Empirical support underneath that sentence and cite the pattern that defines or tests each supporting claim.
Not this pattern when. Do not use E.14 to decide whether a relation obtains, Work occurred, a result was constituted, evidence or assurance passes, a method applies, work is ready, a gate passed, or permission is current. Use the pattern that defines or tests that claim.
Use E.14 for the human-first publication order and the recoverability of support; it supplies none of those domain results.
Intent
Establish a single, human-centric Working-Model that practitioners can read, discuss, and evolve without first reading the full assurance account. Publication alone does not require an assurance field on a direct Working-Model statement.
When a publication elects B.3.5 or another named current assurance requirement applies, declare its posture and supply the support its branch requires.
Under B.3.5, covered claims declare validationMode: structural claims carry their required construction account; a permitted postulate links an admissible evidence object and states scope and challenge cues; an inferential claim links a followable argument.
These supports justify or challenge the claim without creating its subject or relation. Recover the Work, result, provenance and currentness facts consumed by the reliance, following §5.5. Assurance remains beneath the readable claim.
Problem Frame
Teams need one shared Working-Model to make timely decisions. Historically, teams have faced two failures:
- Jargon. Teams use different terms for one shared Working-Model value, slash labels, and partly overlapping terminology.
- Excessive formality. The model becomes too formal for day-to-day design and review.
Both failure modes create friction between two audiences: (1) working users (engineers, programme managers, policy owners) who need a small, stable Working-Model text, and (2) assurance authors (ontologists, methodologists, auditors) who need to inspect the basis and limits of the Working-Model claims.
Give the two audiences a shared working text and access to its assurance:
- A Working-Model layer: curated kinds and relations expressed in plain terms, with simple human rules for using them.
- An Assurance stack beneath it — Mapping, Logical, Constructive, and Empirical when needed — that keeps arguments and accounts recoverable without crowding the Working-Model narrative.
This arrangement supports the framework's unification principle: keep the Working-Model small and its foundations rigorous. Constructional mereology uses sum, set, and slice to make independently grounded assembly, collection, and aspect facts inspectable. Those forms do not create a relation occurrence or decide whole identity. The Kernel stays minimal and meta-only.
Problem
A reader may need to decide, design, review, or coordinate with FPF terms before they are ready to inspect mapping tables, constructive traces, evidence records, or proof arguments. If the working text exposes all of that machinery first, the model becomes unusable; if it hides the machinery completely, the model becomes arbitrary.
Publish one human-facing Working-Model with its assurance available underneath.
Forces
-
Cognitive economy vs. semantic precision. Managers and engineers must navigate with a handful of names and relations; assurance authors must still check that each name has one intended model value, each relation claim has the required world-side basis, and identity conditions are explicit.
-
Speed of change vs. guarantees. The Working-Model must accommodate revision; support must be updated or its limits exposed before a use that depends on it.
-
Parsimony vs. expressivity. The Working‑Model should not proliferate relation types or ad‑hoc categories; fine‑grained distinctions live in the Assurance layers and are shown only when they materially change a decision.
-
Downward grounding vs. upward contamination. Present the readable claim before its needed Mapping, Logical, Constructive, or Empirical support. Assurance notation and record structure do not dictate the Working-Model's wording or layout. If support defeats or narrows the claim, revise that claim under its direct governing rule and make the change visible.
-
Trans‑disciplinary unification vs. local dialects. The Working‑Model must reconcile different disciplines’ habits without erasing them; Mapping captures dialects, while the Working‑Model exposes a single usable choice.
-
Auditability vs. readability. Any Working-Model statement can be audited on request under the pattern that defines or tests its direct claim and any assurance profile selected for the current use; day-to-day views hide the scaffolding unless summoned.
Solution
Human-Centric principles
Recognition text and assurance text
Keep the target assertion, its EntityOfConcern and its scope recoverable in both recognition and assurance. A supporting premise may concern a part, another kind of subject or another model. Show which relation lets the reader use that premise for the original assertion and under which conditions it holds.
If the pattern distinguishes an EntityOfConcern, the interpretive or operational move applied to that object, and the wider review or work process around it, the text should distinguish them explicitly instead of changing nouns as though the object, move and work were interchangeable.
Begin the Working-Model text in the subject domain. If a pattern is meant to help with a review, design, cultural, research, or operational problem, the recognition text should open from that problem-owning moment before internal taxonomy or package architecture.
If a broader umbrella and a narrower working branch are both live, say plainly what each names, what object is being discussed, what move the reader makes, and what wider work remains outside.
Recognition text and assurance text name two reading-order text blocks inside one pattern.
For human-facing canonical patterns, Working-Model-first discipline should appear in a two-part reading order. The recognition text is the working text that a cold practitioner, manager, or researcher should be able to understand first: what situation this pattern is for, what it buys, what it is not for, and what ordinary mistake it helps prevent. The assurance text is the heavier text that carries declaration, object discipline, modeling lens, law, return conditions, and other assurance work.
The assurance text may justify, tighten, or audit the working text, but it must not silently replace or strengthen the recognition-text claim.
Keep the target assertion, its subject and its scope visible through both blocks. Supporting premises may concern other subjects. For Pump-37, calibration of the instrument and the measurement Work can support the vibration reading at 09:00. That reading completes an order for the measurement. A service decision also needs the applicable criterion and operating conditions. A conclusion about a fleet needs grounds covering the relevant pumps. Where episteme-publication-heavy or transform-heavy patterns need a compact ontological account, the assurance text should expose three things explicitly:
- the ontic target or EntityOfConcern;
- the modeling substrate or mathematical lens when one is load-bearing;
- the publication face or working text by which the claim is presented.
Present the Working-Model text first. Readers can consult the assurance text underneath when they need to inspect a claim's grounds. That text remains available for audit.
When empirical evaluation is current, put the ordinary subject claim first. Keep a proposed evaluation distinct from completed performance; use U.WorkPlan when the receiving question requires coordination of future Work, and retain the actual Method and any relied-on MethodDescription edition.
When a claim asserts or relies on a dated evaluation Work, recover its required performer basis under A.13 and independently admit the Work under A.15.1.
Add F.6 when precise assignment-bound attribution is consumed, keeping the assignment and any local system-role-kind classification independently grounded.
An ordinary result may omit identifiers it does not consume while keeping all facts needed for its use recoverable.
E.14-P.1 – Working-Model first, assurance when current. State the direct claim in one Working-Model for human-facing discussion. The reader can stop when that claim and its ordinary explanation answer the commissioned question. Supply additional grounds when the reader needs them to understand, choose or act. Add the formal assurance called for by the publication or current requirement.
When assurance is current, declare only the posture and forms of support required by the applicable pattern: Mapping to align a term with the chosen model value it names; Logical to state label meaning, scope, constraints, and limits; Constructive to make independently grounded construction facts inspectable; or Empirical Validation to support a bounded reliance on a domain result.
Under
B.3.5, covered claims declarevalidationMode.For each selected shoulder, name only the objects, scope, and qualification window the current use consumes. None creates the model value, subject relation, Work occurrence, or result it supports.
E.14‑P.2 – Downward‑only dependency. Keep assurance apparatus subordinate to the readable Working-Model claim: its notation and record structure do not define the public vocabulary.
This presentation rule does not prevent correction. If a proof, trace, or empirical result exposes a false or overbroad claim, revise the Working-Model claim and any affected wording under the direct governing pattern; do not silently leave the recognition text asserting the defeated claim.
E.14‑P.3 – Small working text, big proof. The Working-Model exposes a small set of names chosen through E.10's Plain and Tech registers for the intended reader, and a compact family of relations used in everyday reasoning; the assurance text makes their meanings, basis, limits, and support inspectable below.
E.14‑P.4 – Human registers first. Terms in the Working‑Model are chosen for human legibility (register‑badged, synonym‑aware). Synonym capture and language variance belong to Mapping; only the chosen canonical label appears in the Working-Model text.
E.14-P.5 – Required assurance postures are explicit. A Working-Model relation covered by an elected
B.3.5profile declaresvalidationMode ∈ {axiomatic, inferential, postulate}. Another named current assurance requirement may require its own declared posture. A direct relation outside such a profile needs no E.14 assurance field. axiomatic means that the author relies on the linked construction account required by the elected branch; inferential means that the author relies on a followable reasoned chain.Under
B.3.5, a permitted postulate is a scoped pragmatic claim linked to an admissible evidence object, with empirical challenge cues.Follow another elected profile's own requirements. Empirical Validation may accompany any permitted posture when observation supplies the needed support; a posture declaration does not itself establish performed evaluation.
E.14‑P.6 – Parsimony in the working text. No new Working‑Model relation types are introduced if the existing Logical label-meaning rules plus Constructive grounding suffice to capture the intended meaning.
E.14‑P.7 – A postulate is not completed evaluation. Under an elected
B.3.5postulate branch, authors SHALL state the claim and scope, link the admissible evidence object, and give the profile's empirical challenge cues.That requirement does not by itself assert a new evaluation Work. When reliance consumes an actual evaluation or measurement result, keep its necessary subject, scope, result and source grounds recoverable.
Identify dated Work, performers, Method, bindings, result episteme and qualification window where the claim or applicable profile depends on them.
Use A.10 when the source-to-use relationship must be made explicit or replayable, retaining already sufficient direct support; use F.6 for consumed assignment-bound attribution and B.3 for a current assurance claim. Each remains governed by its own conditions.
E.14‑P.8 – Working-model-first is not explanation-thin. Human-facing parsimony does not license under-explained pattern prose. When a pattern claims a Working‑Model benefit, it SHALL still provide enough problem framing, rationale, and worked slices that readers can tell what the model clarifies, what remains on the assurance shoulders, and when a heavier review path is required.
Layer Standard & Downward Flow (Working‑Model → Assurance)
Each layer has a purpose and a guarantee when selected. Start from the Working-Model statement and attach the support its use requires.
Working‑Model (what humans see)
Purpose. A small, curated graph of kinds and relations that a mixed team can read at a glance.
Elements.
- Kinds — one chosen concept per node (no slash‑labels).
- Relations — a short set of statements intelligible to non-specialists (for example, Component-of, a subject-specific sentence such as “this cartridge belongs to this bank under the bank's rule”, Aspect-of, and a small number of cross-disciplinary ties such as Interface-of or Constituent-of).
- Language registers — choose Plain or Tech wording under E.10 for what the intended reader must identify and use. Keep background synonyms, source labels, and unused symbols in Mapping; expose a technical term in the Working-Model when the current claim or reader use needs it.
Obligations.
- A Working-Model edge or node whose use elects an assurance profile keeps that profile's required support recoverable downward. A direct claim outside such a profile can stand on its direct meaning and truth conditions; E.14 adds no assurance field or separate support account.
- The Working-Model omits constructor jargon, proof terminology, and evidence identifiers that its intended reader does not need. Keep the support available on demand, and include a technical term or identifier in the working statement when it changes the claim or the reader's action.
Assurance-1: Mapping (from words to chosen model values)
Purpose. Consolidate human labels from varied sources and bind them to the chosen model values used in the Working-Model, including admitted U-kinds where kindhood is live.
What to check. When Mapping assurance is selected, verify the Working-Model label's alignment to its chosen model value in the current scope. Recover the synonyms, abbreviations, locales, and registers needed for that comparison in the Mapping account.
This can contribute to Concept-Bridge Assurance (CBA) under B.3.3 when it establishes a claim-relevant meaning correspondence; collecting labels alone does not raise assurance.
Deliverable. When the current use needs source-word alignment, provide a compact alignment account for that scope. Identify the one label used in the Working-Model and the background labels retained as source wording.
Teams can keep their local terms. Mapping shows how those terms correspond to the shared Working-Model vocabulary.
Assurance‑2: Logical (from Working‑Model relations to label semantics)
Purpose. Give each Working-Model relation one precise intended meaning and its admissible use cases, keeping the Working-Model vocabulary small.
Guarantee. When Logical assurance is selected, a Working-Model edge such as Component-of or Aspect-of carries one stated reading, including the scope and relation properties needed for the current use, so an auditor can assess whether that use is legitimate.
Deliverable. When the current use needs an explicit label-meaning account, give a short rule such as: “When an edge is labeled Component-of in the Working-Model text, it intends the direct structural reading whose participants, relation occurrence, construction rule, and identity conditions must be recovered before the assertion is accepted.”
The Logical shoulder ties the human label to that accepted meaning; it does not make the relation obtain.
When a calculus or other expression carries a reasoning or construction step, explain its interpretation, operation and prerequisites or cite their subject guidance under E.5.2. Keep the readable claim recoverable from that expression.
A precise meaning for each label lets the Working-Model stay small without introducing a new relation type for every wording difference.
Assurance-3: Constructive (from a structural claim to its inspectable construction account)
Purpose. Make the construction basis of a published structural claim inspectable without turning the assurance account into the relation or the whole.
Guarantee. When Constructive assurance is selected, one truthful construction trace names the exact whole, collection, or aspect; its participants; the direct relation occurrences that obtain; the applicable assembly, collection, or facet rule; and the direct identity or reidentification conditions.
The same inputs under another assembly may form another whole, while a permitted constituent replacement may preserve the same whole. The trace decides neither case.
Deliverable. For a structural parthood or collection-belonging assertion covered by an elected B.3.5 profile, keep the readable claim first, link it through tv:groundedBy to one current C.2.1 construction-trace episteme in the applicable C.13 form — sum or slice for structural parthood, or set for collection belonging — and declare validationMode=axiomatic.
If another named current assurance requirement calls for a construction account, follow that requirement and use C.13 for the trace content.
Outside those conditions, the direct structural claim has no E.14 mode, link, or trace obligation.
Creating, revising, publishing, or losing a trace changes the account or its availability, not the relation occurrence or whole identity. The trace edition, its warrants and evidence, and the temporal status of the described direct facts retain their own currentness.
(Rationale: constructive assurance makes the facts and identity tests behind ordinary part-whole talk inspectable; it does not substitute an author narrative for those facts.)
Assurance-4: Empirical Validation (from claims to observed world)
Purpose. Make the empirical basis and bounded admissible use of one Working-Model claim inspectable without turning evidence, provenance, or an assurance record into the subject result.
Guarantee. Supply the empirical support required for the target claim and receiving use. A permitted B.3.5 postulate states its scope, links an admissible evidence object and gives challenge cues; an ungrounded proposal does not satisfy that elected profile.
For a relied-on domain result, recover its interpretation, applicability and source grounds under the pattern that defines or tests it. Keep any required scope selectors and temporal qualification; do not add a time window to a non-temporal claim merely for this account.
Where the result or profile consumes a dated Work claim, establish its performer basis under A.13, independently admit the Work under A.15.1 and retain its enacted Method. F.6 supplies exact assignment-bound attribution when consumed; A.2.1 governs the assignment.
Keep any relied-on MethodDescription edition, participant or A.6.1 binding, local classification and C.2.1 result-episteme identity under their own rules.
Use A.10 for a source-to-use account that needs to be explicit or replayable, and B.3 for an assurance claim. Already sufficient grounds remain usable; an unresolved required premise remains open.
Deliverable. Keep the ordinary Working-Model sentence first and expose only the support required for its stated use. A postulate follows the elected profile, including B.3.5's evidence object and challenge cues when that profile applies; no new evaluation is required merely to declare the posture.
A proposed evaluation remains prospective; use U.WorkPlan when coordination of future Work is needed. If the current question concerns the Work's productive treatment, an episteme's identity inception or a production-completion condition, use A.15.PROD for that question.
Re-evaluate only reliance affected by a changed source, method, calibration, result, temporal qualification, provenance or assurance premise. Narrow or withdraw that reliance when its applicable condition fails; age alone does not defeat still-applicable evidence.
Under B.3/B.3.3, Empirical Validation contributes what its result establishes for the target claim and receiving use. A reliability calculation needs its exact domain quantity and model; scope follows A.2.6 and its warrant.
Purpose-selected support for a single Working-Model statement
Start with the direct Working-Model arrow A –Component-of→ B. It can settle which whole a specified part belongs to. To explain how an arrangement works, connect that membership to the relationships on which the result depends. When a profile or current requirement calls for assurance, add its prescribed support:
- Mapping, when source-word alignment matters, shows that A and B are the chosen labels for their model values and records background labels without making them Working-Model names.
- Logical, when the relation reading needs assurance, states what Component-of means here and the boundaries of that use.
- Constructive, when
B.3.5is elected for this structural assertion, links the readable claim to one current C.2.1 trace episteme that reports the participants, direct relation occurrences, construction rule, and identity conditions in a C.13sumform; the author declaresvalidationMode=axiomatic. The direct relation and identity tests remain decisive. - Empirical Validation, when the current reliance needs observation, exposes the empirical claim, supported domain result and scope, together with the Work, result-episteme identity, bindings, temporal qualification and source-to-use account that the claim or elected profile consumes under §5.5. A separately current B.3 assurance claim retains its own grounds.
The selected support stays below the readable claim. It makes the needed basis inspectable without forcing unused assurance machinery into the Working-Model.
Archetypal Grounding (System and Episteme cases)
U.System — Working‑Model first, Constructive grounding available
-
Publication (Working‑Model). Authors state structure using familiar relations (e.g., Impeller ut:ComponentOf Pump; Pump ut:ComponentOf Skid).
For the installed skid, establish the impeller's place in this pump and this pump's place in this skid. A drawing may prescribe these connections for every conforming assembly. To use that drawing for a claim about the delivered skid, connect it to this configuration and its actual assembly. The drawing itself can settle what the design prescribes.
-
Assurance (downward grounding). When the publication elects
B.3.5, first recover the exact skid, parts, direct fastening, coupling, enclosure, terminal, flange, and seal occurrences, the applicable skid assembly rule, and the skid reidentification rule.Then link the readable claim to one current C.2.1
sumtrace that reports those facts and declarevalidationMode=axiomatic.If another named current requirement calls for a construction account, follow its stated obligations instead of borrowing
B.3.5automatically. The account remains below the Working-Model; order and time stay in their own relation families. -
Reading result. Readers see the Working-Model relations first. A construction account supports those claims; the relations retain their own definitions.
U.Episteme - Working-Model first; Logical, Mapping, and exact empirical support as appropriate
-
Publication (Working-Model). Authors connect meaning-bearing epistemes or publications using exact knowledge relations (for example, RepresentationOf or UsageOf) in the same human-oriented style.
-
Assurance (downward grounding). If the direct knowledge relation is sufficient, stop after the readable claim. When interpretation or alignment needs assurance, select Logical or Mapping support.
When observation supplies the needed basis, retain the target claim, supported result, scope and temporal qualification required by the use.
In this case the assurance question also asks which Work produced the observation and under which assignment: recover each required performer's A.13 basis, independently admit the Work and its enacted Method under A.15.1, then establish the exact Work-assignment attribution through F.6 and A.2.1.
Cite any MethodDescription, local classification, participants or A.6.1 bindings and result-episteme identity consumed by that claim.
Retain sufficient direct provenance, use A.10 when its source-to-use account needs explicit recovery or replay, and supply any current B.3 assurance claim.
-
Reading result. Working-Model text remains the public form; the exact result and its support chain remain available underneath. The subject claim keeps its own meaning apart from the method, record, and time claims.
Pump-vibration measurement: short recognition, exact assurance underneath
Recognition text. Pump-37 vibration at 09:00 was 2.1 mm/s with stated uncertainty 0.2 mm/s under the current inspection method. A maintenance reader can use that bounded statement and stop before the machinery below. It does not by itself say the pump passes a maintenance criterion, that work may start, or that any gate or permission is current.
Assurance text. This worked slice elects empirical assurance for a current reliance question.
Pump37InspectionPlan-E3, admitted as a U.WorkPlan, had designated the intended measurement and selected PumpVibrationMeasurementMethod-E2, admitted as a U.Method; it cited PumpVibrationProcedure-E5, admitted as a U.MethodDescription, only for the setup and calibration claims used by the plan.
The measurement domain declares PumpVibrationMeasurementAssignment as the assignment species for this work. RA-ConditionMonitoring-7-E4 is its occurrence, is held by ConditionMonitoringSystem-7, and covers the measurement interval.
That System performed the admitted Work Pump37VibrationMeasurement-2026-07-31T0900 under the assignment, and the Work enacted PumpVibrationMeasurementMethod-E2. Check the Work and enacted Method with A.15.1, and the Work-assignment attribution with F.6.
The applicable A.6.1 bindings identify Pump-37, the sensor indication, calibration coefficients, and returned measurement value. Classification of the System under PumpVibrationMeasurementSystemRole is a separate claim.
Use C.16 to characterize the domain-local measurement result by its exact Characteristic, Scale, unit, uncertainty, time stance, and interpretation basis. C.2.1 identifies Pump37VibrationResult-E4, the episteme that states that result.
A.10 path Pump37MeasurementProvenancePath-E6 cites the calibration, Work, bindings, and source publications; B.3 assurance claim Pump37MeasurementAssurance-E2 qualifies only the stated use and window. Neither provenance nor assurance is the measurement result.
No A.15.PROD claim is needed merely because the result episteme exists; open it when a separately current question concerns the Work's productive treatment, the episteme's identity inception, or satisfaction of a production-completion condition. The inception branch asks whether the exact measurement Work first constituted that episteme.
What changes in practice. A reader sees the usable statement first, can inspect the exact Work, result, and support chain when reliance matters, and uses the applicable maintenance-criterion, readiness, gate, or permission pattern if the next decision asks one of those different questions.
Pattern lesson
The Working-Model layer remains the canonical publication face for authors and assurance readers. A direct claim outside an elected profile carries no E.14 assurance fields. When assurance is current, Mapping, Logical, Constructive, and Empirical support remain purpose-selected shoulders beneath the claim. They preserve a short recognition route while keeping the facts, Work, local results, provenance, assurance, and currentness consumed by that use recoverable through the patterns that define them.
Bias-Annotation
Conformance Checklist (normative author duties in reasoning and writing)
E — Worked examples
-
Exact skid assembly -> “Component Of” For PumpSkid 7, recover the pump, frame, reservoir, valve set, and other constituents; the direct fastening, coupling, enclosure, terminal, flange, and seal occurrences that obtain; the applicable skid assembly rule; and the skid reidentification rule.
A commission to state these Component Of relations can end with the established memberships. A commission to explain the identity of the assembled skid also needs the relation and reidentification conditions below. If the publication elects
B.3.5, keep the readable claim first, link it to one current C.2.1sumtrace that reports the basis, and declarevalidationMode=axiomatic.The same parts unconnected or assembled differently do not thereby form PumpSkid 7. A permitted pump replacement may preserve PumpSkid 7. The direct relations and reidentification rule decide; the trace and posture do not.
-
Cartridges that belong to a bank under its collection rule For a four-cartridge bank, identify the bank and its collection-identity rule, then state which cartridge belongs to it and what makes that belonging begin and end.
A C.13
settrace can report the collection for assurance.Parallel use, physical proximity, a list, or an author's gathering act does not establish that a cartridge belongs to the bank, does not imply Component Of, and does not make the bank an acting system.
-
Bearer, facet rule, and aspect -> “Aspect Of” For the thermal envelope of one reactor, identify the reactor bearer, the thermal-envelope aspect, the thermal-facet rule, the Aspect Of occurrence, and the aspect's identity rule.
A C.13
slicetrace can report those facts.Selecting a view, naming a facet, carving a diagram, or choosing a time window creates no aspect occurrence and no independent system.
Notes across the examples • Keep the ordinary working statement first: Component Of or Aspect Of where that direct relation is admitted, and a subject-specific sentence such as “this cartridge belongs to this bank under the bank's rule” for collection belonging. When an assurance profile calls for a construction account, the linked trace makes that basis inspectable. • Structural assertions covered by an elected
B.3.5profile use Constructive assurance. Direct structural claims outside the profile can stand without E.14 assurance fields; epistemic assertions such as “Representation Of” or “Usage Of” use the direct logical or evidence relation appropriate to the claim.
F — Resulting Context (after you apply the pattern)
What improves
- One readable structural vocabulary. In ordinary work, teams can ask whether a stated relation obtains. Examples include component parthood, belonging under a collection's own rule and the relation of an aspect to its bearer. When a profile calls for support, assurance readers can also recover the participants, direct relation facts, construction rule, and identity conditions behind the published assertion.
- Explicit identity tests. Input lists and traces do not decide identity. Different assembly relations can make the same listed inputs another whole; an admitted replacement can preserve one whole. Collections use their own identity rule and belongs-to occurrences; aspects use the bearer, facet, direct relation, and aspect identity.
- Relation labels and their grounds. Engineers use the labels alongside other relation names. Their warrants and construction accounts stay below the working text, so the labels remain readable and the claims can be audited.
What to watch
- Discipline for structural relation kinds. A published structural assertion is unsafe when its direct relation basis or identity test is missing, even if a trace or
axiomaticflag exists. A knowledge relation such as RepresentationOf or UsageOf does not by itself establish structural parthood. Use the direct logical or evidence relation appropriate to the claim. - Author workload follows the required support. Day-to-day model authors use working labels; specification authors recover the direct relation occurrence and identity test and keep a current construction account when the publication policy requires it. That account adds authoring and maintenance work where needed. It supports review but does not repair missing world-side facts.
Invariants you must preserve
- Parsimony of construction accounts. Use
sumto report integrated assembly,setto report a collection, andsliceto report an aspect. Do not treat them as generative acts or add forms for parallelism or time-slicing; order and time remain with their own patterns. - Relation-kind-specific justification. A direct structural claim needs grounded relation occurrences and its applicable identity test. It needs an inspectable construction account only when an elected profile or named current requirement calls for one. Epistemic claims use the logical or evidence relations they need. No assurance route changes the relation kind being claimed.
Known consequences
- Stable queries, fewer surprises. Working labels retain one direct meaning across disciplines. When a structural assertion is covered by an assurance profile, readers can also follow it to the facts and identity conditions reported in its construction account.
- Audit trail without jargon. When construction assurance is current, reviewers can follow a structural claim to its participants, direct relation occurrences, construction rule, identity conditions, and trace edition while everyday collaborators keep using familiar relation names.
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
Why Working-Model is canonical. FPF privileges human-oriented relations as the primary language and working representation for thinking and communication. This satisfies didactic primacy while preserving conceptual integrity: formal work serves the human layer. The canonical template and style principles make that choice part of ordinary writing while preserving notational independence.
Why grounding is presented underneath the claim. The direct claim stands on the pattern that defines or tests it. When assurance is current, Mapping, Logical, Constructive, and Empirical support sits beneath that claim, and the applicable profile or requirement says what must be declared.
Authors select the contribution that addresses the current assurance question: meaning correspondence for Concept-Bridge Assurance (CBA), a reasoned consequence or constructive argument for Verification Assurance (VA), or empirical support for Validation Assurance (LA) under B.3.3. Each contributes only what its actual argument establishes for the receiving use.
Why patterns teach before they tighten. This pattern places its rule beside System and Episteme cases because the contrast helps readers recognize different conditions of use. E.7 selects the subject, scale and model for each grounding from the rule the reader must apply. The cases make application conditions and the recognition-to-assurance return inspectable. They do not by themselves establish reduced cognitive load, practical effectiveness, or universal applicability.
Why interpretation and operative use remain explicit. E.5.2 keeps conceptual meaning portable while allowing an expression to carry reasoning or construction. An illustrative rendering stays informative; an operative expression retains its explained meaning, operation and prerequisites. A proposed substitution must preserve the claims and conditions needed by the use, and must disclose any relevant loss. Readable presentation does not remove the operation or establish equal effort across notations.
SoTA-Echoing
For the question of how a mixed team can use a relation while its grounds remain inspectable, E.14 selects claim-first presentation with support chosen by the receiving question.
Requiring the full assurance stack before every use delays even a direct relation whose meaning and grounds already suffice; exposing the claim alone fails a use that consumes unresolved mapping, inference, construction or empirical support.
Sections 4–6 preserve the direct stop and add the needed support without changing the subject claim.
This is a bounded synthesis of FPF contributions: adopt B.3.3's distinction among meaning correspondence, verification and validation; preserve B.3.5's elected branch obligations and C.13's construction-account content; adapt their presentation so the readable claim comes first.
This arrangement preserves sufficient direct use and the additional grounds of assurance-dependent use.
Reconsider it when a reader cannot recover a consumed condition or a competing presentation supplies the same claim and grounds with less justified effort.
Relations
Builds on:
- E.8 Authoring Conventions & Style Guide — section order, style principles, and mandatory safety subsections used here.
- E.7 Archetypal Grounding — the Tell‑Show‑Show rule applied in this pattern’s own Grounding section.
- C.2.3 Formality of an Episteme (F) — governs assessment of expression properties and any local ordered summary. The assessment establishes neither readability nor assurance; a notation change preserves it when the relied-on properties and interpretation remain unchanged.
Coordinates with.
- CT2R-LOG — Working-Model Relations and Grounding — supplies the optional elected profile that adds
validationModeand, for covered structural assertions,tv:groundedBy; direct relations outside the profile need neither field. - Compose-CAL (Constructional Mereology) — supplies the
sum,set, andslicetrace content when construction assurance is selected; the trace does not define the Working-Model relation or its identity. - E.10 Unified Lexical Rules for FPF — supplies naming and register rules when the human-facing Working-Model is published.
Constrains:
- All architectural patterns that publish relations SHALL present the readable Working-Model claim first. A direct relation outside an elected assurance profile needs no E.14 assurance field. When
B.3.5or another named current requirement applies, attach only its required support below the claim while preserving relation-family separation and notational independence. (Template conformance as per E.8.)
Informs.
- Part F unification practices (context of meaning, bridges, fit levels) by reinforcing the preference for human‑readable labels with explicit alignment notes rather than silent formal substitutions.
E.14:End
Pattern Change, Edition Continuity, and Impact Analysis
Pattern type. Method pattern.
Status. Stable.
Normativity. Normative unless a passage is marked informative.
Compare one predecessor pattern edition with its proposed successor, describe what changed before classifying it, repair only the uses that depend on it, and use a wider search only when a design choice remains.
Problem Frame
Use this pattern when an existing FPF pattern is being corrected, clarified, reorganized, refreshed from current sources, split, merged, renamed, or changed semantically, and someone needs to know what may continue and what must be reconsidered.
The primary EntityOfConcern—the thing being changed—is one existing FPF pattern edition. The candidate is its proposed successor. The useful result is that candidate plus a bounded account of what changed, which uses may be affected, which predecessor ideas remain, and which checks were rerun. Put that account in the decision, review, campaign, or landing result that needs it; this pattern does not require a separate trace object.
First useful move. Put the predecessor and candidate side by side and finish this sentence in ordinary language:
A reader or user who relied on
<predecessor passage>may now read, do, check, or conclude<difference>.
If the answer is “nothing”, test that claim against the affected passages and stop after the smallest adequate repair and check. If the answer is uncertain because several materially different repairs remain plausible, open the alternative-comparison branch.
For a first pattern seed with no predecessor, use E.8 and the patterns that govern the subject. Do not use E.15 merely to run a wording check, make a design decision, perform a quality review, publish a pattern, or land a candidate: F.19 with E.10 cues, E.9, E.19 or E.21, E.24.PUB, and the landing process own those distinct questions. Return here when one of those activities changes an existing pattern edition and edition continuity or affected use is in question.
Problem
Two failure modes pull pattern change in opposite directions.
- A quick patch can preserve the visible sentence while losing a predecessor idea, changing a direct consumer, or leaving an old instruction elsewhere.
- A safety-minded author can turn a small repair into a full search, scoring, evidence, and publication programme whose records cost more than the decision and still do not prove that the chosen text is better.
Labels do not solve the problem. Calling an edit “lexical”, “minor”, “refresh”, or “refactor” does not say whether practitioner entry, inputs, action, conditions, result, ontology, or assurance changed. A version number communicates an already made compatibility judgment; it does not make that judgment true.
Forces
Solution
Recover the actual change
- Name the predecessor, candidate, question, and receiving use. Use exact editions or recoverable source values. Add a ClaimScope, ReferenceScheme, model-use structure, or other qualifier only when the receiving use depends on it.
- Read the changed passage in both wholes. Inspect enough of each pattern to understand what the passage does in each edition.
- Describe what changes for the practitioner. Ask separately whether the change alters recognition or entry, required inputs, the first or later action, applicability or stop conditions, returned result, normative claims, ontology, dependent uses, or assurance needed for reliance.
- Classify only after that comparison. Choose the smallest Delta-Class in §4.3 that matches the observed effect. The planned label, commit subject, or amount of changed text is evidence to inspect, not the answer.
Keep wording, examples, informative rationale, normative conditions, and public naming distinct; changing one does not silently change the others. Keep ClaimScope and WorkScope distinct when both are current.
If an exact predecessor value needed for this comparison is unavailable, return that bounded continuity gap. Do not reconstruct it from a later edition, a title, or a remembered summary.
Find the affected reach
Start with the changed claim or instruction and ask who uses it to read, act, check, decide, derive another statement, or preserve a public name. Search results and Relations entries help discover candidates, but neither proves dependence.
For every plausible consumer, decide one of three things:
- depends: its action, interpretation, condition, result, check, or public reference would change if the repaired claim changed;
- mentions only: it cites or describes the pattern but its current action remains valid;
- unresolved: the dependency cannot yet be decided from recoverable content.
Repair the exact dependent loci. Reuse an earlier result only when its conclusion and conditions remain unchanged and the changed premise lies outside its actual dependency. Reopen the smallest affected premise, consumer, example, check, or result; do not rerun an unrelated whole programme merely because an edition number changed.
An undeclared consumer can depend on the changed premise, and a declared dependency can be unused in the current question. Check actual and declared reach when the distinction matters.
Classify the actual delta
These are impact classes, not mandatory version-number syntax. If a publication uses SemVer or another version policy, map the already justified compatibility decision into that policy. Do not infer the class from major, minor, or patch.
Refine, rephrase, split, merge, generalize, constrain, rename, add, and retire remain useful edit descriptions. None has a fixed Delta-Class without its actual effect. A split that preserves every use may be Δ-1; a one-word change that reverses an obligation is Δ-3.
Choose the least costly adequate route
Direct bounded repair. Use this ordinary route when the defect and one non-dominated repair are understood. Make the repair, inspect its actual consumers, perform the selected focused checks, and stop. Do not generate dummy alternatives or a search record.
Alternative comparison. Open this branch only when at least two materially plausible designs remain, a current SoTA choice can change the action, or a repeatable search is itself useful. State what the alternatives differ on and which intended use decides among non-dominated candidates. C.18 and C.19 may generate and retain alternatives when novelty or diversity is part of the question; E.22 and E.21 frame and evaluate pattern qualities.
Keep hard constraints separate from quality comparisons. A failed identity rule, broken reference, missing required result, or unreadable first action is a defect to repair, not a low score to trade away. Compare readability, precision, assurance cost, breadth, or other qualities on their applicable scales. Select by the stated intended use and protected trade-offs; do not add heterogeneous values into an undeclared winner score.
Return a decision gap. If the repair depends on an unresolved ontology, authority, source choice, or architecture decision, return that exact gap to its pattern or decision record. More variants do not compensate for a missing governing distinction.
Preserve the predecessor by independent probes
For a material rewrite, derive a predecessor-use inventory from the predecessor before relying on the author's preservation map. Include each distinct working situation, first move, input, condition, result, prohibition, example function, consequence, source-derived contribution, and consumer-facing promise that the predecessor carried.
Then test each probe against the candidate:
- preserved: the same practical or semantic function remains;
- changed intentionally: the successor decision names the new function and why;
- moved: an exact current locus still supplies it without making discovery worse;
- retired: an explicit decision removes it and states the affected use;
- lost or unresolved: repair it or return it before claiming continuity.
For text copied unchanged, identity can establish preservation of wording. To establish preservation of a working function, compare the reader's question, referents, conditions and available support before and after the move. A passage that still names the same operation can lead its new recipient to do a different job. For a deletion, compression, move or rewrite, follow each affected predecessor function to its destination or identify the loss.
The independent inventory need not become a permanent row-by-row file when the receiving workflow needs only the verified candidate and aggregate result, but the inspection must be complete.
Check the candidate proportionately
Select checks from the actual change and intended conclusion. A small Δ-0 repair may need one focused check. A Δ-2 or Δ-3 change may require semantic, ontological, consumer, source, preservation, or independent quality checks. Reuse a current check result when candidate, question, conclusion, and conditions are unchanged.
After a material ontological or formal repair, read the whole changed pattern for what an intended practitioner can recover from the publication and its stated prerequisites. This author walkthrough supplies expert evidence. Obtain an actual reading under F.19 when its conditions require one.
A local token scan cannot establish precise plain language. Check that the working situation, first action, examples, conditions, and result remain understandable without reconstructing the ontology from elsewhere. Simplify the expression, not the distinction. Keep a technical term when it names a needed object or relation; remove stacked qualifiers and formal notation when they do no action-changing work.
Author-side use of E.10, E.19, or E.21 questions is development evidence. It does not become the independent review, complete quality result, admission, or landing conclusion that a later use may require.
Keep one useful change account
Use the receiving workflow's existing record. A compact change account normally needs only:
These answers may live in a DRR, review package, campaign result, source-use account, or landing preservation result as that workflow requires. Cite an existing E.21 or E.22 evaluation, F.15 result, source-use record, or decision instead of copying it. Do not mint a dedicated authoring trace, publish a work log with the pattern, or treat a file or publication occurrence as proof that the change was performed well.
Edition continuity and stop rule
Keep accepted historical editions immutable and recoverable. A successor does not rewrite what an earlier edition meant. A source-edition change reopens only claims and actions that relied on the changed source value; unchanged exact inputs and unaffected premises remain reusable.
E.15 finishes when the candidate answers the change question, every in-scope locus that depends on the change is repaired or explicitly unresolved, material predecessor functions have dispositions, and the selected checks support the claimed Delta-Class and continuity. Publication, acceptance, registration, and landing remain separate later decisions.
Schedule a living refresh only for a high-value claim likely to change and only when someone will use the signal. Name the trigger and affected claim. Retain any review interval required by an applicable publication or maintenance obligation.
Archetypal Grounding
Tell. Change the smallest semantic unit that solves the problem, but judge continuity at the scale where a reader or consumer could be harmed.
Typo with no semantic reach
An identifier is spelled correctly everywhere except one explanatory sentence. The exact identifier, instruction, and checks remain unchanged. The author repairs the sentence, verifies the identifier and nearby reference, classifies the change as Δ-0, and stops. Three alternative phrasings and a DRR would add no value.
Plain-language repair after an ontology correction
A relation passage is ontologically exact but has grown into two pages of qualifications that hide the first action. The candidate restores one readable explanation and keeps the exact relation test in the assurance section.
Because a substantial rewrite can lose ideas, the author independently inventories the predecessor's entry, action, conditions, examples, and prohibitions, then reads the whole candidate for what an intended user can recover from its public explanation. If every semantic function remains, the result may be Δ-1 or Δ-2 depending on whether a normative clarification also occurred; the word count does not decide.
One-word semantic change with dependent consumers
A conformance rule changes may to must. The diff is one word, but admissible use and failure conditions change. The class is Δ-3. The author repairs examples, checklist items, and direct consumers that relied on the optional branch and rechecks their conclusions. Unrelated source and publication checks are reused.
A source edition changes one relied-on premise
A current research edition revises a limitation used by one SoTA decision. The pattern's other sources and practitioner steps do not depend on that premise. The author reopens only that source-use decision and its receiving passage. If the selected action changes, the affected consumers follow; if it does not, the account states why the current result remains supported.
A choice among three architectures
A pattern could model a new distinction as a local value, a direct relation, or a selected structure, and each choice changes downstream use. No single repair is yet selected. The author records the alternatives in the DRR, uses subject-specific criteria and E.21/E.22 qualities, and may use C.18/C.19 to broaden the candidate set. The comparison stops when the intended use supports one non-dominated architecture. Authors need this search only under the conditions in §4.4.
Bias-Annotation
The method biases toward continuity and inspectability (Arch, Onto/Epist). The direct-repair route, record-reuse rule, whole-pattern practitioner-use reading, and first-adequate-result stop protect practical speed and readability (Prag, Did). Decision and review authority remain with their own patterns (Gov).
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits. Small repairs stay small. Material changes show which uses depend on the changed content. Historical editions remain usable. Independent predecessor probes make large rewrites safer. Alternative search remains available where it can improve a decision, and whole-pattern plain-language checking keeps ontological precision usable.
Costs and limits. Determining which uses depend on a change and which functions the predecessor served requires semantic judgment. A repository search cannot fully answer either question. A Δ-2 or Δ-3 repair can be expensive when many consumers depend on the changed premise. This pattern reduces redundant checking but cannot make a broad semantic change local by declaration.
Reopen the method when the four Delta-Classes no longer distinguish action-relevant change, when dependency-focused verification demonstrably misses affected uses, or when a lower-effort method provides equal or better preservation and decision quality.
Rationale
The two routes are asymmetric. The common case receives a short path because extra alternatives and records cannot improve an already understood bounded repair. The strong branch remains because architecture, ontology, and current-source decisions sometimes have several non-dominated answers.
Exact predecessor comparison and affected-reach analysis work together. The predecessor prevents history from being rewritten; actual dependency prevents an edition label from reopening everything. Independent preservation probes answer a different question from an author's change explanation: they test what the explanation may have omitted.
Delta-Class stays useful as a compact impact signal, but only after the change is established. E.21/E.22 own pattern quality, E.9 owns content decisions, E.19 owns review, and lifecycle patterns own publication and landing. E.15 connects these results without duplicating them.
SoTA-Echoing
The selected method combines exact edition comparison with dependency-aware incremental reverification and optional design-space search. No one source supplies the complete FPF procedure.
Relations
Builds on:
- E.8 for first-edition authoring shape and practitioner-facing pattern structure.
- E.10, F.18, and F.19 for wording-use diagnosis, naming, and precise plain language.
- E.9 for a material content decision and E.9.DA when that decision needs adequacy review.
- F.0.1 and F.1 for exact source-local meaning and question-relative source selection.
- A.10 and B.3 when a changed claim depends on evidence use or assurance.
Coordinates with:
- A.10.1 for the general move from a changed source claim to bounded actual uses when cross-use discovery is needed. E.15 is the FPF-pattern-edition specialization of that move: its primary object remains one exact existing FPF pattern edition and its successor, and its Delta-Class, predecessor-function continuity, proportionate pattern checks, and candidate-plus-change-account result remain intact.
- E.19 for pattern review, E.21/E.22 for quality evaluation, and E.23 for repeated improvement.
- C.18 and C.19 for optional candidate generation and explore/exploit control when the alternative-search branch is open.
- F.15 for applicable regression checks and F.9 only when the changed use relates distinct local senses.
- B.4 for later evolution-loop scheduling when a named refresh trigger is worth maintaining.
- E.24.PUB and the landing process for later publication and integration; neither follows from an E.15 result.
E.15:End
RoC‑Autonomy Budget & Enforcement
Status: Stable
Use this when. A claim about a local system-role kind, Method or Service says that a System may decide or actuate without continuous human direction. Start by publishing the claimed autonomy and its limits in an AutonomyBudgetDecl, stating whether it is prospective, action-bound or enactment-bound. The rules below govern permission, override, actual Work accounting and publication.
Tools that only suggest actions fall outside this pattern when a human confirms each action at the point of execution.
Builds on: A.2 / A.2.1 / A.2.5 / A.2.7 / A.15 / A.21; B.3; C.16; E.8; E.10; E.18; F.4; F.6; F.8; F.15; F.17. Coordinates with: A.13 (Agential Role) and A.17/A.18/A.19/C.16/A.10 for current agency characterization, measurement, and evidence; planned C.9 (Agency Characteristic Profile) only as future consolidation; C.24 (Agent-Tools-CAL) where applicable; G.4, G.5, G.8, G.9, and G.10 (method authoring, selection, and shipping).
Problem Frame
A System that performs Work without continuous human direction must stay within declared safety, risk, resource, and remit limits and yield through the stated override path. The same need can be declared prospectively for a Method or Service before a performer or Work item exists. Without a uniform rule, an autonomy claim drifts into tacit norms, cannot be benchmarked or audited, and undermines selection (Part G) and publication (Part F).
Problem
- Opaque autonomy. Patterns assert “autonomous” behavior with no budget or enforcement.
- Un‑gated execution. Systems can perform Work beyond authority or risk limits.
- Ad‑hoc overrides. No standard SpeechAct for pausing/de‑scoping; SoD is unclear.
- Non‑portable publication. UTS (Unified Term Sheet) rows cannot surface autonomy‑critical data for parity or selection.
Forces
Bias-Annotation
- Gov. Bias toward enforceable oversight (hard gates, SoD, canonical override SpeechActs). Autonomous exploration is allowed only within a stated budget and time window.
- Arch. Bias toward gate‑and‑ledger structure (Green‑Gate + Work‑anchored
AutonomyLedger). Mitigation:telemetrySpecRefcan scope what is emitted when full deltas are unnecessary. - Onto/Epist. Bias toward typed, testable constraints (MM‑CHR tokens, explicit admissibility checks). Mitigation: the budget fields marked
?are optional, so low‑risk contexts can start with a minimal budget and tighten its limits over time. - Prag. Bias toward measurable quotas may under‑express “soft” autonomy goals. Mitigation: pair
decision_tokenswithrisk_bandsto capture non‑counting limits. - Did. Bias toward explicit mechanics increases authoring surface area. Mitigation: provide a default
AutonomyBudgetDecltemplate and minimal harness cases in F.15.
Solution — Rule‑of‑Constraints (RoC) for Autonomy
This RoC applies whenever wording about a local system-role kind, Method, or Service claims that a System may perform Work involving unsupervised decision or actuation.
E.16-S1 (Autonomy Budget - mandatory). Any autonomy claim MUST publish a named, versioned AutonomyBudgetDecl. A prospective declaration fixes what is being claimed and how later Work will be bounded. It makes no claim that a performer, assignment, Work item, or authority occurrence already exists. An action-bound declaration supplies the proposed-action and actual allocation references for A.21 permission to start. An enactment-bound declaration additionally refers to already admitted Work.
An unscheduled prospective budget may omit actionBinding, enactmentBinding, nonexistent assignments and authority occurrences. Permission to start a scheduled action uses action-bound with its proposed action, work-entry claim, real holders/assignments and current independent authority.
Use A.2.7's separately declared prospective incompatibility species for that allocation check. enactment-bound adds Work only after A.15.1 admission, retaining the applicable action/permission match. A request, schedule or pass result creates no Work. If authority rotates or a permission-relevant action/window changes, recheck before relying on the permission.
The holder Systems, local kinds, any separate System-classification judgments, assignment occurrences, Work, budget declaration, later override Work, authority relation, and separation-of-duties relation are different objects. A kind reference neither classifies a System nor creates an assignment; an assignment alone grants no authority.
E.16‑S1.A (Scout / probe / commit partition for bounded specialization).
When an autonomy-bearing method uses bounded specialization scouting, the budget declaration MUST keep the scout budget, probe budget and commit checkpoint separate. A successful probe does not by itself authorize a committed route, wider burn, or scope widening. Leaving probe state requires one explicit checkpoint decision through the declared guard or override path, with budget burn and residual budget recorded in the AutonomyLedger. E.16 governs this budget partition plus guard and ledger enforcement; it does not replace the dyadic move of A.15 or the CheckpointReturn plan semantics of C.24.
E.16-S2 (Guarded enactment - Green-Gate).
A Method step requiring autonomy MUST name the required local kind and requiresAutonomyBudget: AutonomyBudgetDecl.id. Green-Gate decides permission for the A.21 prospective work-entry claim and bounded action. It resolves:
- the action's artifact/subject, operation, target and intended window under its identity/continuation rule;
- the real prospective performer System and obtaining A.2.1 assignment whose holder/kind match the declaration, with the required A.2.5 state and any separate classification judgment;
- the actual authorizer/override-authority System and assignment, the applicable prospective incompatibility result, and the independent authority covering this action;
- applicable ClaimScope, qualification/accounting windows, remaining budget and every required guard.
For an action drawing on a shared limit, establish what is still available to commit in that accounting window. Include consumption already recorded and outstanding permissions or other commitments that could draw on the same allowance together. Apply the policy's units and overlap rule. Enforceably exclusive alternatives may share a reservation. One performance that genuinely fulfils two obligations incurs its actual charge once.
A known prohibited allocation blocks the action before performance. Keep every required check present with its source outcome, including missing information, unknown or notRun, then apply the mapping in the applicable A.21 profile. A required missing or unrun result may map only to degrade or block under an explicit rule; an unknown result retains its uncertainty and the policy-qualified consequence. Any accepted uncertainty must name its subject, tolerance, permitted bounded action, consequence and recheck/expiry condition.
Reserve abstain for a gate that makes no decision and states the remaining decision route. Aggregate only after these mappings are known; a pass permits only the bounded action under the checked conditions.
When one token remains and two distinct proposed actions would each consume it, the gate may grant one unconditional permission and reserve that token. It can also serialize commitments or make each permission explicitly conditional on an enforceable final check that secures the token before use. A later check can settle which conditional action may start. Two unconditional promises already made still have to be honoured or renegotiated. Where the policy permits a quantified risk, compare its supported probability or bound with the stated tolerance. A hard cap still governs every permitted joint use.
Resubmitting the same action changes no action identity. Rescheduling reopens window-sensitive permission even when the action-continuation rule preserves a continuing action. Work is admitted separately from actual performance history under A.15.1.
E.16-S3 (Autonomy Ledger). Every Work item admitted under this budget MUST have an AutonomyLedgerEntry:
Existing budget and performance records may supply this declaration and ledger when they retain the required meanings, identities and references. A second accounting store is unnecessary in that case. The ledger is evidence about actual Work. When prior permission is relied on, cite its exact result and the policy-supported match from this performance to the previously bounded action; a shared request label is insufficient. The Work, its performer System, its A.2.1 assignment, and the performedUnderAssignment attribution remain separately recoverable.
When prior permission reserved an amount, match actual consumption to that reservation and account for any unused balance under the policy. A reservation can affect the next permission before any Work has taken place.
For reporting, use Γ_work (B.1.6) for the recorded resource values under the applicable accounting and overlap policy, and Γ_time (B.1.4) for the recovered temporal relations among the Work occurrences.
E.16-S4 (Overrides - SpeechActs, authority, and separation of duties).
Every budget MUST reference an overrideProtocolRef that defines the available SpeechActs:
- PauseAutonomy(budgetId) - stop autonomy-gated steps immediately;
- ResumeAutonomy(budgetId) - resume after the required checks;
- NarrowAutonomy(budgetId, Δscope) - apply stricter limits;
- Escalate(budgetId) - hand over through the declared override-authority path.
Narrowing this authorization can change who may do part of a service while the commission for the service still stands. For a request about the budget, the new limit can complete the answer. If a recipient also needs the service, establish which authorized performer or arrangement can supply the rest, or take the unresolved commitment to the person empowered to revise it.
The declaration names the A.2.7 incompatibility species and its predicate. Before a proposed pause, resume, narrowing or escalation, identify that bounded override action, its real holder/assignment and independent authority. The target navigation Work may already exist; the proposed override is not yet override Work.
Apply the actual-Work species when its condition concerns the existing target Work, or the separately declared prospective species when the policy concerns allocation to the proposed override action. Resolve both actual assignments, holders, relevant subject identity, windows and applicability; reject a known prohibited joint allocation. Different labels or assignment IDs alone do not establish separation.
Check authority for the proposed override independently. Once the override is performed and admitted under A.15.1, record its overrideWork ledger entry and policy-specified zero, negative or other budget delta. Match that performance to prior permission by the policy's action rule. The proposal and its pass result remain distinct from that performed SpeechAct Work.
E.16-S5 (Depletion behavior). When a budget depletes — no tokens remain, an envelope is exceeded, or a cap is breached:
- block ordinary autonomy-gated continuation in the same accounting window;
- emit a DepletionNotice SpeechAct and either Escalate or Park as the policy says; and
- resume ordinary continuation only after an admitted System performs ResumeAutonomy under its exact override-authority assignment, the A.2.7 predicate check over both actual assignments passes, the independent authority relation is current, and the ordinary guards pass.
If the applicable policy permits an exception, recover the condition for that exception, the bounded action it permits, the current authority and every remaining guard. Apply S2 to that action and retain the policy’s accounting and recheck conditions.
A fully reserved allowance can block a new unconditional permission while the ledger still shows unconsumed tokens. Use the policy's release or expiry rule for the reservation and check the resulting availability. Apply the depletion procedure when its defined trigger occurs.
For example, a service policy may permit a qualified security fix while its exhausted error budget still blocks ordinary feature releases. Missing exception authority or a failed remaining safety condition leaves that action blocked. Permission for the exception does not resume ordinary continuation.
E.16‑S6 (Publication in UTS). A UTS row that carries an autonomy claim about Work described through a local system-role kind, Method, Service, or Selector MUST include:
AutonomyBudgetDeclRef(id and version) andbindingState;Aut-Guard policy-id (PolicyIdRef);OverrideProtocolRef;- declared Scope (G) and Γ_time window;
- edition pins for the referenced local system-role kind, Method, CHR, and policies. When action-bound, include the proposed-action and real allocation refs. When enactment-bound, add the actual Work refs needed by the receiving use.
- (optional, if a scale preference is declared)
ScaleLensPolicyRefandScaleLensOptIn ∈ {OptedIn, Neutral, OptedOut}.
E.16‑S7 (Scale & selection — optional lens). When autonomy interacts with open‑ended search (C.18 and C.19), budget consumption and guard violations are selection lenses in Part G (G.5/G.9). Applying a Scale‑Lens / Bitter‑Lesson preference is OPTIONAL. Authors MAY declare a ScaleLensPolicy for the autonomy claim; when declared, it MUST state:
- Trigger criteria — evidence that expected utility‑of‑scale is monotonic/non‑saturating on held‑out tasks, and a threshold at which scaling beats structured heuristics.
- Budget fit — compute/latency/cost targets within the declared
AutonomyBudgetDecl(Γ_time, resource_caps). - Safety invariants — scaling must not weaken guards or SoD. No policy may bypass E.16 gates.
- Fallback — a degrade‑gracefully plan if scaling fails to clear the trigger criteria within budget.
If no ScaleLensPolicy is declared, selection remains neutral with respect to Bitter‑Lesson; RoC does not authorize ignoring scale‑safety guards under any policy.
Archetypal grounding (human-centric)
Show-A (enactment-bound mobile robot).
The autonomy claim names navigation Method Navigate_v3. Its enactment-bound budget names NavigatorSystemRole as the consumer kind, robot Robot_R7 as holder, assignment R7-NavigatorAssignment-2026, and the current warehouse-navigation Work item. It also names the warehouse policy, ClaimScope and shift window, FloorSupervisorSystemRole as the override-authority kind, supervisor System Mina, her exact assignment, and the independently obtaining authority relation for pause and resume Work.
The declared A.2.7 relation is NavigatorSupervisorIncompatibility; its predicate prohibits the same System from holding both assignments for the same navigation Work during overlapping windows. The gate resolves both A.2.1 assignments and their holders and admits the override path because the actual pair does not match that prohibited case and the independent authority relation is current. The budget then supplies action_tokens=10 k steps/day, risk_bands={maxSpeed <= 1.2 m/s, minDist >= 0.5 m}, and resource_caps={battery >= 20%}. Ledger entries decrement the action budget and record distance checks. Depletion stops autonomous movement.
For a proposed pause/resume of that current navigation Work, name the override action and check its permission before it occurs. The navigation Work and its existing ledger remain unchanged by the proposal. Only the performed, independently admitted override adds override Work and its actual budget delta.
Show-B (unscheduled, action-bound, then performed release).
An unscheduled budget names the deployment/authorizer local kinds, policy, limits and prospective incompatibility species without inventing assignments or release Work. When release promotion is proposed, its local action rule fixes artifact Release-E7, target Production-East, operation promote, and intended window 14:00–14:15. Use its existing WorkPlan action locator if available; the rule otherwise identifies those values directly. Duplicate request IDs for these same values name the same proposed action.
ReleaseDutyProspectiveIncompatibility has the unordered kind pair {DeployerSystemRole, ReleaseAuthorizerSystemRole}. Under its adopted policy it prohibits the same actual holder's obtaining assignments to those kinds for the same proposed release action during overlapping allocation windows. Its applicability and meaning-changing policy edition are explicit. The kind relation obtains independently of whether anyone attempts a prohibited allocation.
The illustrative ReleaseEntryProfile-E1 requires allocation, authority, budget and the action’s other declared guards. It maps known prohibition, unknown allocation, and missing/unrun required allocation checks to block; satisfactory current results map to pass. Its hold consequence is to establish the allocation facts and recheck within the proposed action window. This profile belongs to the release example. A.21 sets no such default.
The declaration's decision_tokens=3/day and error-budget burn <= 2%/day remain typed limits checked under their policy. This branch neither weakens the actual-Work incompatibility species nor treats a gate as Work admission.
Conformance Checklist (SCR - E.16-CC)
Consequences
- Testability. Budget use is checkable against declared tokens and envelopes. Work-anchored ledger entries support audit. PauseAutonomy stops autonomy-gated steps.
- Comparability. UTS publishes the autonomy metadata used for fair selection and parity.
- Safety. Required guards remain binding; depletion stops ordinary continuation, with only independently permitted, action-specific exceptions.
SoTA-Echoing — bind a budget decision to the proposed action
Practice question. Is an existing release policy and error-budget ledger enough to permit an autonomous promotion or override? The selected line reuses those controls while checking the proposed action, current allocation and authority, available budget and applicable guards before performance. It then records the actual Work and consumption.
A lighter alternative gates releases solely by the existing service-level error-budget policy and records their effects afterward. That is sufficient when the release policy already supplies every action/authority condition needed by the use.
The Google SRE Workbook's Example Error Budget Policy, 2018 provides a comparison. Its four-week budget can freeze changes, explicitly excepts P0/security fixes and supplies an escalation route. Adopt policy-bound depletion and explicit exception handling in S2, S4 and S5. Adapt the budget to the exact autonomous action and current authority in S1–S2, rather than inferring those facts from the remaining service budget. The example supplies a bounded release policy. It neither sets a general autonomy standard nor establishes that a given gate is complete.
In Show-B a release can be within its error budget while the required allocation overlap remains unknown. Budget-only permission would miss that unresolved premise; ReleaseEntryProfile-E1 blocks until it is established. A changed window reopens the affected permission, while S3 records consumption only after actual performance. Show-A adds two controls: an authority for pause and resume, and a stop when the budget runs out.
The same existing policy, ledger and allocation records can supply these inputs; a second accounting system adds no value when their meanings and exact references are already present. The extra work is resolving a current action/allocation question, not copying the budget into another form.
Reject treating a declared pause interface as proof that a learning agent has no incentive to resist interruption. Orseau and Armstrong's Safely Interruptible Agents, revised October 2016 supplies the distinct learning-theoretic question and results under stated agent and learning assumptions. It does not establish E.16's runtime enforcement or override authority. S4 requires an effective override path; any stronger incentive or learning claim needs its own applicable analysis. S1.A supplies a local budget-control rule for scout, probe and commit. This pattern cites no agentic-search study as evidence that the rule works.
At comparable operational effort, the selected line wins only when an action-sensitive premise could change permission: it uses the same budget evidence and adds that check before harm can follow the proposed action. For a policy already covering it, reuse the existing decision and accounting basis while respecting their currentness. Reopen when the action rule, authority/allocation policy, budget window or override mechanism changes, or when evidence shows a bypass or a cheaper control retaining the same required distinctions. No error-budget policy, interface declaration or interruption theorem alone establishes general autonomous-system safety.
Common Anti-Patterns and How to Avoid Them
Rationale & E‑/F‑/G‑links
- E.10 — uses LEX‑BUNDLE: Scope via ClaimScope (G), time via Γ_time, and L‑AUTO for autonomy wording.
- Mint/reuse authority (policy-ids). Mint/reuse authority is expressed via F.8:8.1 (
PolicyIdRef:PolicySpecRef+MintDecisionRef?) and explicit GateCrossing checks (E.18) evaluated by the active GateProfile/GateFit (A.21); no tier ladder is required. - Part F — integrates with F.4 Role Description (RCS includes AgencyLevel; RSG gates), F.6 for exact performed-Work attribution after independent A.15.1 admission, F.15 SCR/RSCR (harness includes depletion/override tests), F.17 UTS (columns, incl. optional ScaleLens fields).
- Part G — G.4/G.5: method authors must declare budgets & guards; G.9 parity includes autonomy consumption & violations; G.10 shipping requires UTS autonomy fields.
Mini conformance checklist (cross-E-F; author's quick use)
- Declare the boundary: name the autonomy claim, consumer local kind, working situation, policy, ClaimScope, window, budget, override-authority kind, and A.2.7 incompatibility relation.
- Identify the proposed action: keep an unscheduled budget prospective; use action-bound for a scheduled action, its work-entry claim, real assignments/holders and independent authority. Apply the declared action identity and continuation rule.
- Gate permission to start. Check the bounded action against the prospective allocation, guards and budget still available after overlapping commitments. Recheck permission-relevant changes and state any condition that must still hold before use.
- Record actual Work: after independent A.15.1 admission, emit the Work-anchored ledger entry, attribution and actual delta; relate it to prior permission through the policy-supported action match.
- Check override separately: distinguish the existing target Work from the proposed override; apply the selected actual-Work or prospective-action species to the real assignments and applicable window, and independently test authority before performance.
- Publish what users need: expose the budget edition, binding state, policy, override protocol, scope, and window in the UTS row.
Telemetry and selection lenses are optional.
E.16:End
Viewpoint and View Recognition for Multi-View Describing
Status: Stable
At a glance. A viewpoint specifies what descriptions made for a purpose must cover. To choose one, start from what the reader needs to decide or explain (§4.8). To recognize an existing view, test the identified account against the specified viewpoint’s rules (§§4.1–4.5). Its label, layout, generation history or place in a bundle can help locate the account.
Use this when. A description, model slice, query result, diagram, or other claim-bearing episteme is being called a functional, safety, maintenance, architecture, or other view, and the next reading, comparison, construction, or publication depends on whether that claim is warranted.
What goes wrong if missed. A viewpointRef, familiar face name, generated table, or readable diagram is accepted as a view without testing the viewpoint's concerns and rules. The opposite failure is to rebuild a viewpoint convention, bundle, evaluation package, and publication dossier before an ordinary reuse can proceed.
What this buys. One stable test works for directly authored and derived epistemes. Identify the candidate E, resolve the viewpoint edition P, and test P's fixed conformance predicate. Both epistemes retain their identities.
First action. For selection, identify an alternative that changes the reader's action or the grounds for it, then use §4.8 to choose the distinctions the description must preserve. Use §4.6 if a needed convention must be authored or revised. For recognition, recover candidate E through C.2.1, resolve the existing U.ViewpointRef to P, and read the rules fixed by P. The first result of that test is the conformance judgment below.
First useful result. State one readable judgment: either episteme E conforms to viewpoint edition P, in which case the same E is a U.View, or E does not conform to P, naming the failed fixed rule without inventing a negative relation occurrence. Keep the identities of E and P recoverable. If missing identity or interpretation prevents the fixed predicate from being evaluated, report the unresolved condition rather than manufacturing a negative result.
Ordinary stop. If the next work needs only view recognition, stop after that judgment. Add an occurrence designator, explicit result ValueKind, evaluation package, source-viewing relation, correspondence model, collection or structure, publication occurrence, form, or carrier only when a named receiving use depends on it.
Tech-name:
MultiViewDescribingPlain-name: recognizing viewpoints and views in multi-view describing
MultiViewDescribing names this pattern's method. It is not a public U-kind, a family record, or an extra entity beside the epistemes and relations recovered below.
Builds on: C.2.1 for episteme identity; C.13 for collections; A.22 for selected structures; A.6.5 for relation-signature participant SlotSpecs; A.6.3 for an optional source-to-view construction relation; C.2.1:6 for description and specification use, with E.10.D2 for naming; E.24.PUB for publication; A.6.3.RT/A.6.3.RT.OE for expression construction and C.29 for mathematical representation.
Used by: E.17 publication, E.17.1 viewpoint bundles, E.17.2 TEVB, E.18 transformation-flow descriptions, and domain patterns that compare several views.
Problem frame
An engineer may have several claim-bearing epistemes about one system, method, structure, work occurrence, or another identified entity. A functional description, safety description, maintenance description, and allocation description may serve different concerns. One episteme may also be constructed from another by a query or projection, rendered in several forms, published several times, or compared with another view.
Those uses involve different objects and relations:
- the EntityOfConcern of each episteme;
- the episteme itself, identified under C.2.1;
- a
U.Viewpointepisteme carrying fixed concerns and conformance rules; - an obtaining
EpistemeViewpointConformanceRelationoccurrence; - dependent
U.Viewmembership of the same episteme individual; - an optional A.6.3 viewing relation recording how one episteme was constructed from another;
- an optional viewpoint selected for one current describing use when it changes what that use reads or checks or may conclude;
- exact correspondence relations and epistemes that assert or describe them;
- publication occurrences, forms, carriers, and representations.
Use the list to choose the distinctions your task needs. Ordinary positive recognition needs items 1 through 5; a negative test stops without an obtaining conformance occurrence or U.View membership. Construction, selection, correspondence, and publication stay outside unless the receiving use calls for them.
Problem
How can an engineer recognize and use views under explicit viewpoints while preserving exact episteme identity and direct relation semantics, without treating a selected viewpoint, a query result, a diagram, a family label, or a publication as what makes an episteme a view?
A common error is using the wrong object to justify the next action. For example, an engineer may accept a table as a view because it was generated, accept a published diagram because it is readable, or treat a viewpointRef as proof of conformance. Calling a package of documents a multi-view model without recovering a cross-view relation makes the same mistake.
Forces
Solution
Recognition test. Identify the candidate episteme. Resolve the existing viewpoint episteme. Test their fixed conformance predicate. If it obtains, recognize the same episteme as a view; if it fails, name the failed rule and stop or repair. Add construction, selection, correspondence, evaluation, or publication only for the current use.
When the question concerns choosing what each account must show, use §4.8 before selecting a convention. If available conventions omit a needed distinction, §4.6 governs creating or revising one. For the question whether a particular account conforms to an already specified viewpoint, apply the following recognition test.
See §§4.1–4.11 for the object distinctions, obtaining rules and stopping conditions behind this reminder.
Identify the candidate episteme before calling it a view
Recover candidate E : U.Episteme through C.2.1:
- claim content;
- EntityOfConcern
T; - effective
ReferenceScheme.
These three discriminators identify E. A layout, file, query run, viewpointRef, selected project context, publication form, or carrier does not add another episteme identity discriminator.
Recover a diagram element or graph node under the chosen representation and its correspondence, adding C.29 when a mathematical lens is used. Recover a publication form or carrier under E.24.PUB. Do not promote it to an episteme or view by appearance.
Resolve one exact viewpoint episteme
U.Viewpoint is a same-individual dependent durable kind under U.Episteme. One viewpoint P is the same individual as a C.2.1 episteme.
P has one C.2.1 EntityOfConcern. In the ordinary self-contained branch it is the exact independently admitted durable or local target kind whose membership criterion P uses. For a local kind, recover which candidates it can classify, what intended members must satisfy, what separates relevant non-members, and when a changed declaration still describes the same kind. A practice or source boundary helps find and compare that membership rule; it does not decide kind identity.
Only when separately versioned convention components and their organization change a named reuse, comparison, or maintenance action is P instead about the selected S_viewpoint : U.Structure. Neither branch introduces a new public kind or organization record.
Existing-P route. Resolve one current U.ViewpointRef to P and inspect P's fixed target criterion, admitted kinds, concern coverage, semantic-form, completeness, consistency, and omission rules. Do not reconstruct P's constituent collection or authoring history merely to use the already admitted edition.
Run the ordinary E/P route contiguously
- Identify the candidate episteme E under C.2.1.
- Resolve the existing viewpoint edition P and its fixed rules.
- Apply the five-condition test in §4.4 to fixed
<E,P>. - State exactly one readable result: positive, negative with the failed fixed rule, or unresolved with the missing identity or interpretation.
- Stop unless a named receiving use triggers exact occurrence designation, warrant, new-viewpoint authoring, evaluation, selection, construction history, multi-view organization, or publication detail.
Test the direct conformance relation and state the result
EpistemeViewpointConformanceRelation is a direct species of U.Relation. Plainly: the episteme conforms to this exact viewpoint.
Its only two participants are independently identified before the test:
- candidate episteme
E : U.Episteme; - viewpoint episteme
P : U.Viewpoint.
EpistemeViewpointConformanceRelation(E,P) obtains exactly when:
- E is one independently identified episteme and P is one independently admitted viewpoint episteme;
T := EntityOfConcern(E)is recovered only from E's C.2.1 constitution;- T satisfies P's fixed
EntityOfConcernKindCriterionthrough the cited public durable-kind membership rule or the direct identity and membership rule of one exact independently admitted local kind; an exact C.3.2KindSignatureedition and one exactU.ContextSliceare separate test inputs only when P's local membership test needs them; - E has at least one independently admitted episteme kind referenced by P's admitted-kind claims, excluding
U.Viewand every kind whose membership depends on this same conformance; and - E's fixed claim content, interpreted under its effective reference scheme, satisfies P's fixed concern-coverage, semantic-form and consistency rules, including each completeness rule and each admitted omission or loss condition named by P.
T is recovered from E, not guessed from a use qualifier, topic, P, label, reference spelling, or evaluator input, and it is not a hidden third participant. Changing T changes E.
Changing P's target criterion, admitted target kind or cited membership rule, any exact KindSignature edition or U.ContextSlice named as a separate test input, admitted-kind claims, or conformance rules changes P.
State the result immediately after the five tests:
- positive: all five conditions hold, so the pair-determined positive relation occurrence obtains and the same E is a
U.Viewrelative to P; - negative: at least one evaluable fixed condition fails; name that condition, do not mint a negative relation occurrence, and do not claim
U.Viewmembership through P; - unresolved: missing or ambiguous E identity, P identity, kind criterion, local sense, or interpretation prevents evaluation; name that missing condition and claim neither positive nor negative conformance.
Ordinary stopping rule. Stop with that readable result when the next work needs neither an exact occurrence designator nor warrant. Add an occurrence designator, assertion episteme, evaluation episteme or local result value, evidence path, work record, or decision-use episteme only for the named receiving need. A readable assertion is distinct from occurrence identity.
For fixed E and P, one positive occurrence is participant-determined by <E,P>. Discovering, warranting, or using the judgment may involve a classifier, evaluation work, assertion, evidence path, result value, operational state, publication, audience, current use, or newly selected slice; none adds a participant or identity discriminator to the conformance occurrence.
If conformance could change while E and P remain fixed because another current object changed, route that condition to a separately identified adequacy or evaluation claim or reopen the relation architecture.
Conformance covers E's semantic content relative to P's fixed convention claims. Truth about T, decision fitness, stakeholder satisfaction, evidence-backed adequacy, publication usefulness, and operational usefulness remain separate evaluations. Evaluation never makes the direct predicate obtain or produces another occurrence for the same fixed pair.
Exact declaration and public designation of conformance
EpistemeViewpointConformanceRelationSignature is a separate RelationSignature episteme about the direct kind and declares exactly:
The declaration, SlotSpecs, references, and relation-participant designations alone do not establish that the relation obtains. Its positive occurrence is identified by the E/P pair under the fixed predicate. P remains the ordinary episteme about its C.2.1 EntityOfConcern; P is not this signature.
The F.18 NameCard for the direct conformance kind identifies its naming settlement and current F.17 public row:
The card, label, candidate list, and former placeholder are naming evidence only. None is relation admission, occurrence identity, or proof of obtaining.
Recognize the same episteme individual as U.View
An episteme is a U.View exactly when EpistemeViewpointConformanceRelation(E,P) obtains for at least one viewpoint P. This is same-individual dependent-kind membership of E under U.Episteme, not a second view individual, wrapper, form, carrier, result value, or identity discriminator.
One unchanged E may conform to zero, one, or several viewpoint editions through different pair-determined occurrences while remaining one episteme.
Direct authoring and A.6.3 construction—including identity viewing—are separate histories: either may be present or absent, and neither grants membership. Selection, transformation, bundling, naming, rendering, publication, audience, or current use also grants none.
Membership survives the end of reading, selection, use, evaluation, bundle membership, or publication.
P_old and P_new are different C.2.1 epistemes when they differ in fixed claims, effective scheme, or exact EntityOfConcern—the target kind in the self-contained branch or selected S in the structured branch; an obtaining EpistemeEditionRelation relates them but transfers no conformance. A current use may select P_new while unchanged E still conforms to P_old; adequacy and conformance for <E,P_new> are judged separately.
If E's claim content, EntityOfConcern, or effective scheme changes, C.2.1 identifies another episteme and its membership is judged anew.
Author or revise a reusable viewpoint only when existing P cannot serve
New-viewpoint authoring has two branches. Use one self-contained viewpoint episteme P by default. Open the convention-structure branch only when separately versioned convention components and their organization change reuse, comparison, maintenance, or another named action independently of P's fixed conformance claims.
Two viewpoint-authoring tasks.
The first row sets the ordinary architecture. The second demonstrates the narrower case in which structure pays for itself. Formality, assurance, or the wish to make a diagram does not trigger the second branch.
Author the smallest self-contained viewpoint episteme
Start from what a reader must be able to decide or explain from a description made under P. Which subjects can that question concern? Which changes to a subject or its use would require a different conclusion? Use those answers to choose the target and concerns and to justify which omissions a conforming description may admit (C.39.SA:4). Identify the independently admitted target kind K_target that the candidate epistemes' EntitiesOfConcern must satisfy. For a local kind, recover its candidate domain, operative membership condition, intended member/non-member boundary, and continuity rule. Use a practice or source boundary only to find or compare that membership rule. Use an exact C.3.2 KindSignature edition and U.ContextSlice only as separate inputs when the membership test needs them.
Constitute P under C.2.1 as <ClaimGraph(P), K_target, ReferenceScheme(P)>; the target kind is P's EntityOfConcern because P states how epistemes about members of that kind are to be read. P's fixed ClaimGraph:
- states the target-kind criterion and cites its direct authority;
- names stakeholder or audience referents only when they change the concerns, and states the concerns;
- names the independently admitted episteme kinds allowed for candidate E;
- states fixed concern-coverage, semantic-form, completeness, consistency, omission, and conformance rules without circular use of
U.View; and - states the describing-use frame and fixed applicability qualifiers needed to interpret those rules.
The same episteme P is admitted as U.Viewpoint when those five claim-content conditions hold under its effective scheme. No parent U.Signature, C.13 collection, Q, selected S, A.22 work, organization record, or evaluation result is required. Changed P claim content, target-kind EntityOfConcern, or effective scheme identifies another P edition; packaging, publication, evaluation, representation, or current-use selection does not.
Add a viewpoint-convention structure only when it changes action
Use the structured branch only when at least two convention components remain independently identified or versioned and their obtaining dependencies or organization change a named reuse, comparison, maintenance, or joint-interpretation action. Mere decomposition, citation, co-membership, a graph, or future possibility is insufficient.
Construct C_viewpoint under C.13 from the exact constituent episteme editions. The collection may be heterogeneous: its invariant is exact constituent identity, not uniform declaration power. Give each constituent the least-powerful independently admitted kind that carries its claims.
Preserve every exact edition. A concern question, kind citation, or one-use rule acquires none of SubjectKind, RangedValueKind, Vocabulary, Laws, or Applicability merely to fit a common table. Conversely, a constituent that independently is a reusable relation declaration, kind declaration, or method description keeps that stronger admitted kind. Collection position grants no convention job and no stronger membership.
For this branch:
- identify the least-powerful constituents above;
- construct
C_viewpointfrom those editions under C.13; - recover each selected direct relation occurrence using the pattern that defines its obtaining test and occurrence identity;
- identify ordinary constraint episteme
Q_orgaboutC_viewpointand the admissible describing-use frame; - have an exact system use the applicable A.22 selection method over C, selected obtaining occurrences, applied Q constraints, and the use frame, yielding exact
S_viewpoint; and - constitute P under C.2.1 with
EntityOfConcern(P)=S_viewpointand the same five fixed claim-content conditions from §4.6.1.
In this branch, changed P claim content, S, or effective scheme identifies another P edition. S is not U.Viewpoint; P is the claim-bearing individual. No viewpoint record, wrapper, organization object, context entity, bundle position, package ID, publication grouping, or parent U.Signature grants membership.
When the action-changing trigger disappears, use or author self-contained P rather than preserving C/Q/S as ceremonial structure.
Keep explicit evaluation values optional
The fixed E17ViewpointSemanticsSlice@FPFEdition selects the exact FPF and E.17.0 declaration editions, effective ReferenceScheme, and Γ_time. In that slice the admission predicates defined here permit exactly two optional C.3.2 local ValueKinds, each carried by its own C.2.1 KindSignature episteme:
The four values remain distinct from their designators. Each KindSignature declaration states its admitted value domain and exact-equality predicate with a fixed interpretation under C.2.3. That predicate determines membership. The kinds have no SubkindOf, and definedness fails closed. Incomplete evidence or interpretation leaves the optional evaluation unsupported or undefined; it supplies neither a negative result nor a third unknown member.
Omit both local values from P, direct relation obtaining, and—when the structured branch is active—Q_org and structure identity unless the named consumer needs one. Without such a consumer, state the direct conformance judgment or the structured branch's Q_org constraint judgment.
In the structured branch, state Q_org and select S without hidden organization
Q_org is one ordinary C.2.1 constraint episteme with EntityOfConcern(Q_org)=C_viewpoint. Its ClaimGraph carries the applied semantic constraints under its effective reference scheme and the named admissible describing-use frame. Q is not C, a selected relation occurrence, S, P, a result value, Signature, MethodDescription, organization record, actor, or method.
When the structured branch is triggered, Q carries these eight organization constraints by value:
- One target criterion. Select exactly one
E_targetby its claim content and cited target-kind membership rule; a raw kind label, viewpoint name, or collection position proves neither selection nor membership. - Concerns depend on the target. Every
E_concern[i]depends onE_target. When stakeholder attribution changes the concern, cite one exact stakeholder referent recovered as an independently identified System, local system-role kind, obtaining system-role assignment, collection-as-whole, or another exact local kind whose members are the concern referents. A responsibility claim remains a separately defined direct relation and never follows from the kind or assignment. - Coverage depends on the concerns and claim families it evaluates. Each coverage constraint depends on the concern constituents and claim families it evaluates; a heading, graph edge, unresolved family label, or coverage result is neither the dependency nor proof of coverage.
- Semantic form depends on the admitted kind. Each semantic-form constraint depends on the independently admitted-kind constituent to which it applies; notation, form, or a raw kind reference grants no admission or dependence.
- Method conventions depend on the method descriptions they use. Each method-based convention depends on one
D_method[i]whose EntityOfConcern is an independently admitted A.3.1 method. The raw method remains outside C, and description, method, dependence, and performed work remain distinct. - Completeness, consistency, and omission name their subjects. Each such constraint depends on the concern or claim components it constrains and names any admitted omission condition by value; a bare status or whole-P label is insufficient.
- Resolution does not establish a relation. Resolve every designation and reference under the effective scheme, while keeping spelling equality, lookup, graph adjacency, compatible schemes, token presence, and reference resolution from counting as direct-relation obtaining.
- No circular view admission. No admitted-kind constituent may depend on
U.Viewmembership or the same conformance judgment being established. Every mutually dependent group needs one named joint-interpretation method or fixed-point criterion.
Replay mutually dependent groups through stratified or witnessed joint/fixed-point semantics. Without that witness, the candidate fails the A.22 selection criterion for the named use. A graph, strongly connected component, iteration syntax, or fixed-point diagram is at most a C.29 representation of already judged occurrences and semantics; it is not the witness, criterion, or selected structure.
An identified System uses the applicable A.22 structure-selection method over C, the obtaining occurrences r_1,...,r_n, the applied Q constraints, and the admissible-use frame. The symbols r_1,...,r_n are local notation, not an O object or collection kind. The selection yields S under A.22; C remains the C.13 collection, and each r retains the predicate and occurrence identity defined by its relation pattern.
Identity and change stay local:
- Q changes only with its claim content, C as its EntityOfConcern, or effective reference scheme; another graph, form, carrier, representation, or publication leaves the same Q edition unchanged.
- Replacing a selected obtaining occurrence changes the organization used to identify S. Replacing only its assertion, occurrence description, D, J, result, production or use relation, provenance, or graph leaves that occurrence unchanged, although use-specific admissibility may need reevaluation.
- S changes when C, any selected obtaining occurrence, the applied semantic constraint set, or the admissible-use frame changes. Replacing only Q while those discriminators remain semantically unchanged leaves S unchanged.
- In the structured branch, P changes only with its fixed claim content, S as its EntityOfConcern, or effective reference scheme. In the self-contained branch, the target kind replaces S as that discriminator. P is neither its EntityOfConcern, a reference, a bundle position, a publication object, nor an evaluation result.
Resolve P's target criterion, admitted kinds, coverage, semantic-form, completeness, consistency, and omission rules through exact constituent claims and selected obtaining occurrences cited by P. Do not leave them as untyped fields, mandatory Signature constituents, or graph edges treated as occurrences.
In the structured branch the selected public individual is episteme P about S. These nearby alternatives remain rejected:
- S is not
U.Viewpoint: consumers require the claim-bearing edition P, whileEntityOfConcern(P)=S. - An episteme about one method is a neighboring
U.MethodDescriptiononly when M and that description independently pass A.3.1 and A.3.2; it is not the viewpoint genus. A method-description constituent does not retarget P from S to M. - No viewpoint record, wrapper, organization object, context entity, or non-entity value is needed; P, S, C, and selected relation occurrences already exhaust the identity-bearing objects.
- A catalogue or local family-declaration position, catalogue edition, package ID, or publication grouping does not constitute P or grant membership.
- P requires no parent
U.Signature, is not a public C.3 local kind, and is notEpistemeViewpointConformanceRelationSignature. A reusable kind declaration, a local-kind classification judgment, and a direct-relation declaration are different jobs with different subjects.
U.Viewpoint is therefore the same P under the complete positive predicate above.
Author progressively and stop at the needed assurance
Use as much of the authoring path as the new viewpoint needs. For self-contained P, identify target kind, constitute P with the five fixed claim-content conditions in §4.6.1, apply the positive viewpoint-membership predicate, and mint or reuse U.ViewpointRef. For the structured branch only:
- identify every constituent edition and state each proposed dependent-to-base claim readably;
- resolve both endpoint designations, apply the direct obtaining criterion, and construct C from those editions under C.13;
- add D only for a named A.22 selection-use claim, and J or evaluation only when that receiving use needs the additional assurance;
- apply the Q constraints and have an identified System use the applicable A.22 selection method over C and the selected obtaining occurrences, producing S; and
- identify ordinary episteme P about S, apply the positive viewpoint-membership predicate, and only then mint or reuse
U.ViewpointRef.
Citation, collection membership, graph adjacency, and displayed edges never close step 2. Selection identifies an existing selected object; it does not construct another constituent episteme. Viewpoint authoring requires neither five fixed stages, one composite method, empirical/formal evaluation, nor J.
Identify every cited method under A.3.1 and use B.1.5 only when an order-sensitive method whole independently obtains. Stop as soon as the named receiving use is served; add no assurance artifact merely because a longer path exists.
Keep viewpoint-convention dependence direct
Use ViewpointConventionDependencyRelation(E_dependent,E_base) only when interpreting or replaying the fixed claims of dependent constituent episteme E_dependent depends on an exact criterion, law, public name, or method claim carried by base constituent episteme E_base, and replacing that base edition or making the content used by the dependent episteme unavailable can change the interpretation or replay. It is the A.6.6 base-dependence case specialized to viewpoint-convention constituents.
Citation, co-membership, reference resolution, compatible schemes, or a graph edge alone does not establish this predicate. For fixed endpoint editions, one positive occurrence r is participant-determined by <E_dependent,E_base>. Scope, time, status, evaluator, evidence, result, use, selection, representation, and publication are neither participants nor occurrence-identity discriminators.
ViewpointConventionDependencyRelationSignature is a separate RelationSignature episteme about the direct relation kind. It declares exactly:
The SlotSpecs declare reusable participant meanings and polarity. Their declaration does not make the relation obtain or identify an occurrence.
The current A.6.6 vocabulary resolution chain is viewpointConventionDependsOn -> current vocabulary entry -> ViewpointConventionDependencyRelationSignature -> its EntityOfConcern, ViewpointConventionDependencyRelation. The NameToken, its separate NameCard, vocabulary entry, signature episteme, direct kind, and occurrence remain distinct; spelling or citation proves none of them equivalent and makes no occurrence obtain.
Local designation of the direct relation kind
Within this pattern, ViewpointConventionDependencyRelation designates semantic dependence of one exact viewpoint-convention constituent episteme on one exact base constituent episteme, dependent first. Replacing the base edition or losing its exact used content can change interpretation or replay. The Plain designation is “this viewpoint-convention constituent depends on that exact base constituent.” The signature above retains the participant meanings and polarity.
Add only the neighboring object the receiving use needs
The compact positive statement may stop at “this exact constituent depends on that exact base constituent.” Add the following objects only under their positive trigger; do not flatten them into one witnessed-base record or add their fields to the two-participant relation.
D_dependencyUse is therefore the C.2.1 episteme identified through obtaining EpistemeConstitutionRelation(G_dependencyUse,r,S_decl). The ordered triple names the ClaimGraph, EntityOfConcern, and effective ReferenceScheme participants; it is not a self-constituting card or record and does not make the relation obtain.
When the structured branch is active, G_dependencyUse designates r and the receiving A.22 use: C_viewpoint, the Q_org constraints applied, and the named admissible-use frame. It carries two separate claim values:
c_dependencyObtains: the direct predicate obtains, independently of use and evidence;c_dependencyAdmissibleForSelection: r is admissible among candidate organizing occurrences for that named use frame.
Both are claim values in G, not C.2.1 epistemes, occurrences, or decision results. Changing the use frame can change the second claim while r remains unchanged.
Add U.ClaimScope or a time qualification to G only when it changes the represented claim; neither becomes a participant. Cite the exact current A.6.6 vocabulary entry and exact RelationSignature as declarations, not as r or proof of r. D is reidentified only when <G_dependencyUse,r,S_decl> identifies a different claim graph, relation occurrence or effective reference scheme; a changed claim value changes D only through changed constitutive G.
When J is present, keep separate conclusion nodes for the two claims and at least these distinct premises when the justification relies on them:
E_baseunderS_basecarries the criterion, law, public name, or method claim used to interpret or replayE_dependent;- an identified System in the interpretation or replay Work, enacting an admitted method, resolves and applies that base content to
E_dependentunderS_dep; and - replacing base edition
E_baseor making the base content used byE_dependentunavailable can change interpretation or replay of fixedE_dependent.
Designation, citation, graph location, co-membership, scheme compatibility, version difference alone, or a failed lookup supplies none of those premises. If the interpretation is method-dependent, cite the exact U.MethodDescription, but identify the acting system, admitted method, and work occurrence separately.
Keep empirical and formal evaluation local
When empirical interpretation or replay testing is current, identify separately:
H_dependencyEvaluator : U.Systemunder A.1 as performer;RA_dependencyEvaluator : DependencyEvaluationWorkAssignment <: U.SystemRoleAssignmentunder A.2.1, withH_dependencyEvaluatorinHolderSystemSlot, declaration-local assigned-kind domainDependencyEvaluatorSystemRoleKindDomain, andDependencyEvaluatorSystemRoleas RA's assigned-kind value admitted by that domain; the value, assignment, holder System, and Work remain distinct, and neither the value nor assignment acts;M_dependencyTest : U.Methodunder A.3.1 and, when needed,D_dependencyTest : U.MethodDescriptionunder A.3.2; D describes M but is neither method, work, RelationSignature, nor OperationAlgebra, and a separate A.6.1 operation declaration is cited only when typed application is current;W_dependencyTest : U.Work: A.13 first recovers H as the actual performer through the already named obtaining RA; A.15.1 independently admits W as enacting M. Add F.6 only when the receiving claim needs precise attribution of W through that same RA. An attribution failure alone leaves independently admitted W intact; if the finding invalidates a consumed performer or Work-admission premise, recheck that premise and its dependent conclusions;B_dependencyEmpirical, a C.2.1 episteme identifying the model, calibration, assumptions, and interpretation basis; and- result episteme
T_dependency = <G_dependencyTestResult,E_dependent,S_test>, whose ClaimGraph designatesE_base, predicate, method, conditions, basis, and positive or negative result.
Establish actual participation of E_dependent, E_base, each parameter, and B_dependencyEmpirical during W only through the exact relations that define those participation positions or A.6.1 operation-application bindings. A MethodDescription or compatible SlotSpec establishes no participation. Open a local A.15.PROD claim only when the receiving use needs to say W first constituted T or later completed its declared production; inception, completion, episteme identity, and dependency obtaining remain distinct.
When formal interpretation is current, constitute formal-evidence episteme E_dependencyProof = <G_dependencyProof,E_dependent,S_proof> and B_dependencyFormal identifying the theory, axiom set, proof semantics, and interpretation basis. Its ClaimGraph designates E_base, proof obligation, formal method, basis, and result. Preserve entailment, refutation, malformed input, timeout, and checker failure as different outcomes; neither a refutation nor a checker failure fabricates positive r. The proof episteme is distinct from r and its participants.
If reusable target claims are needed, constitute them separately under C.2.1:
C_dependencyObtainshasc_dependencyObtainsas its principal claim and concerns the exact endpoint pair and predicate;C_dependencyDoesNotObtaincarries a distinct negative principal claim and is not a state of the positive episteme; andC_dependencyAdmissibleForSelectionconcerns r under the named use frame and remains distinct from both obtaining claims.
Co-representation in one ClaimGraph does not merge these epistemes. T carries its empirical conclusion locally; E_dependencyProof carries its formal conclusion locally. If a target-claim episteme separately represents one conclusion, recover its correspondence to that conclusion; add C.29 when mathematical inference or transfer matters. That correspondence does not establish the conclusion's truth, its use, or r. Mint no duplicate evidence-bearing relation and no new A.10 ontology.
Keep these three cases distinct:
- r obtains while support for
c_dependencyObtainsis unknown; a selecting system may decline reliance without deleting or reidentifying r; - a negative empirical or formal result may support
C_dependencyDoesNotObtainwithout presupposing r, fabricating D, or becoming a positive occurrence; and - T may support the claim that r obtains without supporting use-specific admissibility; a later decision method may consume empirical and formal result epistemes in separate declared premise slots and produce a separate C.11 result.
For historical reliance on a claim or result, recover its obtaining premise, decision-use, reference-use, or operation-argument relation for the named bounded use. Use A.10 when the source-to-use account is implicit or must remain replayable.
Recover the Work and its enacted Method only when the current claim concerns that dated Work or the expressly selected empirical evaluation branch above requires them. Retain that branch's A.13 performer basis, independent A.15.1 Work admission, and F.6 when precise assignment-bound attribution is consumed.
Storage, inspection, citation, attachment, production, graph membership, or adjacency alone does not establish that use. Keep empirical and formal algebras distinct; keep provenance and assurance with A.10, G.6, and B.3. Retain a missing-governor blocker instead of inventing a generic evidence, use, or acceptance relation.
Schemes, scope, transformation, and change
Recover S_dep from E_dependent, S_base from E_base, and S_decl from D. They are three uses of existing ReferenceScheme, outside r and its RelationSignature. G may designate exact endpoints, claim values, and declared names through those schemes; designation is neither occurrence obtaining, truth, nor historical participation.
Keep claim-scope widen, narrow, and refit under A.2.6 when no local-sense translation is needed. Use translate only when scope membership must be expressed between exact local senses: require an obtaining F.9 Bridge between their exact SchemeSenseCell values, the separate affirmative claim for that translation's direction, rule, and tolerance, and the current A.10 or B.3 reliance branch. Scheme difference, same spelling, token reuse, or translation intent triggers no Bridge.
Use A.6.3.RT when the dependency use depends on a changed representation language or reasoning medium. Recover exact source and receiving epistemes and their construction, preservation, loss, and admissible use when that reliance-facing use requires them.
When the receiving use depends on the transition occurrence, apply RepresentationTransitionRelation@Context. Recover the six participants separately. They are the independently selected BoundedModelUseStructure : U.Structure, the preserved EntityOfConcern, source episteme X, receiving episteme Y, an episteme describing the effective interpretation rules of X, and one describing those of Y. Establish the actual representation-transformation Work as required by A.6.3.RT:4.1.b. The six participants determine the occurrence’s identity. The Work is part of the test that the occurrence took place. @Context is the retrieval label for that A.1.1 use. Apply A.6.3.RT’s conditions for preservation, loss or recoverability. Add a description of the transition only when the receiving use needs it.
Add C.29 only for a current mathematical lens. Use F.9 separately when the claim needs a Bridge between distinct local senses. If interpretation or the proposed dependency use lacks a required basis, name that gap and block only that use; the absence of a six-participant transition occurrence or Bridge by itself blocks neither the direct dependency judgment nor a supported exact construction.
Changing only J, an assertion or occurrence description, evaluation result, basis, provenance, production, later-use relation, or representation leaves r unchanged while its endpoint pair is fixed. It also leaves D unchanged while <G_dependencyUse,r,S_decl> is fixed. Unknown support does not make an obtaining r non-obtaining, and support for a negative claim creates no positive r. A changed representation transition invalidates judgments that depended on that transition, but changes r only when an endpoint episteme or the direct predicate also changes.
Progressive stopping rule. Add only what the receiving use requires: readable dependency assertion; reusable RelationSignature when declaration reuse matters; D only for a named A.22 selection-use claim; J only for inspectable inference; evaluation work and exact participation only when evaluation is current; local A.15.PROD only for a needed result-inception or completion claim; provenance, assurance, representation transition, mathematical lens, scope translation, and Bridge only at their own triggers. Adding one of these objects does not by itself prove that an underlying relation obtains.
Keep selection for one describing use separate
For one current describing use, name the question the reader must answer. When you are to choose the viewpoint, start with a difference that could change the reader’s action or its explanation. For each kind of holon whose composition matters, identify the whole and constituents on which that difference depends. At each needed scale, ask which cases the description may treat alike. Choose a viewpoint whose concerns require the necessary distinctions (C.39.SA:4). For a claim that model C strictly refines model A, fix the inquiry and interpretation. Establish that C has an admissible completion and that every such completion, read in A’s terms, satisfies A’s assertions. Also identify an answer admitted by A that no completion of C can give (B.5.MR:4.2).
One description may supply the entire answer. When it takes several, trace each needed conclusion to what the descriptions establish for the original question. Identify what relates their subjects and conditions. If their claims must hold together, establish that they concern compatible objects or situations. If the descriptions express alternative readings, check what follows on each. In a question about joint work, a change in one participant’s choice can alter what another must do for the common result. Follow that dependence through the descriptions the participants can use. A similar label on both pages can leave it invisible. Use C.37.MM:4.1.1 where the inference needs a correspondence. If the commission asks only whether E conforms to a specified P, §§4.1–4.5 give the complete test. Add one singular viewpointRef : U.ViewpointRef only when selecting P changes what the use reads or checks or what a relying use may conclude; otherwise omit the reference. When present, resolve it under the effective reference scheme to P. ViewpointId is P's designator; designator, reference, episteme, and describing use remain different objects.
When the viewpoint matters, select P only for that describing use. The selection does not establish conformance or U.View membership, give E a new C.2.1 identity, reidentify E, or create a universal selection relation, legacy context tuple, bounded-context object, or generic model-use identity field. Another use may select another P while E remains unchanged. A use needing several viewpoints first identifies the C.13 collection and its exact membership; it does not overload viewpointRef with a collection value.
The architecture therefore keeps exactly two positive dependent-kind rules—P as U.Viewpoint by its fixed self-contained content about an exact target kind or, conditionally, by fixed content about selected S; and E as U.View by obtaining conformance—and two direct relation kinds: viewpoint-convention dependence and E/P conformance.
D remains optional for a named A.22 use; the two local explicit-result ValueKinds remain optional for named evaluation consumers. Families carry exact U.ViewpointRef values only when a named use needs viewpoint selection. C.2.1 identity, MethodDescription, A.6.3 construction, E.24.PUB publication, C.29 mathematical representation, and unrelated interfaces retain their separate identities and rules.
Add viewing construction only when its history matters
A.6.3 defines a viewing relation from a source episteme to a separately identified receiving episteme. It preserves the same EntityOfConcern. Claim content and the effective reference scheme may be preserved or changed only within A.6.3's declared construction law. If the EntityOfConcern changes, the move requires A.6.4 rather than counting as viewing construction.
Keep these claims independent:
- constitution: C.2.1 identifies the receiving episteme;
- construction: A.6.3 states an obtaining source-to-receiving viewing relation when one exists;
- membership: E.17.0 states whether the receiving episteme conforms to the identified viewpoint;
- work: A system may perform query, authoring, or rendering work;
- production: use A.15.PROD only when a local work/change/entity-identity-inception or completion claim about the receiving episteme is current.
Do not infer one claim from the label generated view.
Recover multi-view organization and correspondence only as needed
Several conforming views do not automatically form one new entity. For ordinary comparison, exact view epistemes, exact viewpoint epistemes, and their conformance occurrences can remain a plurality.
When the work depends on the collection as a whole, construct it under C.13. When it depends on an organization among those views, recover the exact direct relation occurrences and select one U.Structure under A.22. A package, table, graph, or shared EntityOfConcern is not that structure by appearance.
When cross-view correspondence matters:
- name the participant epistemes or represented entities;
- state the direct correspondence, consistency, realization, trace, or change-impact relation that is claimed;
- apply the concrete pattern that defines and tests that relation, including its obtaining and occurrence-identity rules;
- identify a C.2.1 assertion or description episteme only when the correspondence claim itself must be reviewed or used;
- recover the correspondence by which a graph, matrix, or diagram represents those objects and relations; use A.6.3.RT/A.6.3.RT.OE for its construction and C.29 when a mathematical lens is required.
Plain correspondence model may describe such a claim-bearing episteme after its EntityOfConcern and direct relations are recoverable. It is not a universal U.CorrespondenceModel kind, a substitute for the relations, or proof that they obtain. If no pattern defines and tests the needed direct relation, report the missing definition or test for that relation, or use A.6.RCD; do not close the case with linked, mapped, or consistent.
Temporary inconsistency is represented by exact evaluation claims and, when current, repair work. It does not silently weaken the conformance predicate or erase an obtaining correspondence relation.
Keep publication and conceptual form outside view identity
E.24.PUB keeps three direct relation occurrences distinct:
PublicationExpressionRelation(selectedEdition,expression,boundedUseDeclaration)states that X conveys the required claims of E for U;PublicationExpressionBearingRelation(presentationCarrier,expression)states that an actual state or realization of independently identified C yields that same X; andEpistemePublicationRelation(selectedEdition,audienceDeclaration,boundedUseDeclaration,expression,presentationCarrier)makes E available through X and C to the declared audience for U while expression, bearing and audience access obtain.
Expression has the three named participants, bearing the two named participants, and publication the five named participants. Each occurrence retains its own maximal continuous obtaining or availability interval. Changing a participant identifies another occurrence; an availability gap followed by restoration creates a later publication occurrence. None of those changes reidentifies an otherwise unchanged C.2.1 episteme.
When reusable form F matters, establish X/F conformance. E.24.PUB retains form-mediated expression and bearing with the same-X witness, and broader availability over permitted expressions or carriers. A changed exact publication need not interrupt that broader availability.
Rendering, upload, or carrier manipulation is U.Work only when an exact system performs it. A.6.3.RT and A.6.3.RT.OE govern construction and correspondence of expressions; C.29 supplies the additional mathematical-lens conditions. A form, carrier, representation, rendering, or publication occurrence grants no U.Viewpoint or U.View membership and makes no world-side relation obtain.
In this pattern’s Plain wording, published view means an already recognized view episteme participating as the selected edition in an exact publication occurrence. It is not another durable kind. One unchanged view may participate in several publications through different audiences, uses, expressions, carriers, and availability intervals.
Worked cases
Directly authored architecture view
Architecture episteme E concerns system T. Maintainability viewpoint episteme P concerns its selected viewpoint-convention structure and states the target-kind, concern, admitted-kind, coverage, and semantic-form rules. E satisfies those fixed rules, so EpistemeViewpointConformanceRelation(E,P) obtains and the same E is a U.View. No source episteme or A.6.3 viewing is required.
Query output that is not yet a view
A query over source episteme X constructs episteme Y, and A.6.3 records the source-to-Y viewing relation. Y omits a concern component that viewpoint P requires. The construction relation obtains, but conformance does not; Y is not a U.View under P. A later repair may create Y2 with different claim content and a new C.2.1 identity.
One episteme, two viewpoints, one selected use
Unchanged episteme E conforms to safety viewpoint P1 and maintenance viewpoint P2. Two participant-determined conformance occurrences obtain, while the named current review use selects only P1 through one viewpointRef. E remains one U.View; the selection neither creates the P1 conformance nor removes the P2 conformance.
Viewpoint revision and library repackaging
Adding a reference to unchanged viewpoint episteme P to another E.17.1 local family declaration, or carrying it in another catalogue edition, changes only the catalogue declaration and provenance; it does not change P. Revising P's conformance rules creates another episteme P_new; conformance of E to P_old does not imply conformance to P_new. An EpistemeEditionRelation may relate the P editions, but it is not a conformance occurrence.
Two publications of one view
View episteme E conforms to P. A web publication uses expression X1 following reusable form F1, borne by carrier K1; a printed publication uses X2, F2 and K2. Expression, bearing and selected form conformance obtain for each. Two exact publications make the same E available under their own audience, bounded-use and maximal availability intervals. E remains one view episteme. Broader availability of E can continue when an alternative expression or carrier replaces one of these exact publications.
Cross-view correspondence
A functional view names transformation F and a structural view names module M. A project claim says M realizes F. The shared system EntityOfConcern and aligned diagram positions do not establish realization. Identify F and M, apply the direct realization-relation pattern, then identify an assertion episteme about that occurrence if review needs it. A traceability matrix may represent the assertion and occurrence through an explicit correspondence; C.29 applies when that use requires a mathematical lens. Its cell is not the realization relation.
When a procedural view also describes a Method
A team reads a procedural episteme E about its factory to learn how the factory recovers after a stoppage. E is a U.View if it conforms to the selected procedural viewpoint P. E mentions the restart Method in its account of the factory. Its EntityOfConcern is the factory.
In a second case, an engineer writes an episteme M about the admitted restart Method itself. M qualifies as a U.MethodDescription when it describes how that Method is carried out under A.3.2. For a view claim, select a procedural viewpoint Q whose target criterion admits the Method and check whether M conforms to Q. When both conditions hold, the same M is a U.View under Q and a U.MethodDescription. If M omits a recovery branch required by Q, it can still qualify as a MethodDescription. A claim that a crew actually restarted the factory needs evidence of the work.
Consequences
Reopen the pattern when either conformance participant kind changes, the fixed predicate changes, or a proposed condition makes occurrence identity depend on an object other than E and P. Reopen the affected use when its candidate episteme, viewpoint edition, selected describing use, direct correspondence, or publication occurrence changes.
Rationale, lineage, and current FPF basis
ISO 42010 supplies vocabulary lineage. The remaining rows explain FPF design choices using its construction, representation, relation, evaluation and work rules. They supply no external state-of-the-art evidence for those choices.
Relations and contribution boundaries
- C.2.1 identifies episteme, claim-content, EntityOfConcern, scheme, and edition identity for P, E, D, and any optional assertion, description, result, basis, or target-claim episteme. E.17.0 adds dependent
U.ViewpointandU.Viewmembership to those same individuals. - Use C.13 to construct
C_viewpointonly in the action-changing structured-viewpoint branch, and any separately needed collection of selected viewpoints or views. - A.6.6 defines the reusable
viewpointConventionDependsOnvocabulary entry; A.6.5 declares the four SlotSpecs inside the two RelationSignature declarations. E.17.0 defines direct dependency and conformance obtaining tests and positive occurrence identity. - C.3.2 admits the two optional local explicit-result ValueKinds and any exact local target or stakeholder KindSignature; their values do not determine direct judgments.
- Use A.22 to select
S_viewpointonly when separately versioned convention organization changes a named action, and to select any separately current multi-view structure. E.17.0 supplies Q and candidate relation occurrences for that branch. - A.6.3 defines optional source-to-receiving viewing construction, including identity viewing; it does not define view membership.
- E.10.D2 defines description epistemes and specification use. A describing use is always named; it selects a viewpoint only when that choice changes reading, checking, or a permitted conclusion. Selection does not establish conformance.
- F.18 governs the two relation-kind designations. F.9 applies when a separately needed semantic relation between distinct F.17
SchemeSenseCellvalues is at issue. - E.17.1 defines exact catalogue epistemes and local family declarations whose members are exact
U.ViewpointRefvalues. E.17.2 supplies the four-position project-local engineering viewpoint authoring template and, only after local materialization, its four exact bindings. A declaration, template position, or reference grants no viewpoint or view membership. - E.24.UK admits dependent
U.ViewpointandU.Viewonce for public use; E.17.0 supplies their stable positive membership predicates. - E.24.PUB defines exact expression and bearing, selected form conformance, form-mediated availability and publication recurrence. A.6.3.RT and A.6.3.RT.OE govern expression construction; C.29 governs a mathematical lens. Their use retains the independent predicate for any represented subject relation.
- A.1, A.2, A.2.1, A.3.1, A.3.2, A.15.1, and F.6 distinguish performer Systems, local system-role kinds, exact assignments, Methods, MethodDescriptions, Work, and attribution. Responsibility remains under its direct predicate. Use A.15.PROD only for a separately needed local inception or completion claim.
- A.10, G.6, and B.3 retain provenance and assurance. Use A.6.RCD when no pattern defines a needed cross-view or use relation.
Conformance checklist
- Candidate E has recoverable C.2.1 claim content, EntityOfConcern, and effective reference scheme.
U.ViewpointRefresolves to one viewpoint episteme P; designator, reference, P, P's exact target-kind or structured EntityOfConcern, and any catalogue position remain distinct.- Self-contained P has the exact admitted target kind or exact local-kind subject as EntityOfConcern and carries the complete fixed conformance test in its ClaimGraph. Only the action-changing structured branch uses an A.22-selected
S_viewpointover least-powerful exact constituent editions and obtaining relations, with Q carrying the eight organization constraints. - Every dependency occurrence has only the identified dependent and base epistemes as participants; assertion, description, D, J, evaluation, work, scope, time, scheme, representation, publication, and use remain conditional neighbors.
EpistemeViewpointConformanceRelationhas exactly E and P as participants, the fixed five-condition semantic predicate, and pair-determined positive-occurrence identity.U.Viewmembership follows only from an obtaining conformance relation, never from authoring, identity viewing, query, selection, packaging, form, carrier, rendering, or publication.- The current describing use is named. A singular viewpoint reference selects P only when that choice changes reading, checking, or a permitted conclusion; omission otherwise changes neither episteme identity nor conformance. Multi-selection uses a C.13 collection with exact membership.
- Optional local result values, evaluation, evidence, occurrence designation, decision-use D, and J exist only for a named receiving work or decision need; unsupported evaluation is not a third or negative value.
- For every multi-view collection, selected structure, or cross-view relation, apply the pattern that defines its identity or obtaining test; a table, graph, matrix, or shared subject proves none.
- Exact expression, carrier bearing, optional form conformance, five-participant publication availability and recurrence, rendering work, and mathematical representation under C.29 remain distinct from E, P, view membership, and every represented world-side relation.
- Ordinary use stops at a readable direct judgment unless a named consumer needs more structure; authoring stops at the shortest progressive path that produces P and its reference.
E.17.0:End
Viewpoint Bundle Library - Reusable Viewpoint Reference Bundles
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative
Tech-name. ViewpointBundleLibrary (pattern and catalogue form, not a U-kind).
Plain-name. Viewpoint bundle library.
Use this when. The same coherent family of already admitted viewpoint editions recurs across projects, schools, or publication uses, and users need one editioned catalogue from which they can import viewpoint references without restating or reidentifying the viewpoints.
First action. Resolve one already admitted catalogue edition L and its family designator, retrieve the local declaration, and resolve only the U.ViewpointRef members needed now. If L or the declaration is new, missing, or disputed, use §4.2 to recover <G_L, K_L, R_L> and verify L's C.2.1 constitution for that edition. Reuse that result while the edition, effective scheme, and relied-on premises stay unchanged.
First useful result. One catalogue edition L, one ordinary family designator retrieving a local declaration claim block, and one finite non-empty member set of U.ViewpointRef values that each resolve to an E.17.0 viewpoint episteme edition. L retains its C.2.1 identity. The compact locator <editionDesignator(L), familyDesignator> aids retrieval under R_L but is neither L's identity nor a separate bundle kind or entity.
Ordinary stop. Stop when L, the declaration, and the needed reference subset are recoverable. Do not reconstruct L's constitution, instantiate every member, select an A.22 structure, prove conformance, or publish the catalogue merely to import an admitted family.
Admission boundary. E.24.UK admits U.Viewpoint and U.View; it does not admit U.ViewpointBundleLibrary or U.ViewpointBundle. E.17.1 therefore defines an ordinary catalogue-episteme form and local bundle declarations in its claim content.
Do not use this when. A describing use needs only to select one viewpoint or a small one-off set that has no recurring family-level purpose. Keep the references local; collecting them in a bundle adds no conformance, U.Viewpoint or U.View membership, structure, publication, or correspondence.
What changes in practice. Authors reuse references and preserve their bundle provenance. Reviewers can detect silent member substitution, alias collision, and package-driven membership claims.
Builds on.
A.6.2-A.6.4 (episteme morphism classes), A.6.5 relation-declaration slot discipline, A.7, E.7, E.10, E.10.D1, E.10.D2, and E.17.0 MultiViewDescribing.
Used by.
E.17.2 (TEVB engineering viewpoint bundles), E.18:5.12, and domain-specific viewpoint families for architecture, governance, safety, research, or assurance.
Problem frame
Selected-family discipline. A local declaration states its target-kind compatibility condition as either a by-value criterion or a reference that resolves to the ClaimGraph defining or constraining the admitted target kind. Bundle labels, aliases, annexes, files, and publication faces never supply that criterion or select an actual entity by themselves.
MultiViewDescribing lets engineers recognize several epistemes about one entity as views under viewpoint editions and recover cross-view relations only when those relations obtain. Many viewpoint families recur across projects and schools. Engineering teams reuse functional, procedural, structural and interface viewpoints. Governance teams reuse risk, control, compliance and operations viewpoints. Research teams reuse theory, experiment, inference and limitation viewpoints.
E.17.1 provides a pattern for packaging reusable viewpoint families. With it, authors can import them, name them stably, review them once, and keep viewpoint-family identity separate from document labels, publication faces, and publication forms.
Problem
When recurring viewpoint families lack a stable catalogue:
- Each domain invents local viewpoint families.
Similar families reappear under slightly different labels, but no stable catalogue
U.Epistemerecords whether the underlying viewpoints are the same. - Viewpoint identity drifts.
A family called
functional,capability, oroperationalmay differ only in name or may differ in meaning. Without a stable catalogue, users have no systematic way to tell. - No catalogue entry to import. Each describing use that needs the family must restate it locally instead of importing an existing bundle.
- No reusable local declaration. Available viewpoints are known through external terminology or project lists, but no catalogue supplies the local references needed for repeated import.
- Reader-facing labels leak into semantics. Authors reuse the same name for viewpoints, views, publication faces, or folders, and the boundary between EntityOfConcern and Description episteme becomes unclear.
Forces
Solution - one catalogue episteme with local bundle declarations
E.17.1 defines a reusable form for one ordinary C.2.1 catalogue episteme L whose local bundle declarations package U.ViewpointRef values resolving to E.17.0 viewpoint episteme editions. L, a declaration claim block within L, its ordinary family designator, each reference, each viewpoint designator, and P remain distinct. Neither the catalogue nor a declaration redefines viewpoint identity or membership, grants U.View membership, or creates publication forms and carriers.
Core role
A conforming viewpoint-bundle library makes three things explicit:
- which family is being named, by an ordinary family designator interpreted under
R_L; - which
U.ViewpointRefmembers resolve to the viewpoint episteme editions packaged by that family; - which target-kind compatibility condition and catalogue-edition discipline constrain the family.
This lets MultiViewDescribing import a finite viewpoint family from a stable catalogue U.Episteme instead of restating it ad hoc in every local description family.
Reuse an admitted catalogue; open full constitution only when needed
Existing-catalogue route. Resolve the already admitted catalogue edition L, retrieve the local declaration by its family designator under L's effective R_L, and resolve only the member references needed now. Do not reconstruct L's complete C.2.1 constitution merely to import an admitted edition.
Open the complete constitution below for the affected catalogue edition when authoring or admitting a new L, when the identity of L or its edition, or the resolution of a reference, is disputed, or when a named later use needs the catalogue's ClaimGraph, subject, or scheme as inspectable premises. Reuse an existing check while that edition, its effective scheme, and the relied-on premises stay unchanged. The complete constitution has three participants:
G_L : NonemptyClaimContentis the catalogue's complete interpreted claim content (ClaimGraph), stating its scope, local family declarations, referenced viewpoint editions, target-kind compatibility conditions, and edition-change rule.K_Lis the catalogue subject: the independently identified finite C.13 collection of already admitted viewpoint episteme editions whose recurring reuse groupings L describes. Its collection identity, members, obtaining membership relations, and identity rule are established before L. Neither the catalogue nor a declaration creates them.R_L : EpistemeInterpretationSchemeis the effective rule collection under which the catalogue's ordinary library, edition, and family designators resolve; eachU.ViewpointRefresolves to P; target-kind criteria and compatibility claims are interpreted; and reference, omission, provenance, and edition-change rules are read. C.2.1:4.1.2 supplies both ranges; the catalogue requires the full content and rule coverage stated here.
An eligible replacement for R_L must also resolve the catalogue’s designators and interpret its criteria. Range membership alone supplies neither result. A missing rule leaves the affected retrieval or compatibility judgement unresolved. Each referenced P retains its own effective scheme and E.17.0 membership. Resolving its reference under R_L does not transfer the catalogue's scheme to P.
EpistemeConstitutionRelation(G_L, K_L, R_L) must obtain. The participant-determined triple <G_L, K_L, R_L> identifies catalogue episteme L. If a proposed catalogue has only a file, label, list, or card but lacks an identifiable K_L or an effective R_L, stop. L has not yet been constituted.
G_L makes at least these claims recoverable:
- one ordinary library designator and one ordinary edition designator interpreted under
R_L; - a finite set of local family-declaration claim blocks, each retrievable inside
G_Lby one ordinary family designator interpreted underR_L; - the
U.ViewpointRefmembers and target-kind compatibility claim for each declaration; and - only maintenance claims currently needed, using the branch that matches the present claim:
- for a current maintenance-System claim, cite the admitted maintenance
U.System; cite a local system-role kind and its independently evaluated classification only when that classification is current; - for actual maintenance Work, recover the actual performer through A.13 and let A.15.1 independently admit the dated
U.Work. Add F.6 only when the catalogue claim or receiving use expressly relies on precise assignment-bound attribution through the same obtaining A.13 assignment. F.6 identifies neither assignment nor performer. Missing or failed F.6 alone leaves independently admitted Work intact. A short catalogue claim may omit identifiers its bounded use does not need; - for current maintenance responsibility, cite its direct admitted predicate and actual participants or return the missing governor; assignment establishes no responsibility; and
- for prospective maintenance guidance, retain only the change-control note, intended maintenance condition or
U.WorkPlan, and scope tag; this content asserts no performed Work, current assignment, or responsibility.
- for a current maintenance-System claim, cite the admitted maintenance
The catalogue entry cites these values. Their definitions or constraints come from elsewhere, and the entry creates none of them.
Library, edition, and family designators are lexical values under R_L, not local ValueKinds, public U-kinds, episteme identity discriminators, or entities by spelling. A local family declaration is claim content in G_L, not by itself a separate entity or episteme. Its compact locator <editionDesignator(L), familyDesignator> is a retrieval aid under R_L; it does not replace L's C.2.1 identity.
If a receiving use needs one declaration as a separately identified episteme, constitute that new episteme independently under C.2.1 rather than inferring it from a row.
Normative constraints:
- Within one
G_L, every family designator SHALL retrieve exactly one local declaration claim block underR_L. - A catalogue SHALL NOT define new kernel episteme kinds, id kinds, reference kinds, or publication-face/form kinds merely to type its fields.
- A catalogue MAY be a core FPF catalogue or an organization-local extension when the same constitution, resolution, and family-declaration discipline remains recoverable.
Local bundle declaration and its ordinary family designator
A bundle declaration is a bounded claim block inside G_L. It states one finite, non-empty recurring family of U.ViewpointRef values resolving to members of catalogue subject K_L. Under R_L, each reference resolves to one viewpoint episteme edition P already admitted to U.Viewpoint membership under E.17.0. The declaration neither admits P nor changes P's C.2.1 identity.
Its minimum claim content is:
- one ordinary
familyDesignator, unique withinG_LunderR_L; - one target-kind compatibility condition: either the by-value criterion used for this family or a reference that resolves to the ClaimGraph defining or constraining the admitted target kind. If member viewpoints use different fixed target-kind criteria, the declaration states the compatibility rule rather than inventing a common superclass token;
viewpointRefs, one finite non-empty set ofU.ViewpointRefvalues;- optional references that resolve under their applicable schemes to archetypal-grounding examples or sections, with their intended recognition use stated;
- optional alignment claims naming the source and relation when a correspondence is asserted; and
- optional references that resolve under their applicable schemes to annex assets, each with its local role such as lexical note, Bridge material, A.16 move-publication note, example, or SoTA companion.
The family designator retrieves the declaration claim block inside L. A member U.ViewpointRef resolves to its P edition, and any reader-facing viewpoint token is only P's designator. The family designator, declaration claim block, reference, viewpoint designator, P, and L are distinct. No token, list position, prefix, alias, or member spelling substitutes for an episteme or reference.
The compatibility condition neither selects an actual EntityOfConcern for a describing use nor supplies or changes any member P's fixed target-kind criterion. Those claims remain in P and E.17.0 conformance.
A bundle is not a bundle of views, files, forms, carriers, or publication occurrences. If a receiving use needs an A.22 structure among the member viewpoints, it separately recovers obtaining relations and selects that structure. Declaration adjacency or order is not structure.
Changing the member-reference set, family meaning, compatibility condition, or the interpretation supplied by R_L changes G_L or the effective scheme and therefore identifies another catalogue episteme. Repackaging, annex layout, publication form, carrier, or audience does not reidentify unchanged L or any unchanged member viewpoint episteme.
Import discipline into MultiViewDescribing
When you name a family designator for a describing use, resolve catalogue edition L and its effective R_L, retrieve the designated declaration inside G_L, then name the imported reference subset Sigma. If L or the declaration is not already recoverable, use §4.2 to establish <G_L, K_L, R_L> before import:
Sigmais a subset of that declaration'sviewpointRefsin L;- every member is a
U.ViewpointRefresolving to one admitted viewpoint episteme edition P; - every candidate episteme E used under a member is independently identified under C.2.1 and is a
U.Viewonly whenEpistemeViewpointConformanceRelation(E,P)obtains; and - every one-viewpoint selection for one describing use carries a single
viewpointRef; importing the family neither selects P for that use nor establishes conformance.
A local subset identifies its source catalogue edition L, family designator, members used and omissions. When the project also asks you to choose a subset for a practical question, use §14.1.
A multi-library use preserves each <editionDesignator(L), familyDesignator> source and member provenance rather than flattening everything into one unnamed family. If one use selects several viewpoints, it constructs their C.13 collection with membership. It does not overload one reference or infer a new family from adjacency.
Construction, identity viewing, transformation, declaration membership, selection, naming, rendering, or publication grants neither U.Viewpoint nor U.View membership.
A local overlay may add didactic or publication material without changing L. Changing a member viewpoint's meaning, the reference target, membership set, or family meaning requires a new local catalogue edition carrying the revised or new family declaration rather than silent mutation under the inherited family designator.
Guard and naming discipline
- A viewpoint bundle is a family of viewpoints, not a bundle of views or documents.
- The family designator is an ordinary lexical value under
R_L, not a local id kind, publication-face/form kind, reference, or entity. - Engineering viewpoint designators and publication viewpoint designators may coexist, but their namespaces SHALL remain disambiguated.
- Bundle semantics come from the viewpoint episteme editions resolved by its member references, not from the spelling pattern of the family designator.
Publication and representation stay outside the bundle
A published library is the same selected C.2.1 episteme edition participating in E.24.PUB relations:
PublicationExpressionRelationrelates that selected edition, one independently identified expression, and one bounded-use declaration;PublicationExpressionBearingRelationrelates one independently identified physical or digital carrier to that expression;PublicationFormConformanceRelationrelates the expression to any selected reusable publication form whose rules it satisfies; andEpistemePublicationRelationrelates the selected edition, audience declaration, bounded-use declaration, expression, and carrier for one maximal continuous availability interval.
When form-mediated expression and bearing are claimed, retain E.24.PUB's corresponding relations with the same expression as witness. A change of exact publication need not interrupt broader availability of the library through its permitted expressions and carriers.
Changing a participant or restoring availability after a gap yields another publication occurrence under E.24.PUB; it does not reidentify unchanged L or any member viewpoint.
Rendering, printing, or uploading is separate system-performed U.Work. For a diagram or catalogue rendering, recover the correspondence between its elements and independently identified declarations or viewpoint epistemes. Use A.6.3.RT/A.6.3.RT.OE when constructing an expression or changing its representation language or reasoning medium for a particular use. Add C.29 when choosing or using a mathematical representation raises an inference or transfer question. Publication, representation, form, carrier, or rendering grants no viewpoint or view membership and makes no represented world-side relation obtain.
Archetypal Grounding
In FPF, an author can instruct a project to use an already defined viewpoint family without confusing that family with the views or publication faces the project uses later.
System description: hypothetical template instance. E.17.2 can guide one project to bind local references r_functional, r_procedural, r_allocation, and r_module to project P editions inside one constituted catalogue L. Until those bindings and their resolution under R_L exist, these names are variables and no reusable TEVB family value is present.
Governance account: hypothetical family shape. A project could bind local references for risk, control, compliance, and operations viewpoints in one catalogue declaration. The labels alone are not references or P editions. This example becomes reusable only after that project supplies complete <G_L, K_L, R_L>, bindings, and one ordinary family designator.
Bias-Annotation
After a recurring family-level use is established, the pattern favours reusing the catalogue and discourages teams from silently recreating the same family under local labels. For a one-off selection, keep the references local. The catalogue cost is justified only when reuse, comparison, or maintenance changes a named practitioner action.
Conformance Checklist
CC-VBL-0The triple<G_L, K_L, R_L>constitutes L; ordinary import resolves an admitted L and its declaration without reconstructing that triple, while authoring, admission, or disputed identity opens the complete §4.2 check. WithinG_L, each ordinary family designator retrieves exactly one local declaration claim block and remains distinct from L, member references, P designators, views, forms, and carriers.CC-VBL-1Every member is aU.ViewpointRefresolving to one independently admitted viewpoint episteme edition whose fixed target-kind criterion is compatible with the bundle constraint.CC-VBL-2Bundle membership, position, spelling, alias, packaging, or publication admits no P asU.Viewpoint; E.17.0 alone defines the membership test.CC-VBL-3A describing use imports a subset from<editionDesignator(L), familyDesignator>, preserves omissions and provenance, and selects any one P through a single reference.CC-VBL-4Every candidate E is independently identified and gainsU.Viewmembership only through obtaining E/P conformance—not through construction, selection, bundling, naming, form, carrier, rendering, or publication.CC-VBL-5A family designator is not used as an id kind, publication-face/form kind, carrier kind, viewpoint reference, or substitute for a member.CC-VBL-6Changes to member references, targets, family meaning, or compatibility constraints create another catalogue edition carrying the revised or new family declaration; publication or annex-only change does not reidentify unchanged P.CC-VBL-7Multi-bundle imports preserve catalogue provenance. A comparison identifies its predicate, participants, scope and operative interpretation and applies the subject method. A cross-local-sense use additionally retains the F.17 endpoints, obtaining F.9 Bridge and separate<u,d,r,t>claim; affirmative reliance needs the same-use A.10 provenance and the applicable ordinary-reliance or named-assurance result under F.9:4.2. Keep only the conclusion supported by these grounds.CC-VBL-8E.24.PUB expression, bearing, publication, recurrence, rendering work, and mathematical representation under C.29 remain distinct, grant no viewpoint or view membership, and make no represented world-side relation obtain.CC-VBL-9A bundle intended for non-expert reuse should provide references that resolve under their applicable schemes to archetypal-grounding examples or sections for its member viewpoints. Grounding aids recognition but grants no membership.
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
MultiViewDescribing supports viewpoint plurality. E.17.1 supplies packaging and provenance discipline for that plurality, including cases where viewpoints are used to re-express positions in a language-state chart declared under C.2.2a or language-state history accounts through the LanguageStateMoveTrajectory form under A.16.0. It makes member provenance explicit across repeated uses.
Semantic correspondence is a separate result: same-scheme comparison states its predicate and participants, while cross-context comparison uses F.9 and a bounded-use reliance path.
Source status, local rationale, and reopen condition
ISO 42010 is retained only as historical vocabulary lineage for the words view and viewpoint. It is not current architecting SoTA, does not supply FPF identity or conformance laws, and does not justify this catalogue architecture. No current external problem-solving source or reusable source comparison is claimed by this E.17.1 edition.
This design has a local FPF rationale. The problem is repeated use of the same E.17.0 viewpoint episteme editions. Users need to resolve references, preserve source catalogue and omission provenance, and avoid reidentifying P or turning a family label into membership.
One C.2.1 catalogue ClaimGraph with local declaration claim blocks is the least additional object that answers those actions while reusing C.2.1 identity, E.17.0 membership, C.13 collection, F.9 comparison, and E.24.PUB publication boundaries.
Reopen the local architecture if a current source or exercised project catalogue demonstrates a simpler way to preserve reference resolution, P identity, subsets, omissions, provenance, and cross-context comparison without losing any of those practitioner actions, or if project replay shows that one catalogue ClaimGraph with local declarations adds apparatus without changing a practitioner action. Until then, describe this as provisional local design, not source-established SoTA.
Relations
- Builds on:
C.2.1for library and member-episteme identity;E.17.0for P membership, reference resolution, singular use selection, and sole E/P view-membership rule;C.13for explicit imported collections;A.22for any separately selected organization;A.6.2-A.6.4for optional episteme-construction histories;A.7,E.7, andE.10for carrier, authoring, and naming discipline;E.24.PUBfor publication;A.6.3.RT/A.6.3.RT.OEfor expression construction and representation transition; andC.29for mathematical-lens use. - Constrains: E.17.0 consumers whenever they import a reusable family; an import narrows eligible references but neither selects one P for a use nor proves conformance.
- Coordinates with:
C.2.2a,A.16.0,E.17,E.17.2,E.18:5.12,F.9,F.9.1, and domain-specific families requiring stable reuse. - Protects: separation among catalogue triple
<G_L, K_L, R_L>, catalogue episteme L, local declaration claim block, ordinary family designator,U.ViewpointRef, P designator, P, candidate/View E, any A.22 structure, form, carrier, publication occurrence, and their representations.
Resolvable annex references for thin bundles
An ordinary project family designator may be accompanied by references that resolve under the applicable source or reference scheme to annex assets. Each reference states its local role, such as lexical, bridge, movePublication, examples, optional sota, or optional pilotTrace. Neither the field spelling nor the role value creates a new reference kind, manifest entity, or typed annex asset.
This keeps the declaration claim block thin while allowing A.16 move-publication notes, lexical material, Bridge material, and examples to remain explicit rather than folded into the core family claim.
Bundle Anatomy and Member Discipline
A viewpoint-bundle library becomes thin and reusable only when the bundle stays stable while the member viewpoints remain explicit enough to review independently. The bundle must be coherent as a family and clear about each member.
What a viewpoint member should make explicit
Each U.ViewpointRef member inside a reusable bundle resolves to one viewpoint episteme edition whose claim content makes explicit at least:
- the concern family it brings into focus,
- stakeholder or audience referents only when they change the concerns,
- the target-kind criterion it carries and the compatibility condition under which this family can reuse it,
- the independently admitted episteme kinds whose membership rules allow candidates under that viewpoint,
- any bundle-specific conformance notes later users must retain, plus a reference that resolves to the comparison claim or F.9 Bridge when either has independently been established; a note or reference creates no correspondence.
E.17.1 does not redefine the internals of U.Viewpoint. It states what must remain visible if a viewpoint is to be reused as part of a bundle rather than as an undocumented local label.
Bundle-level coherence
A shared prefix does not make a coherent bundle. A coherent bundle should answer a recognizable family-level question, such as:
- which engineering concerns are standard for holon description?
- which governance perspectives are required for a service review?
- which research-method viewpoints recur across inquiry reports?
If the member viewpoints do not share that family-level purpose, the result is not one bundle but an uncurated catalogue fragment.
Thin bundles, rich annexes
E.17.1 allows bundles to stay thin. Projects may link to companion material by references that resolve to the annex assets under their applicable schemes. State the local role of each reference. Examples include:
- lexical discipline notes,
- bridge overlays,
- A.16 move-publication notes,
- worked examples,
- or SoTA references
This preserves a stable declaration claim block while letting the reuse package carry enough teaching material and review guidance.
Import, Subset, and Multi-Bundle Coordination
Importing a family, selecting a subset and coordinating several families show why viewpoint bundles are useful. Make each choice explicit so that other users can tell what the project has reused and what it has changed.
Subset selection
A MultiViewDescribing use may import only a subset of a bundle's viewpoint references. When it does so, it should declare:
- which ordinary family designator is the source,
- which viewpoint members are in local use,
- and whether the omitted members are unused or are intentionally excluded because the local scope does not require them.
To choose a subset, recover the question the descriptions must answer. Which claims would let the recipient act or explain the result? Select the viewpoints whose rules call for those claims. A single member may cover a whole service. If the decision depends on how two parts of the service work together, establish that the descriptions can be related under the conditions of that decision. If the catalogue leaves a needed claim uncovered, identify where to obtain it. A commission confined to importing a specified subset ends when its members and provenance are recoverable.
Local overlays vs new bundles
A local project often wants a small adaptation: one extra concern note, one narrower stakeholder emphasis, one local naming convention. E.17.1 prefers explicit overlays or new editions over silent mutation.
A practical rule is:
- if the local project selects a subset or adds only didactic/publication material, keep catalogue edition L and its declaration unchanged and declare the local subset or annex; do not treat the overlay as declaration content;
- if the local project changes viewpoint membership or meaning, publish a new local catalogue edition carrying the revised or new family declaration.
Multi-bundle coordination: provenance first, comparison separately
Many description families need more than one bundle, for example:
- one engineering viewpoint family,
- one safety or assurance family,
- and one governance or publication-oriented family.
Preserve the provenance of every imported U.ViewpointRef and resolved P as <editionDesignator(L), familyDesignator, member reference>. That tuple identifies where each member came from. It establishes no semantic sameness, difference, correspondence, translation, substitution, or admissible comparison by itself.
Identify the P editions or claim subgraphs being compared, the comparison predicate, polarity, scope and participants. Recover how each source interprets the values used by that predicate, then apply its subject method. Different schemes may require a representation or quantity conversion without requiring a semantic Bridge. If the material supports only a lexical or structural contrast (such as members, omissions, order, target criteria or claim-shape differences), state that limited result.
When the comparison or proposed reuse depends on a semantic relation between distinct local senses, resolve their F.17 SchemeSenseCell endpoints and establish the F.9 Bridge. State the proposed use separately as one bounded C.2.1 claim about that Bridge with <u,d,r,t> and polarity.
For reliance on an affirmative claim, retain the same-use A.10 provenance. Ordinary reliance requires RelianceDisposition=pass; when a named assurance claim is current, require its B.3 result with disposition=supported-for-use for that bounded use.
If a required meaning, correspondence or reliance ground is missing, withhold the dependent conclusion and retain the supported contrast or exact inquiry. Catalogue provenance identifies the sources in every branch.
Engineering vs publication families
Some contexts need both engineering viewpoints and publication viewpoints. A shared family designator must preserve the distinction between engineering and publication viewpoints. A family that imports both kinds must keep the namespaces and catalogue origins explicit so that authors can distinguish how the holon is being understood from how a publication face/form chooses to expose that understanding.
Worked family shapes, not shipped catalogue values
Hypothetical TEVB project binding
E.17.2 supplies an authoring template, not a repository-shipped family. One project may constitute catalogue L and bind four local variables:
r_functional -> P_functional,r_procedural -> P_procedural,r_allocation -> P_allocation,r_module -> P_module.
Only after those are U.ViewpointRef values resolving to admitted P editions under L's effective scheme can the project's ordinary family designator retrieve a reusable local declaration. Another project with similarly spelled variables or labels has not imported this family unless it resolves the same L and references.
Hypothetical governance and risk shape
A project may author a governance-oriented declaration with local reference variables such as:
r_risk -> P_risk,r_control -> P_control,r_compliance -> P_compliance,r_operations -> P_operations.
This is an example of a possible declaration shape, not a current family. Each left-hand variable must be bound to a local U.ViewpointRef; each right-hand variable must be bound to one P independently admitted under E.17.0. L, R_L, and the family designator must exist before reusable import is claimed. The four positions recur together but remain non-interchangeable.
Hypothetical research-method shape
A project may likewise consider local variables r_theory, r_experiment, r_inference, r_limitations, and, where appropriate, r_reproducibility. This list teaches a candidate family shape only. A local inquiry note can import a subset only after the project has constituted L, bound each retained variable to a reference and P, and made omitted members visible in one declaration claim block.
Cross-family description relation positions
A project may use one materialized local TEVB instance for its design family, another local governance family for program oversight, and another local publication-oriented family for publication faces and forms. E.17.1 keeps these relation positions reviewable by preserving which catalogue and declaration each viewpoint came from and by preventing a final publication face or form from masquerading as the catalogue itself.
Authoring and Review Guidance
For bundle authors
Bundle authors should ask:
- what recurring family is being named,
- which viewpoints belong together in that family,
- what local didactic publications or examples belong in annexes instead of the bundle core,
- and whether the bundle is stable enough to deserve a reusable family designator.
For reviewers
Reviewers should inspect both levels:
- member level - are the included viewpoints individually explicit enough to be reused?
- bundle level - do they form one coherent family rather than one convenient list?
They should also check whether a local project has silently forked the bundle while still using the inherited family designator.
For integrators and librarians
Integrators should keep libraries small, curated, and editioned. Publish only the smallest declaration set the current reuse needs:
- one stable core declaration when a recurring family is established,
- one explicit local extension only when local membership or meaning changes,
- and one clear subset declaration only when the current use imports a subset.
Do not create all three by default.
Edition and Migration Notes
Rename vs semantic change
A lexical rename that leaves viewpoint meaning and membership unchanged may be treated as a naming-layer migration. A change in membership, concern, admissibility, or member semantics requires another catalogue edition carrying the revised or new family declaration.
Migration from local Sigma lists
Legacy MultiViewDescribing uses often publish only one local list of viewpoints. Migration should proceed by:
- identifying recurring families across several such local lists,
- publishing those families as explicit bundles,
- then rewriting the local families to import the new ordinary family designator and declare any subset selection.
This sequence preserves provenance and avoids pretending that the reusable family had always existed.
Migration from publication-face/form-bound naming
If a legacy practice uses one label interchangeably for a viewpoint family, a viewpoint, a report section, and a publication face, distinguish those uses of the label during migration. The ordinary family designator remains at the declaration layer. U.ViewpointRef values resolve P while any reader-facing viewpoint token is only P's designator. Publication-face names remain publication-layer vocabulary.
Boundary to annex growth
Annex references are useful, but a declaration should not become a thin shell hiding all of its meaning elsewhere. The core declaration must state enough about its members and their organization as a family to stand on its own. Annexes deepen reuse, but they do not replace the declaration's primary claims.
Import Collision and Alias Discipline
A family designator is not a synonym bag
An ordinary family designator does not mean that all member viewpoints are interchangeable labels for one concern. One declaration states that this reviewed family of viewpoints is intended to recur together. Authors should keep the family designator distinct from each member.
Import collision rule
When two imported bundles contribute viewpoints with overlapping lexical names, preserve the originating viewpoint designators and catalogue provenance rather than silently merging the members. Inspectable collisions make provenance adequate; they do not show that the local senses correspond or that either member may substitute for the other.
Alias boundary
Local teaching aliases may be added for readability, but the alias must identify the member viewpoints it denotes and must not erase bundle provenance. If the alias alone is used to select the bundle, the selection is unsupported. The project should replace the alias in that use with explicit member references.
Bundle Projection and Comparative Use
Projection to local subsets
A description family may project only a subset of a reusable bundle. Identifying omitted members makes the extent of the import clear. Use §14.1 to judge whether the subset will serve a particular question. Projection keeps bundle provenance intact while acknowledging that local publication rarely uses every member.
Comparative bundle use
Name the members or claim subgraphs, comparison predicate, polarity, scope and participants; recover their operative interpretation and apply the subject method as in :14.3. If only names, member sets, omissions or structures can be compared, state that bounded contrast and stop.
When the comparison needs a semantic relation between distinct local senses, use their F.17 cells and the obtaining F.9 Bridge. A proposed comparison, translation or reuse needs its separate bounded-use claim with direction, correspondence rule, tolerated loss and polarity. For reliance on an affirmative claim, retain the same-use A.10 provenance and the applicable ordinary-reliance or named-assurance result under F.9:4.2. Use F.9.1 only when a separate stance episteme about that bounded-use claim is needed. The note explains this use and leaves the Bridge and its bounded-use claim unchanged. Assess a different correspondence or a wider use under F.9. Retain an unresolved required basis as the limit on the dependent use.
Boundary to publication-face design
A publication face may render one composite presentation of several viewpoints, but the face is not the bundle. E.17.1 therefore requires the underlying member structure to remain recoverable even when a public-facing document flattens it for readability.
Review Matrix and Catalogue Maintenance
A reviewer can test a viewpoint bundle library with five questions:
- Do the member viewpoints still have explicit standalone meaning?
- Does the declaration retrieved by the family designator describe one coherent recurring family rather than one convenience list?
- Can the reviewer see which members were omitted and, if the subset is recommended for a task, how the retained descriptions can support that task?
- If several bundles interact, is provenance preserved, and does each comparison use its subject method with F.9 added only for a semantic relation between distinct local senses?
- Has a publication face started impersonating the library?
Prefer small, provenance-preserving declarations inside editioned catalogues over lexical mega-families that are easy to name but hard to reuse truthfully.
E.17.1:End
TEVB - Project-local Typical Engineering Viewpoint Bundle Template for Holons
Status: Stable Type: Authoring template; no TEVB catalogue value is shipped by this pattern.
Use this when. A project wants to author a small local family of engineering viewpoints for descriptions of holons, so that functional, procedural, allocation-responsibility, and module-interface claims remain distinguishable and comparable.
What goes wrong if missed. One functional, procedural, responsibility or structural label, or a label for a diagram or report, can come to mean several things at once. People may take it for the viewpoint, the view, the described holon, a publication face, or proof of an engineering relation. The opposite failure is to require all four viewpoints and their full authoring machinery for one local reading.
What this buys. TEVB supplies a four-position authoring template. Once a project has constituted its own catalogue L and bound four local references to four viewpoint epistemes, that project can reuse the resulting local family while keeping candidate episteme, described holon, conformance, cross-view relations, and publication separate. One use may select just one bound member.
First action. Resolve the already admitted project-local catalogue edition L and the local declaration designated by f_eng, then use the engineering question to choose which member references to resolve (§4.3). If L or the declaration is new, missing, or disputed, use E.17.1:4.2 to constitute or verify <G_L, K_L, R_L> for that edition. If a needed P edition is missing, author and admit it under E.17.0 before binding its reference.
Reuse those results while the catalogue edition, effective scheme, declaration, and relied-on premises remain unchanged.
First useful result. A materialized local family has one project-local L, ordinary family designator f_eng, four local references r_functional, r_procedural, r_allocation, and r_module, and four local P targets to which those references resolve under R_L. For recognition of one view, the result is the admitted L and declaration, the reference resolving P, and the E/P conformance judgment. When the question concerns the engineering work itself, use the claims of the relevant views to answer it.
Before the four bindings exist, the result is only an authoring template, not a reusable family value.
Ordinary stop. For materialization, stop when the local catalogue triple, declaration claim block, four reference bindings, and four P targets are recoverable. For recognition under one specified viewpoint, stop at its E/P judgment. For an engineering question, stop when the claims and any relations the answer depends on warrant that answer. Reopen full catalogue constitution only under the E.17.1:4.2 triggers.
Add another member, structured viewpoint-authoring witness, C.13/A.22 organization, construction history, cross-view relation, evaluation, or publication object only when a named receiving use depends on it.
Not this pattern when. Keep a one-off viewpoint local when no recurring four-position family is needed. Author another E.17.1 declaration for safety, assurance, information, mission, deployment, business, publication, or architecture-framework-specific concerns outside the four TEVB positions. TEVB is not a universal architecture framework.
Tech-name:
TEVB— the template name, not a family designator or catalogue value Plain-name: project-local typical engineering viewpoint bundle template for holons
Product-form boundary. This pattern ships no catalogue edition, effective scheme, family designator, U.ViewpointRef, or viewpoint episteme edition. Every L, f_eng, r_*, P_*, C_*, Q_*, and S_* symbol below is a variable in the template until one project supplies and verifies its binding.
Equal labels in two projects establish no shared family or cross-project reuse. Such reuse begins only when both uses resolve the same L and member references.
The template does not by itself constitute an architecture framework, a U.Method, a set of publication forms, or an additional entity alongside catalogue L and its referenced P editions. It prescribes no modelling notation, storage format, or tool API.
Builds on: E.17.0 for U.Viewpoint, EpistemeViewpointConformanceRelation, and U.View; E.17.1 for bundle packaging by U.ViewpointRef; C.2.1 for episteme identity; C.13 for the constituent collections of viewpoint conventions; A.22 for their selected structures; A.6.6 and E.17.0 for constituent-dependency relations; A.6.3 for optional view construction; E.24.PUB for publication.
Used by after a project materializes the bindings: E.18 transformation-flow descriptions, E.17 multi-view publication, architecture-description patterns, and domain patterns that need that local engineering concern family for holons.
Problem frame
Engineering descriptions repeatedly ask four different questions about one holon:
- Functional: what transformations, capabilities, and effects characterize what the holon can or is intended to do?
- Procedural: what methods, orders, states, concurrency, failures, and recovery rules characterize how relevant behavior unfolds?
- Allocation-responsibility: which admitted Systems and local system-role kinds relate to the holon’s behavior? Which current C.3.2 judgments, obtaining assignments, capabilities, transformations, and separately governed responsibility relations or selected structures relate to it? Each such C.3.2 judgment says that one System counts under a kind for one
KindSignatureedition and context slice. - Module-interface: what constituent holons, interfaces, dependency structures, substitutability conditions, and change rules characterize its construction?
The questions recur across hardware, software, organizations, and mixed systems. Their answers may appear as prose, models, diagrams, cards, or publications, but those forms do not identify the viewpoints or make an episteme a view.
Problem
How can engineers reuse a compact family of these four concern-bearing viewpoints while keeping all of the following distinct:
- the holon described by a candidate episteme;
- the viewpoint episteme and its target kind or, only in the triggered structured branch, its selected convention structure;
- the candidate episteme and any dependent
U.Viewmembership; - a viewpoint selected for one describing use;
- the Methods, transformations, selected structures, local system-role kinds, assignments, modules, and interfaces mentioned in the claims;
- any viewing construction, evaluation, cross-view relation, publication occurrence, form, representation, or carrier?
Without that separation, a label such as functional view can stand for a concern convention, a diagram, a query output, a report section, or a claim about a system. The next engineering action then relies on the wrong object.
Forces
Solution
Local mantra. To materialize a local instance, constitute L and bind f_eng, four references, and four P targets. To use an admitted instance, resolve L, its declaration, and only the needed reference. Then identify the holon-centered candidate E and test its conformance under E.17.0. For any additional engineering or publication claim, keep its objects and relations distinct and use the applicable pattern.
Use the mantra to recall the procedure. The rules below specify the positions, how to materialize a local family, how to use the conformance test and when to stop. Their variables do not denote repository-shipped values.
Bind one project-local declaration without embedding viewpoint values
One project instantiates the template only by supplying these bindings:
The four r_* variables must be bound to local U.ViewpointRef values. The four P_* variables must be bound to already admitted viewpoint episteme editions. f_eng and any reader-facing names are ordinary designators under R_L. Designator, reference, viewpoint episteme, any optional selected viewpoint-convention structure, declaration claim block, and catalogue L remain distinct.
The template does not admit P as U.Viewpoint, make another episteme a U.View, or establish publication. Use E.17.0 for both dependent-kind membership tests, E.17.1 for L and its declaration claim block, and E.24.PUB for publication.
The four positions are fixed for a project declaration that claims conformance to this template. Safety, assurance, information, mission, deployment, business, and publication-oriented viewpoints use another local E.17.1 declaration or a later project catalogue edition with an explicitly revised declaration. A recurring label alone neither binds nor extends f_eng.
Materialize each local viewpoint before binding its reference
Each P_* variable must be bound to one C.2.1 episteme that independently gains U.Viewpoint membership under E.17.0.
Start with E.17.0's self-contained branch: give P its admitted target kind as EntityOfConcern and put the complete fixed target-kind, concern, admissibility, semantic-form, coverage, consistency, completeness, omission, and describing-use test in its ClaimGraph.
Use the structured C/Q/S branch below only when separately versioned convention components and their organization change a named project reuse, comparison, or maintenance action.
For any one of the four positions:
- identify the target kind and the complete fixed claims required by E.17.0;
- in the ordinary self-contained branch, constitute P with that target kind as EntityOfConcern;
- in the independently triggered structured branch, identify convention epistemes under their least-powerful admitted kinds, construct collection C under C.13, recover every selected obtaining direct relation, state ordinary constraint episteme
Q_org, let a system perform the A.22 selection work, and identify selected structure S. Constitute P with S as EntityOfConcern; - apply the five E.17.0 viewpoint-membership conditions to the resulting P, then bind its project-local reader designator and
U.ViewpointRef; and - record the resolution under
R_Lin the local declaration claim block.
No constituent, Q_org, or P becomes a U.Signature merely to fit this template. A constituent is a U.MethodDescription only when it describes one independently admitted method under A.3.2.
Selection work and its result remain separate from C, S, P, and selected relation occurrences. The table below gives optional recipes with template variables. A project supplies the values.
The four template positions use these concern objects and patterns when one project authors its P editions:
Keep these claim boundaries explicit:
- Functional: functioning status, input/output boundary, and functional-port coverage remain claims in
E_rule.functionalCoverageunless the claim identifies a separate EntityOfConcern and states its predicate, participants, and obtaining test. A concern episteme about a Transformation, a holder's ability or a transformation-flow Structure has respectively that Transformation, holder System or Structure as EntityOfConcern. State the ability in its claim content. Select an ability-assertion episteme itself as subject only for a question about that assertion. When several of these concerns are needed, keep their subject accounts distinct. There is no universal function entity or one multi-subject concern episteme. - Procedural: every method, order, state, concurrency, failure, and recovery claim designates its operational subject and the admitted method, state-transition, or transformation-flow relation that gives the claim meaning. A bounded coverage rule may remain in P, but a candidate E cannot satisfy it through vocabulary alone. For a MethodDescription claim, check that E describes how its own admitted Method subject is carried out (A.3.2). For a View claim, check conformance to the selected P. State wording alone establishes no Structure, and procedural content alone establishes no performed work. Add safety evidence only for a safety-bearing claim or named reliance.
- Allocation-responsibility: holder System, local system-role kind, four-input C.3.2 classification judgment, optional extension representation, assignment, transformer relation, allocation, segregation, capability, and responsibility remain separate typed claims or concern objects. A local system-role kind is not a classification judgment or assignment. Classification or assignment establishes neither responsibility nor Work.
- Module-interface: A.6.M
ModuleInterfaceClaimremains claim content. Whole-holon, candidate-module, boundary, independently identifiedInterfaceSpecificationepisteme and its resolving reference, substitutability, and change-policy content stays in the coverage-rule episteme until a module-relation declaration supplies participant kinds, predicate, obtaining rule, and occurrence identity and current facts satisfy it. The claim record is not that relation and a module topic is not an EntityOfConcern.
Split any phrase spanning several subjects into separate concern epistemes, or retain it as one constraint claim over candidate content.
Give each stakeholder constituent exactly one referent—System, local system-role kind, claim-bearing C.3.2 classification-assertion episteme when that judgment is current, obtaining system-role assignment, C.13 collection-as-whole, or other independently governed subject.
A KindExtension remains an optional representation for a named set-consuming use, not the kind or judgment. Cite any responsibility concern through its separately governed direct predicate. Do not coerce heterogeneous constituents into Signatures merely to make the rows uniform.
The following four rows are structured-branch recipes. Every symbol is a template variable until a project binds values. A project using the ordinary self-contained P branch does not instantiate the structured recipe for that position.
Each project-bound structured witness remains independently recoverable. Constituent editions identify C. Every selected dependency occurrence passes the E.17.0 predicate. Optional D_dependencyUse states obtaining and named-use admissibility as separate claims. A.22 selects S from C, selected occurrences, applied Q constraints, and the use frame. P is then identified by its ClaimGraph, S EntityOfConcern, and effective scheme.
Changing only the Q edition leaves S unchanged when those selection inputs remain semantically unchanged. No topic list, citation, displayed edge, hidden O, D, template variable, or neighboring witness establishes that another structured witness meets these requirements.
The dependency relation in this table is ViewpointConventionDependencyRelation from E.17.0. It obtains only when interpreting or replaying the fixed claims of the dependent episteme relies on a criterion, law, public name, or method claim of the base episteme, and replacing the base edition can change that interpretation or replay. Co-membership, citation, or a visible arrow is insufficient.
When an A.22 selection judgment needs an explicit claim that one obtaining dependency occurrence is admissible for that use, identify the separate decision-use episteme described by E.17.0. Do not insert that decision, its evidence, or its evaluation result into the dependency relation or S identity.
Keep the four concern conventions distinct
Choose among these concerns from the engineering question. Which difference in the holon, its use or its parts could change the answer? For example, a plant-wide flow bound during a pump change can establish a minimum total flow. If each outlet has a separate minimum, the same total can hide a loss at one outlet. Trace which paths serve which outlets during the change. Where supply depends on switching order, relate the operations of the switching Method to the equipment available at each stage. A claim that yesterday’s crew actually switched in that order draws on its work. A single view can already carry all the grounds for the particular answer.
Functional. A conforming candidate episteme foregrounds transformations, capabilities, effects, functional elements, or transformation-flow relations of its holon. It does not identify a module structure by functional vocabulary and does not mint U.Function. Any neighboring responsibility claim keeps the admitted System, local system-role kind, current C.3.2 classification judgment, assignment, kind-relation structure, capability, transformation, and direct responsibility relation separate. Use A.2, C.3.2, A.2.1, A.2.7, or the direct responsibility pattern for the claim made.
Procedural. A conforming candidate episteme foregrounds methods, order, state, concurrency, failure, and recovery related to its holon and designates the admitted method, state-transition, or transformation-flow relations on which each claim depends. A view about a factory may cover its restart Method while keeping the factory as its subject. If the holon described is the admitted restart Method itself, the same episteme also qualifies as a U.MethodDescription when it meets A.3.2. Ordinary operational recovery needs no safety package unless the claim is safety-bearing or a named receiving decision relies on one.
Allocation-responsibility. A conforming candidate episteme foregrounds Systems, local system-role kinds, current C.3.2 classification judgments, obtaining assignments, relations among those kinds, capabilities, transformations, and separately governed responsibility relations or selected structures related to its holon.
A label creates no kind, classification, or assignment. A classification judgment needs its candidate, kind, KindSignature edition, and context slice but no assignment. The view may state that judgment, but it does not make the criterion true or create an assignment or responsibility relation.
Module-interface. A conforming candidate episteme foregrounds constituent holons, dependency structures, boundaries, interfaces, compatibility, substitutability, and change policy. It remains distinct from the functional viewpoint. Many modules may support one transformation, one module may support several transformations, and either description may be incomplete without becoming the other.
The following are practitioner recognition and claim-shape cues, not embedded StakeholderFamilies or AllowedEpistemeKinds fields. A reader label creates neither a system-role classification nor an assignment and enters neither viewpoint nor view identity. Every example needs its EntityOfConcern, the predicate and participants of each claimed relation, its obtaining test, and its E.17.0 conformance result.
Recognize holon-centered TEVB views by conformance
TEVB keeps two subjects explicit:
EpistemeViewpointConformanceRelation(E,P) must obtain under the fixed E.17.0 predicate. Only then is the same episteme E a U.View. Direct authoring, query execution, A.6.3 construction, a reader-facing label, declaration membership, or publication does not establish that membership. A reader-facing system-role label also establishes neither the local kind nor a C.3.2 classification judgment or assignment.
For one current describing use, its use qualification carries a single viewpointRef : U.ViewpointRef resolving P under the effective reference scheme. Any reader-facing viewpoint name is only P's ordinary designator. The use qualification, designator, reference, and P remain distinct. Selection identifies neither E nor H, establishes no conformance, and adds no conformance participant or episteme-identity field.
Recover H only as EntityOfConcern(E) from E's C.2.1 constitution. Do not import a legacy context tuple, generic bounded-context object, or model-use identity field into E, P, S, conformance, or selection. Another use may select another P while E remains unchanged. Several selected viewpoints require a C.13 collection of their references rather than one overloaded reference.
If the required EntityOfConcern changes from H to one of its Methods, a local system-role kind, a system-role assignment, a transformation, a responsibility relation, or a structure, identify a candidate episteme about that subject. To use it as a view, choose a viewpoint that admits the subject and establish conformance. If H already is the required Method, the existing candidate can serve under either or both classifications established in §4.3.
Import, subset, and extend one materialized local instance
An E.17.0 multi-view use can import TEVB only after it resolves one admitted project catalogue edition L, retrieves the declaration claim block designated by local f_eng, and resolves the imported r_* members or subset.
Open <G_L, K_L, R_L> under E.17.1:4.2 only when L or the declaration is new, missing, or disputed, or a named later use consumes the catalogue constitution as premises.
f_eng is only an ordinary designator inside L: it identifies neither L nor any viewpoint by itself and is not a member reference. Each imported reference resolves to P under R_L; any reader-facing viewpoint name is only P's designator.
A local subset names retained references, preserves <editionDesignator(L), f_eng> provenance, and records whether each omission is unused coverage or an intentional exclusion.
If local work changes only reader-facing aliases or adds examples, keep those as naming or annex content.
If it changes a viewpoint's target criterion, concerns, admitted episteme kinds, or conformance rules, identify another viewpoint episteme edition and bind another reference as needed.
If it changes declared family membership, identify another catalogue episteme carrying the revised declaration claim block. Do not keep an old designator while changing the P it resolves under the same effective scheme.
Several local families may be used together, but each member retains its catalogue provenance and resolved viewpoint edition. Similar labels do not merge members. Two projects can claim use of the same reusable family only when they resolve the same L, declaration, and member references. Independent instances of this template remain different local families even when all four labels match.
A project may bind its local four positions to reader names such as Functional, Procedural, Allocation-Responsibility, and Module-Interface. Those names do not perform the binding. A different reference-to-position mapping is another local declaration and must not silently reuse the earlier f_eng under R_L.
Keep cross-view relations and publication separate
A materialized local TEVB instance provides four project references. The template alone provides none. Neither instance nor template asserts correspondence among resulting views. When a later engineering use depends on a relation between a functional claim and a module claim, or between a procedural claim and a system-role-assignment claim:
- identify the participating entities or epistemes;
- state the realization, allocation, dependency, consistency, trace, or other direct relation claimed;
- use the pattern that defines and tests that relation, including its obtaining law;
- use A.6.RCD when no existing direct or derived relation is sufficient;
- recover any representation's correspondence to that relation; use A.6.3.RT/A.6.3.RT.OE for expression construction or representation transition, adding C.29 when a mathematical inference or transfer question requires it.
If a separate receiving claim asserts dated U.Work, recover each actual performer through A.13 and use A.15.1 to establish the Work, Method, time, and containing System independently. Add F.6 only when that receiving claim also consumes precise assignment-bound attribution through the same obtaining A.13 assignment. Missing or failed F.6 alone leaves independently admitted Work intact. Those Work and optional attribution facts are neither participants in the cross-view relation nor prerequisites for identifying it.
E.17 and E.24.PUB may publish a selected TEVB view edition through PublicationExpressionRelation(selectedEdition,publicationExpression,boundedUseDeclaration), PublicationExpressionBearingRelation(presentationCarrier,publicationExpression), and the five-participant EpistemePublicationRelation(selectedEdition,audienceDeclaration,boundedUseDeclaration,publicationExpression,presentationCarrier). Identify the expression and carrier independently under their own domains.
When a reusable publication form is selected, establish PublicationFormConformanceRelation(publicationExpression,publicationForm). The same expression witnesses the corresponding form-mediated expression and bearing relations.
Each exact relation retains its own participant set and maximal continuous obtaining interval; changing a participant or restoring its predicate after a gap yields another occurrence without reidentifying unchanged E or P. Broader availability through permitted expressions or carriers can continue when an exact publication changes. For a team using different publications, check that the permitted routes together reach each required reader when needed under E.24.PUB:4.2.1.
Rendering, printing, upload, or carrier manipulation is separate system-performed U.Work. A representation retains its correspondence to independently recovered objects or relations. C.29 adds the conditions for a mathematical inference or transfer.
A publication-side viewpoint, when current, is another viewpoint episteme selected by reference—not a TEVB position label reused as a form or file name. View episteme, viewpoint episteme, construction, conformance, form, carrier, publication, rendering, and representation remain distinct. Publication and representation make no represented world-side relation obtain.
Worked cases
The cases below assume one hypothetical project has already constituted L_local, bound f_eng, and resolved r_functional -> P_functional, r_procedural -> P_procedural, r_allocation -> P_allocation, and r_module -> P_module under R_L. They demonstrate a materialized local instance. They do not assert that these values exist in the repository or in another project.
Four views of a processing plant
The plant Plant_X : U.System is the EntityOfConcern of four separately identified epistemes.
- E1 states transformations, capabilities, material-flow effects, and functional boundaries.
r_functionalresolvesP_functional; E1 conforms to that P and is a functionalU.View. - E2 states claims about admitted method
PlantOperation, A.19.SPR operational-state structurePlantRunState, and E.18 transformation-flow structurePlantRunFlow. Its order, failure, and recovery claims designate the transition conditions and flow relations in those structures. It conforms toP_procedural. It is not a method description because its EntityOfConcern is the plant. No safety-bearing claim or named reliance is present in this case, so no safety-analysis, A.10, or B.3 branch is opened. - E3 states that
PumpUnit-3counts as local kindCoolingCirculatorSystemRolein the plant slice through a separate C.3.2 judgment over the candidate, kind,KindSignatureedition, and slice. That claim needs no assignment. E3 separately states any obtaining assignments, capabilities, transformations, and governed responsibility structures that are current. It conforms toP_allocation. Neither E3 norP_allocationmakes the classification criterion true or creates an assignment or responsibility relation. - E4 states constituent equipment holons, dependency structure, pipes, interfaces, substitutability, and change policy. It conforms to
P_module. E4 remains separate from the diagram rendering in which it is published.
The four conformance occurrences make E1-E4 views. Their shared holon and common local declaration do not establish any cross-view realization or consistency relation. Those claims are tested separately.
Query output missing a required concern
A query constructs episteme Y from plant model X, and A.6.3 records that construction. Y is labelled functional view, but it omits the output-condition coverage required by P_functional. Y does not conform to that P and is not a U.View under it. A new edition with repaired claim content becomes a U.View under that P if it passes the predicate.
Ordinary non-safety jam recovery
Candidate procedural episteme E_jamRecovery concerns conveyor system H. Its ClaimGraph designates admitted method ClearJam, an operational-state structure with Running, Blocked, and Resetting positions, the transition conditions between those positions, and the E.18 flow relation that resumes only after the blockage sensor is clear.
These method, state, and flow facts supply the operational basis for its failure-and-recovery claims. If EpistemeViewpointConformanceRelation(E_jamRecovery,P_procedural) obtains, E_jamRecovery is a procedural U.View.
No claim in this case is safety-bearing, no receiving decision relies on a safety analysis, and no evidence or assurance result is requested. Therefore neither an A.10 evidence path nor a B.3 assurance branch is opened. A later actual clearing remains separately identified U.Work.
Safety-triggered recovery use
Suppose a second claim says that restarting H after the same jam is safe for an exposed operator, and a named restart decision relies on that proposition. The project now identifies the safety-analysis episteme and its hazard, guard, and recovery claims; relates the relied-on evidence through A.10; and uses B.3 only when a named assurance claim for the restart use is current.
If the applicable safety rule or restart decision requires assurance, state that claim, its target, and its named use before applying B.3.
The operational method, state transitions, and flow relations remain the same operational basis. Safety analysis and reliance are added because this claim and decision trigger them, not because every failure or recovery description requires assurance.
Responsibility diagram and actual assignment
A responsibility-diagram episteme E concerns System H. Local reference r_allocation : U.ViewpointRef resolves P_allocation; EpistemeViewpointConformanceRelation(E,P_allocation) obtains.
Diagram cue. One box names MaintainerSystemRole@Plant. That spelling can help locate the plant-side definition. By itself it establishes neither a local system-role kind, an assigned-kind domain, a C.3.2 judgment, nor an assignment.
Classification-only claim. If a current claim says PumpUnit-3 counts as CoolingCirculatorSystemRole for CoolingCirculatorKindSignature-2 and PlantSlice-7, recover J(PumpUnit-3, CoolingCirculatorSystemRole, CoolingCirculatorKindSignature-2, PlantSlice-7) = true under C.3.2. No assignment is required.
Assignment claim. If a separate claim says admitted System S holds an assignment, first recover the local kind (here MaintainerSystemRole) through C.3 and declare the assigned-kind domain (here PlantMaintenanceSystemRoleKindDomain). The diagram cue identifies neither. Then recover RA : MaintenanceWorkAssignment <: U.SystemRoleAssignment under A.2.1, with S in HolderSystemSlot, PlantMaintenanceSystemRoleKindDomain as the declaration-local assigned-kind domain, and MaintainerSystemRole as RA's assigned-kind value.
E can assert or describe RA without becoming RA. Any responsibility of S remains a separately governed direct claim.
One view, two publications
Module-interface view E is published as an interactive model and as a printed inspection sheet. Both publication occurrences select the same episteme edition. Their forms and carriers differ. Both publications involve the same E, its conformance occurrence, and its U.View membership. If an inspector needs the interface’s load limit offline, the sheet can support the decision by carrying the applicable condition and its source or by providing a route the inspector can use in time. A link back to the interactive model can already suffice for a reader who only needs to locate it.
DDD Context Mapping method and product
A team enacts DDD Context Mapping. The way of doing is one independently admitted U.Method under A.3.1. An episteme that substantively describes that method may separately be a U.MethodDescription with the method as its EntityOfConcern. Neither is a TEVB viewpoint or view by its label.
First determine whether the product is a claim-bearing episteme or only a diagram, form, or carrier. A claim-bearing product called a Context Map is separately identified under C.2.1 as candidate episteme E with its own claim content, EntityOfConcern, and effective scheme.
It becomes a U.View only if one viewpoint P admits E's EntityOfConcern and EpistemeViewpointConformanceRelation(E,P) obtains. Method enactment, product naming, diagram form, declaration position, publication, and visual resemblance grant no membership.
If the map represents independently recovered domain regions or relations, recover how its elements correspond to those subjects. Use A.6.3.RT/A.6.3.RT.OE for constructing the expression or changing its representation language or reasoning medium for a particular use. Add C.29 when a mathematical inference or transfer depends on that correspondence. A carrier remains with E.24.PUB, and the drawing makes no represented world-side relation obtain.
Consequences
Reopen the TEVB template when its four positions no longer give a small useful engineering concern family for routine holon description.
Reopen one materialized local instance when a bound P's target criterion or conformance rules change, or when a candidate concern cannot be expressed without changing a local binding. Author another local family instead when the concern is orthogonal rather than a replacement for the four.
Provisional local design rationale and source status
The four TEVB positions are a provisional local choice for authoring routine holon descriptions. They are retained because each changes a different immediate practitioner question and none can reliably be inferred from another by label alone:
The template uses four positions to keep this family small and makes no claim to completeness. The table identifies important concerns outside the four:
Source status. ISO 42010 is historical vocabulary lineage only. Function–behaviour–structure language is also lineage and a recognition aid, not evidence for this four-position cut. Query or projection production uses C.2.1 to identify the candidate episteme and A.6.3 to state its construction. That construction supplies no external basis for choosing the viewpoints.
Responsibility/allocation is retained because it changes the practical question and avoids a recurrent function/actor collapse.
Reopen. Re-run source selection and a bounded actual-use comparison when a current problem-solving source or exercised project result supplies a better reusable family, when routine project replay repeatedly needs one omitted branch at the same frequency and action impact as the four, when two retained positions cease to change different actions, or when the four-position template produces more selection work than it saves. Until such evidence exists, call the cut provisional local rationale and never a computed frontier.
Pattern contributions and boundaries
- E.17.2 provides the four-position project authoring template and its concern distinctions. It supplies no L, declaration, reference, P edition, or membership occurrence. A project materializes those objects through the patterns below.
- Use E.17.0 for
U.ViewpointandU.Viewmembership,ViewpointConventionDependencyRelation,EpistemeViewpointConformanceRelation, and ordinary-use stops. - Use E.17.1 for catalogue L, local family declarations, and packaging by viewpoint references. E.17.1 admits no bundle U-kind.
- Use C.2.1 to identify every constituent episteme, Q, P, candidate E, assertion, and description.
- Use C.13 to construct collections and A.22 to select structures.
- Use A.6.3 only for optional source-to-receiving viewing construction; that construction does not grant view membership.
- Use A.3.1/A.3.2 for Methods and MethodDescriptions, A.3.4 for transformations, A.2 for local system-role kinds, C.3.2 for their
KindSignaturedeclarations, four-input classification judgments, and optional extensions, A.2.1 forU.SystemRoleAssignmentspecies and occurrences, A.2.2 for capabilities, A.2.7 for relations among system-role kinds, the direct responsibility pattern for responsibility, E.18 for transformation flows, and B.1.1 plus applicable module or interface patterns for dependency, module, and interface relations. - Use E.24.PUB for publication objects and relations, A.6.3.RT/A.6.3.RT.OE for expression construction and representation transition, and C.29 for a mathematical lens over independently recovered objects or relations.
- Use A.6.RCD to state or derive a needed relation claim, or return its blocker, when current predicates are insufficient.
Conformance checklist
- The pattern is used as an authoring template until one project supplies
<G_L, K_L, R_L>, ordinaryf_eng, fourr_* : U.ViewpointRefvalues, four P targets, and their resolution path. Labels or variable names fill none of those positions. G_Lcontains one local declaration claim block with the four bound references; it is not an inferred bundle U-kind, separate bundle entity, embedded viewpoint value, view, document, form, carrier, or publication occurrence.- Each reader-facing viewpoint name is only the project-local designator of P; designator, reference, P, any structured-branch S, and declaration position remain distinct.
- Each P passes all five E.17.0 viewpoint-membership conditions. It uses the self-contained branch by default; a C/Q/S witness is required only when separately versioned convention organization changes a named action.
- In a triggered structured branch, each witness names least-powerful constituent editions, every selected obtaining dependency occurrence, ordinary
Q_org, A.22-selected S, and ordinary P; optional dependency-use decisions and evaluations remain named-use neighbors. - Each concern episteme has one independently recoverable EntityOfConcern. Every relation claim names its predicate, participants, obtaining test, and applicable pattern; a multi-subject phrase is split or retained as a constraint claim, never promoted to a hidden group kind.
- Candidate E has one holon H as EntityOfConcern and becomes
U.Viewonly through obtainingEpistemeViewpointConformanceRelation(E,P). - A singular describing-use reference selects P without entering E/P identity or conformance; A.6.3 construction, declaration membership, naming, evaluation, rendering, and publication grant no membership.
- Every procedural failure or recovery claim has an operational subject and admitted Method, state-transition, or transformation-flow basis. A safety-analysis episteme or A.10 evidence path appears only for a safety-bearing claim or named reliance. Within that branch, use B.3 only when a named assurance claim for that use is current. A procedural view also qualifies as a MethodDescription when the same episteme describes how its own admitted Method subject is carried out under A.3.2. An assertion of performed Work requires its own basis. Allocation-responsibility views keep the local system-role kind, any four-input C.3.2 classification judgment, optional extension, assignment, performer System, and responsibility relation distinct. Module-interface views remain distinct from direct module relations or functional views.
- DDD Context Mapping remains a
U.Method; a product called Context Map is a separately identified episteme and becomes a View only through E/P conformance. - Every cross-view relation names its predicate, participants, obtaining test, and applicable pattern; a diagram edge, correspondence label, citation, shared holon, or common template is insufficient.
- Form expression, carrier bearing, five-participant publication and recurrence, rendering work, mathematical representation under C.29, and any publication-side viewpoint remain distinct and make no represented world-side relation obtain.
- Cross-project reuse is claimed only for the same resolved L, declaration, and member references. Equal TEVB position labels or independently filled templates establish no shared family.
- Later reuse for recognition under one viewpoint stops after the readable conformance judgment. An engineering answer must be supported by the claims in the selected views and any cross-view relation on which it relies (§§4.3, 4.6). It reopens full catalogue constitution only under the E.17.1:4.2 triggers.
E.17.2:End
Multi‑View Publication Kit
Status: Stable Type: Part E publication pattern Normativity: Normative unless marked informative
At a glance. Use E.17 when you need to publish one already accepted engineering account in one or more readable faces for different readers without changing its claims.
Use this when. The source account is already accepted for the present work, but a reader needs a plain explanation, technical card, interoperability card, or evidence-facing lane. The publication task is to expose the same account for that reader, not to create a new engineering claim or establish performed Work, gate passage, or assurance by presentation.
What goes wrong if missed. A readable face can silently add, widen, or hide claims. Authors can also make a four-face kit, a newly authored viewpoint or bundle, and an assurance dossier prerequisites for every publication even when one small face would answer the reader's question.
What this buys. Each reader with a current need gets the smallest useful face, the source remains recoverable, omitted detail and bounded use stay visible, and stronger identity or assurance apparatus is added only for a downstream use that needs it.
First action. Point to the current source account and the engineering object or relation it describes, name the reader and what that reader must be able to understand or do, and choose only the face or faces needed for that use. Resolve an existing viewpoint when one already fits; do not author a new viewpoint or bundle merely to start publication.
First output. One useful publication face, or the smallest necessary set, that names the source, intended reader/use, what it preserves or omits, and how to return to the source. No ClaimGraph, formal profile, viewpoint bundle, evidence package, or four-face completion is required for this ordinary result.
Working publication move. Select the current source and the minimum face set for the named readers. Copy or conservatively arrange only source-backed claims. Try the work each reader is meant to do using the face and the sources available in time. If the source supplies only part of the commissioned answer, identify the supported contribution and return the missing question to the source author or the person commissioning the work. Publish that contribution on its own when a permitted partial delivery serves the reader. For a commission on the complete set, keep the missing grounds visible before issuing it as complete. If a face will carry safety, release, evidence, cross-context, or other consequential reliance, strengthen only that face with the relations and records that the reliance needs.
Ordinary formality rule. A source pointer, reader/use line, readable face and visible omissions can suffice for ordinary publication. For source-finding, test whether the reader can reach the source. For explanation, test whether the reader can recover the relation and grounds that answer the question. Add formal identities only when the use depends on them.
When reliance changes the engineering move. Identify the source edition. Resolve the viewpoint and E.17.0 conformance only if U.View membership matters. Identify the E.24.PUB publication occurrence, form, carrier, and bounded use when their identities matter. Cite the concrete evidence, gate, release, provenance, or assurance record that carries the downstream claim. These additions do not turn the face into that record.
Stop condition. Stop as soon as every reader with a current use has a useful face that preserves the needed claims and exposes its return to source. Do not create unused faces, fields, viewpoints, bundles, or assurance records for kit completeness.
When a face carries another claim. Use E.17:5.1d only when a publication-facing unit begins to carry a distinct work, evidence, gate, approval, status, explanation, comparison, or reduced-use claim. Ordinary publication of a source-backed face does not require that boundary map.
At the first screen, keep only the current source, named reader/use, minimum useful face set, visible omissions, and return to source.
Not this pattern when. Use A.15.1 for a performed-work claim, A.10 for an evidence or provenance path, B.3 for assurance or engineering justification, A.20/A.21 for constraint or gate decisions, A.7 when carrier, publication, and Work are conflated, and the relevant release or authority rule when that is the problem. E.17 only publishes the already accepted account and keeps those downstream claims separate.
Tech-name:
MultiViewPublicationKit(MVPK)
General publication-face form: In E.17,
MVPK facerefers by default to the reusable publication-form rules F. A constructed face expression X follows those rules and supplies content. Name X when identifying what was rendered, compared or made available. The selected source episteme, any separately constructed receiving episteme, the bounded-use declaration, the publication occurrence and the carrier remain different objects. Name the one you mean. A face is not a U-kind and does not become aU.View, evidence, assurance, gate decision, work occurrence, authority, or release permission by its label or readability. Source-edition, viewpoint, scope, occurrence, form, carrier, pin, or downstream-record identities are stated when they change the receiving use. USM binding (overview): when publication-scope identity must travel,U.PublicationScopeunder A.2.6 carries that bound; an ordinary bounded-use line can precede that record. See §5.0. Episteme-side view position. MVPK can publish an already recognizedU.View, or it can publish another selected episteme without claiming view membership. WhenU.Viewmembership is material, E.17.0 tests that same episteme against theU.Viewpointepisteme resolved frompublicationViewpointRef;PublicationVPIdis the viewpoint episteme's designator, not the reference. A.6.3 construction, E.17.0 conformance, E.24.PUB publication occurrence/expression/form/carrier, and C.29 mathematical representation remain separate relations.
Intent
Let a practitioner publish the few readable faces that current readers need from one accepted engineering account, without adding claims or turning publication metadata into engineering authority. The optional morphism profile keeps the earlier compositional publication tests for uses that publish morphisms. Ordinary publication does not require this profile.
Problem frame
- Different readers often need different parts or presentations of the same accepted account. One or two faces can suffice for a current task.
- Informal renderings can change the meaning, hide omissions, or sever the return to source. Composite morphisms can also lose traceability when their publication claims are used compositionally.
PlainView,AssuranceLane, and a packagedviewpointRefare easy to overread asU.Viewmembership, assurance, or conformance even though none establishes those claims by itself.- Publication identity, pins, carrier relations, and evidence references matter for some receiving uses, but putting all of them before the first readable face makes ordinary publication unnecessarily hard.
MVPK therefore starts from the current source, reader/use, and minimum useful face set. It then adds viewpoint conformance, E.24.PUB occurrence/expression/form/carrier identity, pins, bridge records, evidence, or assurance only when a named use depends on those distinctions.
The optional morphism profile retains the functorial publication discipline for epistemes used as descriptions, including those admitted for specification use. Part E is conceptual: no machine-exchange formats are specified here.
Problem
- Semantic drift in publication. Unchecked presentations introduce claims not present in the C.2.1 source epistemes, including those about the arrow in the optional morphism profile. Each such episteme, including one admitted for specification use, keeps its claim content, EntityOfConcern, and effective
ReferenceScheme. Publication form, viewpoint reference, scope, or carrier supplies none of those identity discriminators. - Non‑compositionality in the morphism profile. Publishing
g∘fyields faces that do not match composing the faces offandg. - View, viewpoint, and face confusion. A template or face is treated as the view or viewpoint, without conformance between the selected claim-bearing episteme and the viewpoint episteme.
- Unpinned numbers. Numeric claims lack unit, scale, reference‑plane, and edition pins from Part F or Part G, undermining auditability.
Forces
Solution — the MVPK Kit
Publication-scope and face-profile binding (normative)
- Ordinary selection. Start with the current source account, intended reader/use, and the smallest publication-form set needed now. A single form is a valid result. Adding another form requires another current reader/use or a material distinction that the first form cannot carry safely.
- Bounded use before exact scope. Alongside each selected publication form, state the separate bounded-use declaration in ordinary prose. Identify the applicable
U.PublicationScopeunder A.2.6 when scope identity must travel across publication, comparison, exchange, dispute, or reliance. The scope establishes neitherU.Viewmembership nor permission, evidence, work, assurance, or release, and it encodes neither the selected source, viewpoint, publication-form profile, Publication Characteristics, nor carrier. - Resolve before authoring. Reuse an existing viewpoint when its concerns and rules fit the reader/use. Author a new reusable viewpoint, or create a project-local family declaration under E.17.1, only when the current need cannot be served truthfully by an existing viewpoint or a simple bounded publication face. E.17.0 tests viewpoint conformance. Use E.17.1 to identify the catalogue edition, ordinary family designator, local declaration claim block, and needed
U.ViewpointRefsubset. - Optional profile. A formal MVPK profile fixes the publication-form designators, any declared partial order, Publication Characteristics and pins, and any cross-context or reference-plane constraints. These fields apply only to the optional formal or load-bearing branch, not to the ordinary first result.
- Canonical labels.
PlainView,TechCard,InteropCard, andAssuranceLaneare MVPK face designators. None identifies aU.View,U.Viewpoint, evidence object, assurance result, or gate. Use only the designators needed by the current readers. MVPK-Max is the optional profile in which all four have a use.
Terminology (normative)
- View (
U.View): the same C.2.1 episteme individual for whichEpistemeViewpointConformanceRelation(E,P)obtains under E.17.0 for theU.ViewpointepistemeP. A publication-form label,viewpointRef, direct authoring, A.6.3 construction, or publication occurrence does not establish that membership. An ordinary publication form exposes its current source reference and separate reader/use declaration; addpublicationViewpointRef, conformance relation, scope, occurrence, carrier, or pins only when those identities change publication or reliance. - Publication vs expression vs bearing vs presentation vs rendering vs representation (guard):
- Publication occurrence = the E.24.PUB relation among E, audience declaration A, bounded-use declaration U, expression X and independently identified carrier C. Recover their identities when availability, recurrence, exchange or reliance depends on them. Preparing an ordinary face can stop with its useful expression; source construction and viewpoint conformance retain their own predicates.
- Expression =
PublicationExpressionRelation(E,X,U): X conveys the required claims of E for U. A selected reusable form F constrains X throughPublicationFormConformanceRelation(X,F).PublicationFormExpressionRelation(E,F,U)states that an adequate X conforming to F exists. That three-participant occurrence can continue when the witness X changes. - Carrier bearing =
PublicationExpressionBearingRelation(C,X): an actual state or realization of independently identified C yields X.PublicationFormBearingRelation(C,F)states that C bears an X conforming to F. Form-mediated availability requires the same X to witness expression, bearing and conformance. - Presentation = the order, grouping and emphasis arranged for a recipient in an expression. In a source-preserving MVPK publication, these choices must preserve the required source claims and their qualifications. A different arrangement can change X while preserving E.
- Rendering produces or displays an expression through a carrier. Name the rendered construction when comparing X and the system-performed Work when its production matters. Its decoding, layout or graphical choices can affect recoverability and need the preservation test for the selected use.
- Representation relates an independently identified construction to its subject through a recoverable correspondence. A.6.3.RT and A.6.3.RT.OE govern constructing and using that expression; C.29 supplies the mathematical-lens branch. The correspondence and intended operation determine what can be returned to the represented subject.
- Architecture-description mapping note. An architecture viewpoint maps to one
U.Viewpointepisteme.PublicationVPIdorEngineeringVPIddesignates it and aU.ViewpointRefresolves it. An architecture view maps to one episteme that passes E.17.0 conformance. An MVPK face is the separate publication form through which that episteme may be exposed for a separately declared bounded use and, when material, in an exact publication occurrence. - Publication work: Build, rendering, upload, or delivery may be
U.Workperformed by the System identified through A.13. When such Work is current, A.15.1 independently identifies the Work, performers, Method, time, and containing System. Add F.6 only when the publication account expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment. Missing or failed F.6 alone leaves independently admitted Work intact. Name a carrier relation separately only when the claim or downstream use depends on it. - Viewpoint (
U.Viewpoint) - a claim-bearing episteme edition recognized under E.17.0's dependent-kind rule. Resolve an existing viewpoint when it already states the relevant concerns and conformance rules. Author a new reusable viewpoint, or create an E.17.1 local family declaration inside an identified catalogue episteme edition, only when the current reader/use cannot be served truthfully without that separate result. The declaration packagesU.ViewpointRefvalues. The declaration is not another bundle entity.PublicationVPIddesignates the viewpoint episteme.U.ViewpointRefresolves it. - Explanation-use profile values. An existing publication form can be paired with a bounded explanation-use profile value such as
SourcePinnedExplanation,SourceLinkedExplanationReconstruction,DidacticRetelling, orSpeculativeRetelling. The profile value is neither the form nor a new face, explanation, or carrier-rendering kind. Pins, provenance references, and no-new-A.6.B-boundary-claims discipline apply only when the source, transformation, or receiving use makes them material.
Episteme-publication relation-position binding (normative)
For functional-description publications, E.17 covers only the publication relation.
Publication relation position. A principle scheme, functional diagram, comparison table, screen, export, scenario, explanation, or code-like method description can help interpretation, source-finding, comparison, selected-method inspection, or work-planning preparation.
Unsupported neighboring claims. The publication does not by itself assert performed U.Work, a work claim, gate passage, evidence, assurance, engineering justification, supervisory relation or control relation, authority, release permission, or a new transformation-flow kind.
Interface and protocol proximity. When interface, protocol, schema, boundary, or API wording appears beside a functional-flow description, keep that operational claim with its project claim set and exact reference. Apply the boundary, interface, protocol, or transformation rules in A.6.B, A.6.C, or E.18 as the concrete claim requires; do not absorb it into the publication by layout proximity.
Retargeting. If the publication changes the EntityOfConcern or retargeting target from an already described component, recovered transformation, method, work occurrence, transformation-flow structure, material U.Entity, or source claim into a functional, control, or flow architecture claim, this is not a same-entity publication-use change. Use A.6.4, OntologicalReframing, or E.18 as applicable.
Source recovery. When a requested use requires a project-side object or relation beyond the publication face, first recover the existing reference that carries that claim. The entries below are separate checks. Listing them here establishes neither a single route nor a generic relation between the patterns. Choose those that apply to the claim at issue:
- source wording, publication construction, carrier-relation construction, source relation, project-side reference, or explicit non-use disposition under
C.2.P; - appearance-based reliance repair under
A.15.4; - project
U.Method,U.WorkPlan, or work-result record underA.15; - evidence and provenance path under
A.10; - engineering-justification record under
B.3; - constraint or gate decision under
A.20orA.21; - supervisor-subholon feedback record under
B.2.5, control-structure-view record underC.30.LCA, or a record of the architecture claim underC.30or selected-structure claim underA.22, as applicable; - carrier, export, OCR, or front-end distinction under
A.7, followed by the applicable carrier, front-end, or Work pattern; - same-entity textual relation under
A.6.3.CR; - representation relation under
A.6.3.RT; - reduced-use-rendering relation under
A.6.3.CSC.
No backdating. Recover the grounds for the claimed earlier object or relation from the available sources. A present account may establish what happened earlier without claiming that its record already existed then. Do not invent an earlier source, permission, decision, or relation. If the required grounds cannot be recovered, retain the specific gap and withhold only the dependent claim or effect. A prospective repair, decision, or work-plan request can then supply the next action.
Ordinary orientation and source-finding can stay as an inline note.
Functional-description guard (CC-MVPK-FD). A functional-description publication separates the source U.Episteme or episteme-side U.View, the particular expression and the reusable publication form (the MVPK face), any present carrier or rendering work, the separate bounded-use declaration, and unsupported neighboring use. The guard applies only when a functional-description face is present. It is not the first universal MVPK conformance gate.
MVPK inherits the distinction among U.Episteme, contingent published-episteme use, publication occurrence, particular expression, reusable publication form, U.View, carrier, and authority-reference relation. It introduces no durable published-episteme kind or other generic semio kind. A publication face does not define another relation's claim, supply authority, or become the source claim merely by being published. Use the source or authority relation when one is current.
When a morphism publication is encountered or reused, name only the relation positions needed by the current use:
- the selected source
U.Episteme,Depisteme, orSepisteme edition and the claims the publication exposes from that source. Identify its ClaimGraph only when claim identity must travel; - the
PublicationExpressionRelationoccurrence among that selected episteme, expression X and bounded-use declaration, with form conformance when the use depends on F; - the
PublicationExpressionBearingRelationoccurrence between independently identified C and X; use the form-mediated predicates when that quantified granularity answers the receiving question; - the five-participant
EpistemePublicationRelationoccurrence when that selected edition is available to the declared audience for the bounded use; - the selected episteme's independent E.17.0
U.Viewmembership when the publication use calls it a view, plus any separate A.6.3 construction history when current; - the system-performed carrier or rendering Work; any A.10 evidence/provenance path or G.6 path citation needed to replay it; and any G.11 currentness result, only when those neighboring facts are current; and
- the project-side object, reference, or authority relation when the next work or reliance claim depends on it.
Changed claim content, EntityOfConcern or effective reference scheme identifies another episteme under C.2.1. A later-edition claim also requires its source and continuity rule.
Re-evaluate expression, bearing and form conformance when their participants or predicates change. Changing any fixed E/A/U/X/C participant identifies another exact publication. For form-mediated availability, retain the common expression witness.
For broader continuous availability, retain the permitted alternatives under E.24.PUB:4.2.1: changing one expression or losing one copy need not interrupt the broader availability. A file rename alone does not decide carrier identity.
Across these cases, a reader can recover which object or relation is available for reliance: the episteme claim, the published form, the view, the carrier, the typed project-side FPF kind and reference named by value, or the authority-reference relation.
A dashboard tile, generated explanation, card face, credential view, or carrier can guide source-finding, but it does not by itself establish the source claim or effect, gate decision, evidence relation, assurance claim, local system-role kind, separate System-classification judgment, assignment occurrence or state, direct status predicate, responsibility or authority predicate, Work occurrence, or permission.
If its source uses role or status without making one of those claims clear, treat that phrase as unresolved recognition wording and route it through E.10.ROLE. Then use the recovered direct pattern or say which governing rule is missing.
Source-exposure rule. A publication-facing unit or its carrier exposes a project-side object only when that object and its direct relation are recoverable. Appearance or production history alone establishes neither. If a SpeechAct, GateDecision, evidence path, credential or status source, U.Work occurrence, U.Episteme, or publication occurrence is recoverable, rely on that object and relation. Otherwise use the face only for orientation or source-finding.
Shared source-relation and bounded-use vocabulary
Use this vocabulary when a publication face, rendering, generated text, comparison note, narrower-use rendering, source-finding cue, or authority-looking display can be overinterpreted as carrying a wider source relation or bounded-use permission than it carries. The vocabulary names the source relation or bounded-use value for one claim or use. It does not instantiate evidence, gate, assurance, work, commitment, speech act, decision, release, or authority.
Patterns can use shorter local field names such as sourceRelationStatus, explanationSourceStatus, or representationValidityStatus when the local object is clear. Comparative patterns split source-relation status from comparative-relation status instead of using one overloaded field. The local field remains interpretable through the vocabulary above, and the bounded use is named beside it when downstream reliance could change.
For ordinary use, name only the status distinction that changes the next bounded use. The common light states are source-pointer-only, source-relation-unknown, source-relation-not-needed, source-not-recoverable-here, admissible-for-this-use, downstream-use-forbidden, and reopen-trigger-present.
The vocabulary is neither an ordered source-stage scale nor a source-record or authority taxonomy, and it does not substitute for evidence, assurance, gate, or work records. A missing source relation blocks only the unsupported use. It does not prove the underlying world claim false. If independent-verification-present is relied on, name the separate evidence, assurance, decision, work, or bridge record that supplies the independent basis.
Shared use-boundary terms
Use these terms when a publication face, rendering, narrower-use rendering, explanation, comparison note, source-finding cue, or authority-looking display can be interpreted beyond its named source relation. Define these terms once, in this section. In local patterns, use the same terms and link here.
Compact boundary aid for the present claim or effect
When a publication-facing unit, publication face, rendering, narrower-use rendering, explanation, comparison note, dashboard tile, credential view, status view, carrier, or generated unit creates more than one possible interpretation, separate the claim being made or effect being used now and cite the source relation that makes that claim recoverable. Apply the aid to the present claim or effect. It does not classify the whole unit. The same unit can expose several typed records. Handle one claim or effect at a time instead of assigning one source relation to the whole unit.
Mixed-case precedence. When several publication-use patterns appear possible, repair the smallest unstable interpretation that changes the current bounded use before applying a neighboring pattern whose claim or effect is present:
- If one local head is the only unstable part, apply
E.17.AUD.LHRorC.2.Pand stop when the repaired sentence names the local kind, relation, and bounded use. - If the bounded
PublicationUnitor the interpretation of its primary subject is unstable, applyE.17.AUDorE.17.AUD.OOTDbefore usingE.17.ID.CRorE.17.EFP. UseEntityOfConcern(E)for that subject only when the unit carries one identified claim-bearing epistemeEand both name the same entity. - If the unit is stable and the present problem is comparison overread, apply
E.17.ID.CR; useF.9,C.11,A.20, orA.21only for an equivalence, recommendation, selection, decision, gate, or release claim. - If the unit is stable and the present problem is explanation overread, apply
E.17.EFP; useA.10,B.3,A.20,A.21, orA.15.4only for an evidence, engineering-justification, gate, release, work, or reliance claim. - If the present problem is a durable reusable name, UTS row, Core-facing term, or cross-context naming relation, apply
F.18; otherwise keep the lighter local repair pattern.
Evidence-path boundary. An A.10 evidence/provenance path, including one that cites attestation, freshness, or a G.11 currentness result, carries only the claim it instantiates, identified by value. It does not approve or authorize work, pass a gate, perform work, supply release permission, or raise assurance or engineering-justification use unless the typed project-side FPF kind and reference named by value that carries that downstream claim is also instantiated, such as A.15.4, A.15, A.20, A.21, or B.3.
Gate-display boundary. A dashboard tile, status view, or release screen exposes a gate decision only when the GateDecisionResult is recoverable with its decision subject, bounded action, profile application, check-application references, decision, action consequence, scope, window, and recheck condition under A.21.
Add evidence, currentness, replay, or audit references only when the applicable rule or receiving use requires them. Without the required result and basis, the display remains orientation or source-finding only. Color, label, layout, or proximity establishes neither gate passage, release permission, nor performed Work.
Local review fields are not FPF kinds
Local review fields and values in CR, RT, CSC, EFP, ID.CR, or a neighboring publication-use pattern are local aids for one case. They are not U.Kind, RelationKind, evidence, gate, authority, work, publication face, or another project-side object unless the pattern that defines that object establishes its membership. When a local field starts carrying such a claim, cite the object and say whether the cited pattern defines it, constrains it, or supplies its test.
Shared anti-overread invariants for publication-facing units
Use the FPF pattern that defines, constrains, or tests the claim being made or effect under use. Keep any local review field local, preserve reduced bounded use, and address only the unsupported wider claim or effect through the source relation it requires.
Source-relation minimality. Name the smallest direct relation sufficient for the live use. A source reference, publication occurrence, evidence path, engineering-justification record, gate decision, and release decision are different objects or relations. Choosing one licenses none of the others. Do not apply A.10, B.3, A.20, or A.21 when the use needs only source-finding, orientation, or inspection of an existing source episteme, publication occurrence, or status-register entry.
Local repair vs publication redesign. A local epistemic precision repair is enough only when it can preserve the current publication face or PublicationUnit while fixing one head, boundary, source relation, bounded use, explanation class, or unsupported downstream claim. If layout, grouping, visual emphasis, comparison arrangement, generated explanation, hidden source limitation, or mixed EntityOfConcern packaging still induces overread after the local relation is repaired, create a redesigned publication face or PublicationUnit instead of adding warning text around the misleading form.
Most-likely careful interpretation constraint. Design and word a publication-facing unit so its most likely careful interpretation does not exceed its named source relation and bounded use. A visible Approved head needs a recoverable approval basis under its governing rule, including the gate result when gate passage is claimed, or a different head. Sorted output needs its comparator or sorting relation visible if no recommendation is intended. Generated explanation separates inferred links from pinned source claims by wording, label, or source reference.
Visual cue claim pressure. Layout, order, color, prominence, icon, grouping, and proximity can imply evidence, readiness, preference, equivalence, approval, or verification. Green can suggest readiness, top position can imply preference, and grouping can imply equivalence. Proximity to evidence can suggest an evidence relation, a badge can suggest approval, and a lock or checkmark can suggest verification. If that implication would change the next action, recover the evidence, assurance, gate, decision, recommendation, bridge, approval, or other record or relation that carries it, or redesign the face so the unsupported overread is no longer invited.
Extraction survival. When a PublicationUnit is excerpted, quoted, screenshotted, summarized, copied into a tutorial, retold by a generator, or moved to a slide, it keeps only the claims, source pins, boundary line, references named by value, and bounded use carried in that extracted unit.
Any use that depended on hidden neighboring context is lost unless that context is carried by source pins, a boundary line, or a reference named by value.
A dashboard screenshot does not carry the underlying gate record, a quoted comparison row does not carry the full comparator or sorting relation unless that relation is included or referenced, a copied explanation paragraph does not carry source pins unless pins remain recoverable, and a pattern excerpt does not carry the whole pattern boundary unless the excerpt states or cites it.
No-extra-pattern case. If a publication-facing unit has bounded use only for ordinary orientation, learning, source-finding, review, comparison, or planning preparation, and no operative work or reliance, evidence, gate, assurance, bridge, source-dispute, or release claim is present, keep the existing publication source relation and proceed with ordinary use.
Pattern-inflation anti-pattern. Do not apply a neighboring pattern merely because the publication-facing unit resembles a worked example. Apply the neighboring pattern only when a claim being made or effect changes the next available project move.
Strategic overread invariant. Apply the same anti-overread rules whether the misleading interpretation is accidental, conventional, incentive-driven, or intentionally induced by publication design.
Green status color without GateDecisionRef, reviewed-looking wording without approval, selective source links without operative-claim source relation, comparison ordering without selection decision, hidden caveats behind a source link, or pins for trivial claims beside unpinned causal linkage do not create evidence, gate, decision, assurance, work, release, or bridge relation by design pressure.
Carrier-travel invariant. A copied, exported, screenshotted, summarized, generated, translated, or re-rendered face can carry orientation or source-finding cues. It carries no evidence, authority, gate, approval, engineering justification, work, currentness, or release relation unless the corresponding object and relation remain recoverable for that use.
Derivative-chain decay. A second-order rendering inherits at most the bounded use that is explicitly carried from the prior source relation. It does not inherit source faithfulness, evidence relation, currentness relation, authority-reference relation, gate decision, work relation, or reliance relation by default.
Publication-face snapshot and refresh identity. A face can keep the same layout, name, or carrier while its source pins, data window, source-relation status, currentness, EditionId, or bounded use changes. Visual sameness is not source, evidence, or use-boundary sameness.
Beyond orientation, identify the face edition or snapshot, the source pins or data window that still carry the claim, and any changed bounded use. If those cannot be recovered, use the face only for orientation/source-finding or reissue it from the source under E.17 and the concrete downstream rule that the new use needs.
Claim-level source relation only. Do not assign one whole-unit source-relation status unless every operative claim in that publication-facing unit has the same source relation named by value for the same use and unsupported downstream uses are explicit.
Modality and deontic-force preservation. Publication-facing transformations preserve possibility, obligation, permission, recommendation status, decision status, confidence, scope, and temporal window when those values change the claim or use. If one changes, narrow the bounded use or apply the concrete definition, constraint, decision, evidence, work, gate, or authority rule that carries it.
Comparison does not become recommendation or decision. Explanation does not become evidence, and a face does not become authority. Do not use a publication unit to assert a downstream effect without the grounds for that effect. Linking a source does not establish its availability for reliance, and looking ready does not establish gate passage.
This preservation rule also applies across extraction, translation, screenshotting, summary, and generated retelling. A translated permission is not wider permission, a screenshot of approval-looking display is not an approval record, a summary of evidence is not an evidence path, and a generated retelling of a decision is not the decision record unless the source relation that makes the operative claim recoverable by value and source pins survive in the new publication-facing unit.
Reader position is not a project system-role kind or assignment. Reader position, audience, target user model, verifier position, review-reader position, and learner position do not become project system-role kinds, U.SystemRoleAssignment occurrences, decision authority, gate authority, issuer relations, responsibility relations, or Work contexts by publication.
If any of those values is current, cite its typed project-side reference and direct predicate separately. Otherwise say which governing rule is missing rather than inferring it from a reader label.
Source-gap states. When a required source relation is missing, identify the gap. The source may be unnamed, named but unavailable, or available but unused. A used source can be insufficient. A source can also be stale, outside its window, or contradicted. Another possible gap is a mismatch among the source-maintenance System, any maintenance Work whose performer is recovered through A.13 and which A.15.1 admits independently, the status register, and a separately established responsibility relation. Add F.6 only when the source-maintenance comparison expressly consumes precise assignment-bound attribution; its failure alone leaves independently admitted maintenance Work intact.
Assignment establishes neither source-maintenance responsibility, classification, nor status-register authority. Block only the unsupported effect and keep any reduced bounded use available.
Measure and display overread. Before relying on a displayed number, score, percentage, color, rank, confidence, similarity, or status, recover the claim and its source. Retain the characteristic or criterion, scale, scope, window, aggregation rule, calibration, or evidence basis only where the applicable pattern requires it or it changes the value's interpretation. A status or ordinal judgment does not become a measurement by display. If the basis required for the intended use is missing, keep only the supported orientation or source-finding use.
Use A.10 when the relied-on source-to-use account is implicit or must remain replayable, B.3 for a named assurance claim, A.20/A.21 for a constraint or gate claim, and F.9 only for a separately claimed Bridge or its bounded use. F.9.1 may add a stance note about an already constituted claim.
World-contact stop. When a source update, revocation, policy change, holon-state change, incident, model update, environmental change, or new observation changes a premise required by a downstream use, revalidate that affected use before continuing it. Refresh the source, reissue the publication, or recover a changed project-side result as that use requires. Reuse unaffected results. A changed label or edition alone does not establish changed meaning or invalidate every publication use.
Functional-description boundary. A functional, architectural, descriptive, representational, or explanatory fit claim creates no permission, obligation, approval, gate passage, release relation, performed-work evidence, or engineering justification. Those uses need the separate work, authority, evidence, decision, gate, release, or assurance object and relation that carry the claim.
Mixed bundle no-shared-evidence-relation rule. A bundle with source-pinned, reduced-use, speculative, didactic, comparison, and evidence-facing parts is not interpreted under one shared evidence relation or use-boundary value borrowed from another member. Each operative claim keeps its own source relation and unsupported downstream use.
Educational usefulness. Didactic, onboarding, tutorial, and workshop material can help readers orient themselves. It is not evidence, gate passage, approval, work occurrence, engineering justification, release permission, or bridge relation.
Comparison exposes conflict; it does not adjudicate it. A comparison note can expose contradiction, asymmetry, different foregrounding, or residue. It does not select an option, approve release, pass a gate, or create bridge or substitution relation unless the corresponding C.11, A.20/A.21, F.9, or other decision relation carries that result.
Same publication-facing unit, multiple interpretations. A green release dashboard can serve as an MVPK face for source-finding, expose an A.10 evidence/provenance path with a G.11 currentness result when those sources are recoverable, or display an A.21 gate decision when its result is recoverable. The face remains distinct from those results; without the required sources, its release cue is unsupported.
A generated comparative explanation can be an E.17.EFP explanation-use case, an E.17.ID.CR comparison case, an A.6.3.CR generated-summary case, or source-finding only. It is never all of those under one shared evidence-relation class or bounded-use value by fluency alone.
Archetypal publication-use cases. Use examples like these to recognize publication uses:
- Green dashboard tile. A tile says
Model ready. Treat the tile as thePublicationUnitwhen that tile carries the present release overread. Recover the A.21 gate result and the source basis required by its profile application before using the tile as gate passage. A release claim also needs its own decision or authority basis. Without those, retain source-finding and status orientation only. Green color or placement supplies neither result. - Generated explanation with source links. A generated text explains a method and cites sources. The explanation rendering is not source replacement. For an operative claim, rely on a link only when it pins that claim. Retain the subject grounds required by any Work or reliance claim; use
A.10when its source-to-use account is implicit or must remain replayable. If the required grounds are absent, keep the rendering as reader help. If the rendering is deliberately reduced-use, useA.6.3.CSC. - Comparison table. A table compares two methods and places one first. Ordering is not selection. The comparator or sorting relation, source references, shared review frame, and unsupported downstream claim remain visible. Choice or decision needs
C.11; equivalence or a Bridge needs F.9, while F.9.1 may add only an optional stance note about an established bounded-use claim. - Unrecovered source wording. A draft uses source-object wording, undeclared interpretive-view shorthand, or generic unit wording without naming the FPF kind. Recover the FPF kind and relation positions instead of minting source-relation pseudo-kinds or undeclared interpretive-view pseudo-kinds. Use
PublicationUnitonly when a bounded reader-inspected unit inside a publication is present; otherwise use the episteme, view, publication, carrier relation, section of a named non-pattern FPF publication form whose reader-help function and reference are recoverable,A.6.Prelation claim, or typed project-side FPF kind and reference named by value. - Translated tutorial. A translated tutorial can improve reader access to an FPF pattern. It is a derivative rendering, not the original source. Operative claims need source mapping for reliance, translated heads can need
E.17.AUD.LHRorC.2.P, andF.18is present only when durable naming, UTS, Core-facing, or cross-context naming work is intended.
What the neighbouring patterns prevent. Use the examples below to see which claims the publication can support and which need another basis.
Blocked overread with useful publication use remaining.
-
A comparison table appears to select option B. Block the selection interpretation when no
C.11ChoiceResult, decision record, or visible selection relation exists. Useful publication use remains: use the table as a bounded comparison underE.17.ID.CR, or applyC.11when selection is intended. -
A green dashboard tile appears to permit release. Block that interpretation when the gate result, its applicable profile and required check results, or the release decision or authority basis is not recoverable. Useful publication use remains: source-finding and status orientation, followed by inspection of the missing source when release is intended.
-
A generated explanation appears to establish a causal relation. Withhold that conclusion when its required causal grounds are absent or insufficient. Use
C.28to establish the causal claim; useA.10when its source-to-use account is implicit or must remain replayable, andB.3for a named assurance claim. Retain any supported explanation or source-finding use. -
C.2.Pprevents the wrong object from being treated as source, the wrong relation from being treated as source relation, and a loose phrase from being treated as an FPF kind. -
E.17.AUDandE.17.AUD.OOTDprevent action on a publication unit whose primary subject, carried publication move, or outside boundary shifted silently. -
E.17.ID.CRprevents a comparison unit from being used as decision, equivalence, bridge, evidence, or release source relation. -
E.17.EFPprevents fluent explanation from laundering unsupported claims into reliance, assurance, gate, or evidence use. -
E.17MVPK prevents a readable publication face from being treated as evidence, gate, work, authority, or release source relation by display quality. -
F.18prevents a local name from becoming global identity without context, kind, lineage, and bridge or cross-context naming relation.
Anti-escalation examples. Do not apply a neighboring pattern when the unit makes none of the claims that pattern governs:
- Do not apply
F.18when a one-off local phrase repair restores the local kind, relation, and bounded use without minting a durable reusable name. - Do not apply
A.10when the publication-facing unit is not being used for reliance, evidence, provenance, currentness, or claim-bound evidence relation. - Do not apply
A.21when a dashboard tile is merely status orientation and noGateDecisionRefor gate profile is present. - Do not apply
F.9when a comparison does not claim sameness, substitution, bridge relation, or cross-context equivalence. - Do not apply
E.17.EFPwhen the text is only a same-entity rewrite or representation change underA.6.3.CRorA.6.3.RT.
Concrete reopen trigger. Name the condition and the nearest source-bearing side or the concrete definition, constraint, test, decision, evidence, work, or authority relation to revisit. A vague reopen if needed does not preserve the source relation.
Declared publication-face kind values at Part E
Part E restricts publication-face kind values to the literals publication face/form and interop publication form. PlainView, TechCard, InteropCard, and AssuranceLane are face designators, not additional U-kinds or automatic U.View memberships.
USM linkage (normative when exact scope identity is current). An ordinary face first states its bounded use. When that bound must be cited, exchanged, compared, or relied on independently, identify U.PublicationScope under A.2.6.
For a face whose admissible use relies on claim c in episteme E, PublicationScope(face_E) ⊆ ClaimScope(c). For a capability-description episteme E, the EntityOfConcern is holder System H (A.2.2). If the face presents a qualified holder-ability claim C about H from E for work admission, PublicationScope(face_E) ⊆ WorkScope(C), where WorkScope(C) is the work-condition basis designated by C. Any relied-on ClaimScope remains a separate constraint. When both bounds apply, PublicationScope(face_E) ⊆ ClaimScope(c) ∩ WorkScope(C). Neither inclusion grants permission to perform work or proves that work occurred.
A cross-context semantic claim separately retains its F.17 endpoint senses, F.9 Bridge, bounded-use claim, and any current A.10 or B.3 reliance result. An optional F.9 CL summarizes evidence strength. It is not a relation or use condition.
Publication-face naming discipline.
- The only
publication-face kindvalues are publication face/form and interop publication form. - Concrete face designators end in ...View, ...Card, or ...Lane only within this family; the suffix does not establish kind membership.
PlainViewis a face designator, not aU.Viewclaim. UseU.Viewonly for an episteme that passes E.17.0 conformance.AssuranceLanecan expose evidence bindings or pins, but it is not an assurance claim, evidence-sufficiency result, confidence verdict, gate, or release permission.- carrier, bearer, and holder retain their exact carrier or relation meanings and do not name a view or publication entity.
- Any legacy
ViewFamilyIdtoken is only the ordinary family designator used to retrieve one local E.17.1 declaration claim block inside the identified catalogue episteme edition. It is not a local id kind,U.View,U.Viewpoint, bundle-membership rule, or face kind.
Profiles select only needed faces.
- MVPK-Min: one selected face, normally a
PlainVieworTechCard-Lite, for one current reader/use. No assurance or interoperability face is implied. - MVPK-Lite: the minimum two or more faces needed by current readers; add
AssuranceLane-Liteonly for an evidence-facing use andInteropCardonly for an identified external consumer. - MVPK-SetReady: add the faces and pins required for replayable or external interchange; concrete exchange formats remain outside Part E.
- MVPK-Max: use all four designators only when all four reader/use obligations are current. It is not the default completeness target.
- A -Lite face removes only optional fields and retains every claim. Enrichment adds fields or pins without retracting, widening, or strengthening the source claim.
The publication-face kit
The optional morphism-publication profile uses representation-side constructors, not another viewpoint ontology:
FaceObj_sis the conceptual object component for publication face designators.F_faceis the finite set of publication-form designators. Its default formality order isPlainView <= TechCard <= InteropCard;AssuranceLaneremains independent.Emit_s(-) : EpMorphism -> FaceMorph_sconstructs candidate face expression content X under an identified expression domain forswhen this formal profile is current. If that work also constructs a different claim-bearing episteme, identify that receiving episteme and its source relation separately.- The coherence rules in section 6 and the pin policy constrain this representation-side construction.
InteropCardmay carry references that identify the interoperability concerns; concrete exchange schemas remain outside Part E.
FaceObj_s, FaceMorph_s, and Emit_s are local conceptual-form symbols. They are not public U-kinds, U.Viewpoint epistemes, U.View individuals, publication occurrences, or presentation carriers.
Result. MVPK(f,F_face) yields a C.13 collection of selected publication forms, each paired with a current source reference and separate bounded-use declaration. The construction also identifies the expression X made under each selected form, with its expression-domain equality and source/use binding.
If publication work constructs a different claim-bearing episteme rather than only an expression of the selected source episteme, identify that receiving episteme and its A.6.3 or other source relation separately; it is not the face.
For each actual publication, E.24.PUB separately tests expression, carrier bearing, selected form conformance and the E/A/U/X/C publication occurrence with its declared audience, bounded use, and recurrence rule. E.17.0 conformance, A.22 organization and any C.29 mathematical representation remain separate and are checked under their own patterns.
If a current use depends on organization among selected forms or among separately identified epistemes, select the U.Structure under A.22 from the corresponding collection, obtaining organizing relations, applied constraints, and use frame; the face collection supplies no structure by itself.
PromoteFace[s->t] changes publication-form explicitness; it changes neither episteme identity nor viewpoint membership and adds no claims. A change to X changes an exact publication when that X is its fixed participant. An unchanged E can remain continuously available through the permitted alternative expressions or carriers.
EntityOfConcern-side input and output vs publication (normative convention)
- Input and Output are signature-side declarations. The Input and Output sections of a morphism describe declared input and output data or episteme types under the morphism signature; they do not depend on any publication face.
- No independent Input/Output specification. In the optional morphism profile, keep the signature-side declarations authoritative. Provide an accessible return to those declarations, or carry the needed source-attributed declarations and material pins when the face must support use without that access.
- Use Signature only for signatures. Use Signature only when the named object is a signature under an applicable signature pattern, such as
U.Signature. On faces, use TechName or PlainName. - Comparison and ordering. A bounded source contrast under E.17.ID.CR retains its shared review frame and row-level comparison criteria. When a face publishes selection or an ordering, preserve the source-defined set or order semantics, make the comparator and any scalarization explicit, and cite
ComparatorSetRefwhere that formal comparator family is used. - Bridge and plane references. A semantic crossing cites its F.9 Bridge and separate bounded-use claim. A plane-dependent value cites its characteristic, selected
ReferencePlane, and applicable C.16 or A.19.CPM transfer or comparison rule. If B.3 is triggered and its assurance claim depends on an integration relation, retain that relation's B.3CLandΦ(CL)reference; infer no penalty from an F.9 Bridge or publication face. - Carrier references and relation positions. When the use depends on them, name the carrier reference, any A.10 evidence/provenance path or G.6 path citation, and the G.11 currentness result; keep
U.Workoccurrences distinct from epistemic claims via relation positions. - Publication is not execution. Publication morphisms carry no time or resource semantics; any build, render, or upload work is separate
U.Work.
Pins and local publication-profile fields (normative)
Intent. Make publication-time numeric, comparison, evidence-reference, and crossing claims explicit and auditable without minting another public kind or importing geometric metaphors. Do not use “axis” as a generic name for a profile field. This is an optional formal branch. An ordinary publication form retains only the source pins needed to interpret claims used from it.
Reuse existing value definitions. A measured aspect is an admitted U.Characteristic with its membership predicate and Scale. A product of characteristic slots is an admitted U.CharacteristicSpace.
A publication-profile field, abbreviated locally as a PC field, is only a field in the selected E.17 formal profile: it points to one of those admitted values or to another value or relation whose meaning is already defined. A PC field is not a U. kind, characteristic, evidence relation, comparator, bridge, or viewpoint by being present.
Initial local fields. Use only the fields that the selected publication form and receiving use consume:
- PC.Number — a displayed numeric or comparable value of a
U.Characteristic; name the characteristic reference and its unit, scale, reference plane, and edition when they affect interpretation. - PC.EvidenceBinding — a reference to an existing A.10 evidence path, evidence carrier relation, F.9 Bridge occurrence or description, or optional F.9
CLnote; the field supplies no evidence, relation truth, or use permission. - PC.ComparatorSetRef — a reference to the comparator family used by a declared partial order.
- PC.CharacteristicSpaceRef? — an optional reference to an admitted
U.CharacteristicSpacewhen the claim is interpreted in that space.
These are local field names, not members of a catalogue of public kinds. A formal profile may declare another local field only by naming the admitted value or existing reference it carries, the predicate or relation that gives it meaning, the consuming publication use, and any required pins.
Norms (E17-PC).
- E17-PC-1 (Exact grounding). A numeric or comparable field resolves the
U.Characteristic, the membership or interpretation predicate or CG-Spec reference that gives the value meaning, and material pins{unit, scale, reference-plane, edition}. - E17-PC-2 (Lexical discipline). Use Characteristic, slot, and CharacteristicSpace for those governed objects. Retain literal mathematical or domain terms such as axis, dimension or plane when they name the actual construction; a spatial metaphor must not hide the characteristic or relation being asserted.
- E17-PC-3 (No hidden arithmetic). A form does not hide aggregation or normalization. It cites the calculation or normalization definition and edition.
- E17-PC-4 (Crossing). For a semantic crossing, cite the F.9 Bridge and separate bounded-use claim. For a plane-dependent value, cite the selected
ReferencePlaneand applicable transfer or comparison rule. Add an F.9CLnote or B.3 penalty reference only when the receiving use consumes it. A field manufactures none of these relations or claims. - E17-PC-5 (Edition pinning). Fields that rely on maps, distances, spaces, set semantics or transfer rules pin the applicable editions. Revalidate an affected use when a relied-on meaning or condition may have changed; reissue when that result or an applicable publication requirement calls for it. Preserve supported historical and unaffected uses.
- E17-PC-6 (Viewpoint conditionality). The separate bounded-use declaration says why each field is present. Resolve
publicationViewpointRefonly when the selected episteme is claimed as aU.Viewor the formal operation depends on that viewpoint.PromoteFace[s->t]may reindex or annotate a form; it adds and widens no claim.
Publication-form responsibilities when this profile is selected. PlainView may show PC.Number only when the characteristic and material pins resolve; otherwise use qualitative wording. TechCard may add PC.ComparatorSetRef or PC.CharacteristicSpaceRef? only for a declared ordering or characteristic use. AssuranceLane may carry PC.EvidenceBinding only as a pointer to the evidence or policy relation. InteropCard remains notation-neutral and points to the references needed by the external consumer.
Extending the profile. Give a new local field a plain and technical label, the value or reference it carries, the predicate or relation that gives it meaning and establishes membership or applicability, the identity-relevant edition, pinning rule, and one named consuming use. If the work instead needs a new public ValueKind, return that as a separate kind-settlement and product decision; E.17 does not admit it by declaring a field.
Adding or changing invariants.
- Put a new invariant in the CG-Spec or other specification-use source that defines it; supply the test there.
- Version any affected
U.CharacteristicSpace, comparator, map, distance, or transfer rule; publish an explicit correspondence relation when semantics change and never mutate slots in place. - Update an
A.21gate check only when a gate consumes the invariant. Publication conformance can warn or block only according to the selected profile and bounded use. It does not create an operational gate. - State the edition-change and Lean-profile downgrade behavior that the concrete consuming use needs.
Author ergonomics (non-normative)
Quick author steps:
- Name source and reader/use. Point to the current source account and say what this reader must understand or do.
- Choose the minimum face set. Start with one face; add another only when a different reader/use needs different detail or form. Copy no claim that the source does not carry, and state material omissions.
- Try the reader's intended use. If the reader must explain how the practice works, follow the relation and its grounds through the face. If the reader must choose a method or arrangement, recover what the source gives them for making that choice. When several readers must act together, check that their faces concern compatible conditions and retain the dependency their work requires. If the reader needs to locate the source, a usable return to it can finish the task. If the reader must decide or act, follow that decision using the face and what the reader can access with the necessary permission and time. Carry a missing condition from the source onto the face, or provide a usable return to it. Stop when the declared use is served. Add viewpoint, scope, occurrence, pin, bridge, evidence, gate or assurance records only where their identities matter.
For the optional morphism profile, declare F_face, pin numeric or comparable content once, and establish the composition or promotion claim at its stated extent under §6. A -Lite face may omit optional fields and retains the strength and scope of the source claims.
Rules and Invariants (normative)
Local publication-composition test. For a claim about specified objects, arrows and forms, check the applicable local properties. For a law over a category, use the domain-wide conditions below.
identity:Emit_s(id_X)is the identity face morphism forFaceObj_s(X);composition witness: compare the face forg o fwith the composition of the faces for f and g under the selected equality. A verified match supports this pair. A failed comparison leaves the composition claim unavailable for this pair.no-new-claim diff: compare the meaning and conditions of the face with the selected source episteme, including any paraphrase, pin or conservative construction.monotone promotion: a richer face adds fields, pins, or typing without retracting or strengthening the source claim;scope non-widening:U.PublicationScopestays within the claim or work scope used by the selected description.
For the optional formal profile, declare the source objects and arrows in scope, the target face category for each form s, and the equality used there. Give the rule that sends each object X to FaceObj_s(X) and each source arrow f : X -> Y to an arrow Emit_s(f) : FaceObj_s(X) -> FaceObj_s(Y). Specify its action on the information retained by the face. If two source inputs have the same face representation but require different represented outputs, that information alone cannot determine the proposed action (C.29.1:4.4).
- Functoriality and typing per face.
- For each object X in the claimed domain,
Emit_s(id_X) = id_{FaceObj_s(X)}. - For every composable pair
X -f-> Y -g-> Zin that domain,Emit_s(g o f) = Emit_s(g) o Emit_s(f). Derive the laws from the action on arrows, check the identities and all pairs of a finite domain, or define the action on a generating presentation and show that it respects its defining relations and extends to the specified action. A use concerning one composite can end at its verified equality. - If
f : X -> Y, thenEmit_s(f) : FaceObj_s(X) -> FaceObj_s(Y)is total in the selected formal substrate. An ill-typed composite blocks that formal claim. Weakening conformance does not repair it.
- For each object X in the claimed domain,
- Face-promotion coherence.
- If
s <= t, the t-face is a more explicit publication form for the same selected source claims. - For a natural promotion give a component
PromoteFace[s->t]_X : FaceObj_s(X) -> FaceObj_t(X)for every object. For every arrowf : X -> Y, establishPromoteFace[s->t]_Y o Emit_s(f) = Emit_t(f) o PromoteFace[s->t]_X. For functors on the stated domain, checking the squares on generators suffices when they generate all arrows. For a claim about just one arrow, its square is sufficient. - Identity and composition of
PromoteFacefollow the selected formal substrate.AssuranceLaneis outside the default formality chain.
- If
- Source episteme and construction.
- Every
Emit_suse names the source episteme edition. It resolves apublicationViewpointRefonly when the selected episteme is claimed as aU.Viewor the formal operation's definition depends on that viewpoint. - When another episteme is constructed from the source, use A.6.3 to identify that source-to-receiving construction relation. The face constructor is not a species of
U.EpistemicViewing, and A.6.3 does not establishU.Viewmembership. - Changed claim content, EntityOfConcern, or effective reference scheme identifies another episteme under C.2.1. Changed form, carrier, or publication occurrence does not by itself.
- Every
- Pin discipline. Numeric or comparable claims used from a face retain the unit, scale, reference-plane, and edition pins required by the applicable characteristic and measurement patterns.
- Publication is not work. Build, rendering, upload, or delivery is
U.Workonly when each performer has its A.13 core and A.15.1 independently admits the dated occurrence. F.6 enters only when the publication account also consumes precise assignment-bound attribution through the same obtaining A.13 assignment; missing or failed F.6 alone leaves independently admitted Work intact. - Publication and carrier separation. E.24.PUB identifies the selected episteme edition, publication occurrence, expression, reusable form and carrier separately. A.10 supplies the evidence/provenance source-to-use path, G.6 supplies addressable path citation, slicing, and local refresh, and B.3 supplies any assurance claim.
- Cross-context and reference-plane use. For a semantic crossing, recover the F.17 endpoint senses, F.9 Bridge, and separate bounded-use claim. For a plane-dependent value, retain the characteristic, selected
ReferencePlane, and applicable transfer or comparison rule. Add A.10 or B.3 only when reliance is current; an optional F.9CLsummarizes evidence strength and never grants use. Visual juxtaposition and scheme difference alone establish none of these claims. - PublicationScope discipline. For a face use v selecting episteme E,
PublicationScope(v)does not exceed the claim scope on which that publication relies. A capability description may also cite a work scope, but the publication scope does not grant work admissibility.PromoteFacedoes not widen either scope.
The equations are conceptual-form constraints on the optional morphism-publication profile. They do not turn face symbols, formulas, or diagrams into world-side relations, viewpoints, views, publication occurrences, or work.
Objects used by the optional formal profile
In the optional morphism profile, the author selects source E and publication-form profile F_face. P is selected only for a material U.View claim or a formal operation whose definition depends on that viewpoint. A system performs any authoring, rendering, checking, or publication work. MVPK names the publication method and constraints.
Archetypal Grounding (SoTA-aligned Local Tests)
Use these examples to test the MVPK invariants. A source’s reputation does not establish that a face passes a test.
Ordinary two-reader publication. An accepted interface account already states the service boundary, messages, and failure conditions. A project lead needs a short explanation and an integrator needs the typed details. Publish one PlainView and one TechCard, both pointing to that same account and stating what they omit. Do not create InteropCard, AssuranceLane, a new viewpoint bundle, or a formal composition witness unless a later use needs them. The face labels establish neither U.View membership nor assurance.
Suppose the interface account permits a retry after an explicit rejection before processing and requires a status inquiry after a lost acknowledgement. A short card for finding the interface account can stop at its reference. A card for deciding whether to retry must make the two cases distinguishable before the next send. An integrator can follow the reference to the identified rule in time. An offline shift can use a checked local copy of that edition too. If the shift has no timely authorized route to the rule, put its operative terms on the card with the source edition, or limit the card to orientation. If the lead must explain why the actions differ, a list of the two instructions can leave the question open. The source may establish that the first request could already have taken effect when an acknowledgement is lost. Carry that ground into the lead’s account. If the source gives only the instructions, publish those and identify the missing ground for the requested explanation.
The remaining examples exercise optional formal or load-bearing branches.
-
Composite service pipeline (
InteropCard+AssuranceLane).f: Parse → Normalize,g: Normalize → Score.InteropCard(g∘f)is an interoperability face whose path claim matches the declared relational composition of the two source claims.AssuranceLane(g∘f)cites the A.10 evidence/provenance path and, only when replay needs a stable path address, its G.6PathIdorPathSliceId. The faces neither establish that composition nor become evidence carriers. -
Control loop morphism (
TechCard+PlainView).- For
h: Setpoint → Actuation,TechCard(h)is a typed card with units;PlainView(h)narrates the same mapping with no new claims. (Monotone formalization echoes refinement‑typed specification toolchains.)
- For
-
Optics-informed composition witness.
- Emit the face for
g∘f, compose the emitted faces forfandg, and compare them using the selected equality. This decides the question about the Parse–Normalize–Score pair. If these are the only two generating arrows in a free path category, extending their assigned images by path composition and sending empty paths to identities supplies a functor on it. For a published mapping defined independently, check that its action agrees with that extension. Additional generators or equations change the proof obligation.
- Emit the face for
-
Functional-description publication (
PlainView+TechCard). A principle scheme or functional diagram can publish a readable relation from signature or principle episteme content to method-family selection, selected method,U.WorkPlan, performedU.Work, work-result record, and result measurement. The MVPK faces can help inspect that relation and prepare a work plan, but they do not become work, gate passage, evidence, engineering justification, or control architecture. When one of those claims is current, recover its concreteA.15/A.15.1,A.10,B.3, orA.20/A.21grounds and any record required by the receiving use, or, for control architecture, the grounds for the architecture claim governed byC.30andA.22. UseC.30.LCAfor a separately current control-structure-view record andB.2.5for a separately current supervisor-subholon feedback record. Recover any earlier object or relation under §5.1a without inventing a prior record or authority. If the required grounds cannot be recovered, withhold the dependent claim and return the specific gap or a prospective repair, decision, or work-plan request.
Bias-Annotation
E.17 blocks publication-face bias, the mistake of treating the readability or layout of a face, card, view, rendering, source pointer, dashboard tile, or generated explanation as creating the underlying claim, evidence, work, gate, authority, or release relation. It also blocks source-proximity bias. A face can point near source material or a source relation. The operative source relation must still be recoverable by value.
Conformance Checklist (normative)
CC-MVPK-FD is the functional-description guard in §5.1a. It is conditional on a functional-description publication face and is not a prerequisite for every MVPK publication.
A conformance check is kept only if it changes the next bounded use of the publication face, blocks a concrete overclaim, or preserves a source reference or reopen condition needed for the declared bounded use.
Core ordinary checks
Conditional checks
Common Anti-Patterns and How to Avoid Them
- “Presentation logic” as semantics. Fix: Keep every claim in the source ClaimGraph. When a reader needs to know how a claim arose, name the authoring, measurement, observation, model, source-use, representation, or refinement relation. Use the specification-use gate, CG-Spec, or KD-CAL when it owns the requirement. Keep views declarative. Publication adds zero claims.
- Publishing only endpoint faces.
Fix: The optional formal profile constructs faces for
g o f, not only endpoint faces forFaceObj_s(X),FaceObj_s(Y), andFaceObj_s(Z). A system performs the construction work. - Unpinned numbers. Fix: Recover the value’s required interpretation and support. Retain the material pins and add CG or CHR references only when the applicable rule or selected profile consumes them. Withhold a numeric use whose required basis remains missing.
- Face presented as a view without conformance. Fix: Resolve the viewpoint episteme and apply E.17.0 to the candidate episteme; redesign or re-emit the face only after the semantic repair.
InteropCardequivalent toTechCardduplication. Fix:InteropCardcan refine typing or shape but cannot contradictTechCard(reindexing monotone).
Consequences
Rationale
Multi-view publication is needed because one account can serve several concerns without any face becoming the whole account. Source return, bounded use, and material omissions must be visible enough for ordinary reading. Viewpoint, correspondence, currentness, publication, evidence, assurance, decision, architecture, and release relations are added through their concrete defining or checking patterns only when the receiving use needs them.
SoTA-Echoing: Adopted And Adapted Invariants And Rejected Shortcuts
SoTA and local-rationale alignment rule. Read each external-source row as source idea -> local FPF invariant -> practical local test -> shortcut rejected. A cited source contributes only the idea translated into this pattern. A row deduced from named current FPF patterns is labelled local design rationale and is not presented as external SoTA evidence.
Relations
-
Architecture ADR projection boundary:
C.32.ADRis the architecture-specific publication projection forArchitectureDecisionDescription@Project. E.17 keeps publication face, source episteme, carrier, scope, and downstream typed value separate for the broader MVPK claim. In that name,@Projectis a compatibility and retrieval cue only. E.17 infers no project entity, composite-work identity, context, authority, viewpoint, or parthood from it.C.30.ADandC.32.ADRmust identify the compositeU.Workand the direct description-use or publication-use relation when project locality is current. -
Builds on:
C.2.1for selected-edition identity;E.24.PUBfor expression and bearing, selected form conformance, form-mediated claims and publication availability;E.17.0for viewpoint andU.Viewmembership;A.22for selected structure;A.6.3.RT/A.6.3.RT.OEfor expression construction andC.29for a mathematical lens;A.7andE.10.D2for carrier, front-end, EntityOfConcern, episteme, and specification-use discipline;A.6.2-A.6.3for optional source-to-candidate construction;E.8andE.10for authoring and publication-language discipline; and Part F and Part G for bridge, terminology, characteristic, and pin discipline. -
Constrains: publication-face-emitting automation and hand-written faces. When another episteme is constructed from a source, A.6.3 supplies the separate construction relation; E.17.0 separately tests viewpoint conformance, and E.24.PUB separately identifies publication occurrence/form/carrier. Readable form creates none of those relations, nor an evidence path, gate decision, work occurrence, assurance record, release source, or bridge declaration.
-
Neighboring-pattern boundary use: use the compact boundary aid in
E.17:5.1dwhen a publication-facing unit starts carrying work, reliance, evidence, assurance, gate, release, bridge, explanation, comparison, retargeting, carrier, or front-end claims beyond ordinary publication use. -
Part F bridge wording boundary: when the publication face uses or invites "same", "equivalent", "align", "map", substitutable, interchangeable, attribute, entity, or profile matching, or other Bridge-wording pressure across contexts, use Part F and
A.6.9to repair the wording. Use F.9 for the Bridge and bounded-use claim, and F.9.1 only for a separate optional stance note about that claim. Neither object follows from a publication face, and no local Bridge taxonomy is introduced here. -
Coordinates with:
C.2.Pfor source-expression and source-to-use recovery before publication-facing wording is relied on;A.15.4for appearance-based reliance repair; C-cluster selection or archive patterns when separately constructed epistemes are selected or retained; CHR and UNM for measurement and normalization semantics; F.9 for Bridge occurrences, bounded-use claims, optionalCL, evidence and loss boundaries, and optional Cards; F.9.1 for separate optional stance epistemes; andA.6.9for sameness wording. Publication faces remain publication forms; their bounded-use declarations, selected or receiving epistemes, occurrences, and carriers remain separate, and face status never establishesU.Viewmembership.
Minimal authoring template (Part E)
Ordinary publication
- Current source/account:
<recoverable source and edition or current subject> - Reader and use:
<the question or choice the reader must be able to handle, alone or with others> - Minimal publication-form set (MVPK faces):
<one or only the needed forms> - Bounded-use declaration for each form:
<reader, permitted use, and blocked stronger use> - Preserved and omitted:
<which source claims and connections support that use, and what remains open for the original question> - Return to source:
<where the reader checks or reopens the source> - Stop:
<why no additional face or apparatus changes this use>
Add only when triggered: publicationViewpointRef and E.17.0 conformance for a material U.View claim; U.PublicationScope; E.24.PUB occurrence/expression/form/carrier identities; pins; F.9 Bridge and bounded-use claim; selected ReferencePlane and applicable transfer or comparison rule; provenance, evidence, gate, release, or assurance references for the concrete receiving use.
Optional morphism profile: declare F_face and the source morphism; use Emit_s and PromoteFace witnesses only for faces that claim compositional publication.
Manager’s one‑page review (copy‑paste)
We publish only the publication forms current readers need, each tied to the same recoverable source and a separate bounded-use declaration, with material omissions visible and no added claims. A selected episteme exposed through a face is a
U.Viewonly through E.17.0 conformance; publication occurrence, form, carrier, work, evidence, gate, assurance, and release remain separate. Exact identities and formal witnesses appear only when the receiving use depends on them.
E.17:End
ExplanationFaithfulnessProfile — explanation-use discipline over existing MVPK faces
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative
One-line summary. ExplanationFaithfulnessProfile classifies the bounded explanation use of a publication expression or representation of one exact claim-bearing episteme. The profile neither identifies the episteme expressed by the text nor turns changed claims into another expression of the source.
Explanation-facing text in plain terms. Consider a text expression constructed under the rules of an existing MVPK face. It may express only the part of an exact source edition needed for a declared explanation use. Identify that edition by its ClaimGraph, exact EntityOfConcern and effective ReferenceScheme. Then test whether this expression recovers enough of that edition for the use under E.24.PUB.
Omitting other source content neither claims whole-source coverage nor by itself constitutes another episteme. A changed assertion, concern or effective scheme needs its own identity decision.
First identify the episteme and test the expression (ontic first screen). Recover the reader’s question from the request for an explanation. An editor may declare a narrower use that the expression can already serve. If so, establish what that use contributes to the requested explanation (C.39.QR:4.1). Identify the source and the claims and grounds the reader needs before assigning a class.
- Recover the exact source edition's ClaimGraph, EntityOfConcern and effective ReferenceScheme under C.2.1. Equal displayed words alone do not establish this identity.
- Test whether the text expresses enough of that edition for the declared use under E.24.PUB §4.2. Preserve every qualification that changes the answer for that use, keep the source return, and disclose use-relevant omissions. A partial expression may qualify. A missing required condition leaves source sufficiency unresolved or fails the test, even when each displayed sentence is individually true. Omission alone does not create a target episteme. Apply A.6.3.RT as well when a representation-change claim is current.
- If the text asserts changed content or concerns another entity or effective scheme, identify the exact target under C.2.1 and the source-to-target relation claimed under A.6.3.CR, A.6.3.CSC, A.6.4 or another applicable pattern. Establish that relation independently before EFP classifies the target expression. The class creates neither participant nor relation.
- A causal or counterfactual addition belongs to a separately identified target with the source relation and causal grounds required by its claim. Reuse an independently supported C.28 conclusion within its limits; use B.5.2 when a new explanatory hypothesis must be generated or compared. EFP may classify the resulting expression when that classification is still needed.
Explanation-use relation in plain terms. State which exact episteme the published text expresses, how that episteme relates to the named source when it is a different target, which explanation-use class applies, and what downstream claim or effect stays outside the profile. Name the exact E.24.PUB publication occurrence, pins, traces, or provenance only when they are material to the present use.
Use this when. Use EFP when a source-pinned, reconstructive, didactic, or speculative ambiguity changes how a published explanation expression may be reviewed or used—especially for generated, retrieval-facing, model-facing, derivative, or interactive explanation. Authorship alone does not trigger the profile.
What goes wrong if missed. A publication expression, a rewritten episteme, and a new hypothesis are all called a rendering of one source. Helpful wording then hides a changed claim-bearing object or an unsupported source relation.
What this buys. First identify the episteme, then classify the bounded explanation use. The reader can tell which episteme is being published, how a changed target was obtained, and which stronger use remains blocked.
Not this pattern when. For an ordinary note, if a source locator plus one natural-language bounded/blocked-use sentence already preserves meaning and prevents the credible overread, use that simpler publication note and stop. To establish rewrite, representation change, coarsening, comparison, retargeting, hypothesis production, evidence, work, assurance, or gate claims, first apply the pattern for each claim. EFP does not establish those claims.
First output. One compact explanation-use note naming the exact source or related target episteme, explanation class, source reference, bounded explanation-reader use, blocked downstream use, and reopen or boundary condition.
The note names a source-to-target relation when the text expresses a separately identified target, including a changed concern or effective scheme. MVPK face, pins, provenance, and other source fields are inherited by reference unless ambiguity or a load-bearing use makes them relevant.
Lower-burden ordinary branch. First try a source locator plus one sentence naming the allowed reader help and blocked stronger use. If that resolves an ordinary case, do not instantiate EFP. When class ambiguity still changes the next action, use the compact EFP result without a fuller field block.
Load-bearing use. Open the fuller explanation review only when the rendering will guide work or reliance, be relied on externally, be disputed, cross contexts, affect a person’s or team’s status, or be cited as evidence, approval, engineering justification, gate, or release reliance.
Stop condition. Stop before EFP when the simpler source-linked boundary sentence performs the task. After EFP is triggered, stop when the class, bounded/blocked use, and reopen condition settle the next action. Add no field or check that does not change it.
Bounded explanation-use examples.
Neighboring patterns and project records. E.17.ID.CR supplies the bounded-comparison discipline for a comparative review unit. A.6.3.CR and A.6.3.RT define same-entity rewrite and representation change. A.6.3.CSC defines the narrower-use result, blocked downstream use, and source-bearing reopen needed after deliberate coarsening. A.6.4 and OntologicalReframing address a changed EntityOfConcern.
A.15 governs System-Role–Method–Work alignment. A.15.1 independently admits dated Work. A.15.4 repairs an appearance-based work or reliance use only while its prerequisite remains unclear. B.3 supplies assurance and engineering-justification tests. A.20 tests a named internal constraint for a stated case under that pattern's subject and applicability conditions. A.21 governs a named gate decision under its applicable profile.
For permission-looking or policy-bearing prose, use A.2.8.PER for strong grants, exercises, weak non-prohibition/non-violation findings, and permission conflicts. Use A.2.8 for obligation, recommendation-as-duty, and prohibition commitments. Use A.2.9 for the communicative Work that institutes or revokes an effect.
Common wrong escalations and boundary transfers. Do not use this profile to hide new claims, bridge-comparison load, action-selection pressure, or gate-bearing guidance inside helpful prose. If the rendering is a bounded comparison, apply E.17.ID.CR. If it is only same-entity rewriting or representation shift, apply A.6.3.CR or A.6.3.RT. If a deliberately coarsened rendering's narrower bounded claim or effect, blocked downstream use, and source-bearing reopen are the problem, apply A.6.3.CSC. If it is already making world, work or reliance, assurance, or gate-bearing claims, leave E.17.EFP for the more exact downstream FPF pattern or project-side record.
Generated-explanation repaired case. Identify the exact source by its claims, concern and effective scheme, then test the generated text for the declared use. A sufficient partial expression may remain an expression of that source. Missing use-relevant content requires source return or repair, not automatic target creation. Changed assertions require an exact target and an independently obtaining source-to-target relation. An identity, sufficiency or relation gap leaves classification for that use open.
After identity and bounded sufficiency are settled, recover the source basis for each operative claim used beyond reader help. Use A.10 when its evidence/provenance account is needed for the named reliance. An assurance, gate, work, permission, approval, or release claim also needs its applicable pattern and exact project record when required. Missing grounds keep the dependent use at reader help or source-finding.
Common wrong first interpretation. A fluent, confident, source-linked, or reliable-looking explanation is treated as evidence. First identify the exact episteme expressed by the text and any required source-to-target relation, then classify its publication expression for reader help or source-finding. Only an operative claim with an A.10 evidence path or another source relation that carries, supports, or exposes the source basis for the operative claim can carry downstream reliance.
If a generated explanation says "reliable" but no operative claim maps to a source relation, the E.17.EFP result is limited to source-finding or reader help. If someone nevertheless proposes relying on it downstream, the receiving A.10, B.3, A.21, or other relation named by value can return evidence-needed or no-bounded-current-use for that attempted reliance. It is not weak evidence by style, confidence, fluency, or citation-like wording.
Generated-retelling survival. Compression or omission may preserve a source edition's inspectable reader-help use, source-finding cue and quoted pins when the partial expression remains sufficient under the ontic first screen. State the limited use and material omissions. If the text strengthens or otherwise changes an assertion, or changes the concern or effective scheme, identify the target and establish its claimed source-to-target relation before classifying the target expression. Fluency and links establish neither identity nor sufficiency, evidence, assurance, permission, gate passage or work authority.
Derivative text and adaptation source-link rule. A fork, adaptation, abridged guide, translation, generated explanation, tutorial or access-format conversion first undergoes the same identity-and-bounded-sufficiency test. A sufficient partial expression may remain a source expression. Insufficient expression returns the exact missing source content or condition. A changed claim-bearing target requires its own identity and an obtaining direct relation. EFP then qualifies explanation use only if needed.
If the result will guide work or reliance, map each operative claim to its source basis. Use A.10 when the evidence/provenance account must remain replayable for that reliance. A missing claim-to-source map permits only reader help, a source-gap note, or prospective evidence work.
Published-expression and episteme identity over revision and regeneration. Reidentify the source or target through ClaimGraph, exact EntityOfConcern and effective ReferenceScheme, then retest expression sufficiency for the declared use. A different selection of source content may remain another sufficient expression of the same edition. Equal words about another concern or under another effective scheme do not establish the same episteme. A changed target requires its direct source-to-target relation. Face, prompt, template, carrier or title alone settles none of these questions.
When use beyond ordinary reader help depends on how the text was produced, identify the exact generation or production relation and the source references it used. Neither relation changes episteme identity by itself. EFP records only the bounded explanation use of the resulting published expression.
Pattern basis. E.17 supplies face discipline; E.17.0 supplies viewpoint/view conformance only when U.View membership is material.
Builds on. E.17.0 MultiViewDescribing; E.17 MVPK; A.7; E.10.D2; A.6.B; F.9; F.18.
Coordinates with. ConservativeRetextualization; RepresentationTransition; E.17.ID.CR ComparativeReviewUnit; A.6.4; A.10; A.15; A.15.4; B.3; A.20; A.21; A.2.8; A.2.8.PER; A.2.9.
Problem frame
The exact source episteme may need more than one publication expression or representation. Explanation work may also produce changed claim content. Identify the resulting target episteme under C.2.1 rather than treating its ClaimGraph as the whole episteme or hiding changed content inside the word rendering. Recurrent cases include:
- a manager-readable expression of the same technical ClaimGraph;
- connective explanation that remains entailed by the source, or else belongs to an exactly related target episteme;
- didactic use of a sufficient source expression, including a partial expression, or of a separately identified target with an obtaining rewrite or coarsening relation;
- exploratory use of a publication expression of a separately constituted hypothesis episteme.
FPF uses C.2.1 for episteme identity, A.6.3 and neighboring patterns for source-to-target relations, E.17.0 for viewpoints and views, E.17 for reusable publication-face rules, and E.24.PUB for exact expression, form conformance, bearing and publication availability. EFP supplies only the remaining bounded explanation-use classification of one expression of the already identified source or target episteme.
Problem
Without a dedicated profile:
- an expression of the source, a rewritten target episteme, and a new hypothesis blur together;
- readers use explanation prose as if it were a second source of semantic rules;
- publication-side reviewers cannot tell which expressions remain within the bounded use of their explanation class;
- source and evidence details are either demanded for every explanation or omitted when a named claim, dispute, derivative, or reliance needs them;
- an EFP class quietly substitutes for C.2.1 identity, an obtaining source-to-target relation, bridge work, or a gate decision.
Forces
- Clarity vs semantic restraint. Explanation can help readers, but it does not mint new semantic commitments on publication faces.
- Face discipline vs reader fit. The same episteme can need different forms, while changed claims identify another episteme even when reader fit motivated the change.
- Traceability vs accessibility. Simpler renderings are useful only if readers can still recover how they relate to the source.
- Didactic usefulness vs policy misuse. A didactic or speculative retelling can help readers. Present it so they can distinguish it from assurance or gate-bearing content.
- Explanation vs interpretation. Some moves still belong to explanation rendering; other uses require interpretation, retargeting, or the FPF rule or project record that defines the world-side or gate claim.
Solution — review profile for explanation renderings on existing MVPK faces
Informal definition
ExplanationFaithfulnessProfileis a review profile for the explanation use of publication expressions or representations of exact claim-bearing epistemes on existing MVPK faces. E.17 supplies face discipline. E.17.0 supplies viewpoint/view conformance only whenU.Viewmembership is material.It does not create a new face family, episteme, or source relation. C.2.1 first identifies the exact episteme expressed by the text. E.24.PUB identifies the expression used here and any selected form conformance, while A.6.3.RT governs a claimed representation transition. When claim content changes, the applicable source-to-target pattern defines the relation and its obtaining test. EFP then states the bounded explanation use of that already identified expression.
Profile, episteme, and published-expression distinction
ExplanationFaithfulnessProfile is a review profile. Its cases concern passive publication expressions or representations of an exact U.Episteme.
Identify the episteme through its claim content, exact concern and effective scheme. Test the expression's sufficiency for the declared use. Then classify that explanation use. Partial expression can preserve the source identity. Missing needed content blocks that use, while a changed target needs a C.2.1 identity and an independently obtaining source-to-target relation. An EFP class supplies none of these prerequisites.
How to read this profile
This profile does not decide whether a claim is true or which claim-bearing object exists. It starts after C.2.1 identity and any required source-to-target relation are recoverable, then qualifies the explanation use of one publication expression or representation.
Faithfulnessnames the review question for that explanation use, not a pass verdict or an episteme-identity rule.- Class names are bounded-use labels for an expression or representation, not merit labels and not source-to-target relations.
- Use E.17 for reusable face rules and E.24.PUB for exact expression, form conformance, carrier bearing and publication availability.
- A changed assertion, concern or effective scheme reopens episteme identity even when the prose remains explanatory, didactic, reconstructive or speculative. Selecting only enough source content for a bounded use does not itself assert a changed ClaimGraph.
- A causal or counterfactual addition requires a separately identified target and grounds for that claim. C.28 governs causal support; B.5.2 supplies a new explanatory hypothesis when that is the work needed.
Local working vocabulary
The following terms apply to this review.
- Source episteme and publication occurrence = the exact source
U.Epistemeedition and, when material, the exact E.24.PUBEpistemePublicationRelationoccurrence through which it is available. Neither is an expression, MVPK face, reusable form, carrier, or arbitrary physical item. - Current claim-bearing episteme = the exact source or target edition identified by ClaimGraph, EntityOfConcern and effective ReferenceScheme. The published expression may express a sufficient part of that edition for the declared use. A target's claimed source-to-target relation must obtain under its direct pattern.
- Published explanation expression = one independently identified sign construction that expresses enough of the current claim-bearing episteme for the bounded explanation use. Identify its expression domain and equality rule under E.24.PUB when comparing what was supplied. It follows the reusable rules of the selected MVPK face. Those rules are not the filled text and do not acquire its explanation-use class.
- Class assignment = the explanation-use class assigned to that published expression on that face.
- Bundle-local class difference = a case where two expressions in one bundle carry different bounded explanation uses.
These are review aids, not new kinds or relation types. EFP neither creates the current episteme nor substitutes for C.2.1, E.24.PUB, A.6.3, B.5.2, or another direct source-to-target pattern.
Core profile fields
Perform the ontic first screen once without copying its result into a metadata record for every note. Most published expressions whose identity branch is already recoverable need only the compact explanation-use note:
The fuller field vocabulary below opens only when ambiguity or load-bearing use is present: different explanation classes for expressions constructed under the selected faces, source linkage dispute, connective reconstruction, reader-fit dispute, interaction or statefulness, derivative rendering, cross-context reuse, cited reliance, work or reliance, evidence, gate, engineering justification, bridge, or coarsening boundary.
faceRuleRef = E.17andviewpointConformanceRuleRef = E.17.0;sourcePublicationOrRecordForm;targetPublicationOrRecordForm;changeTargetRef;entityOfConcernPolicy = preservefor explanation renderings over the same underlying sourceU.Epistemeedition;boundedContextPolicy;viewpointPolicy;referenceSchemePolicy;representationLanguagePolicyidentifies what the rendering must preserve or may change about the language used for its expression. An episteme-specific convention for interpreting a value belongs withreferenceSchemePolicywhen it affects the claim.groundingPolicy;referencePlanePolicy;claimPolicy;claimScopePolicy;publicationScopePolicy;reliabilityTransportPolicy;pinningPolicy;provenancePolicy;lossProfile;claimContinuityClass;microtheoryContinuityClass;onticContinuityClass;bridgeRequirement;worldContactPolicy;evidencePolicy;gatePolicy;workCrossing;sourceRelationRuleRef?,upstreamAuthoritySourceRef?,downstreamUseRuleRef?, anddownstreamAuthoritySourceRef?;boundedFaces;publication-face kind valuewhenpublication face/formorinterop publication formdiscipline is present;publicNamePolicy;explanationSourceRelationClassusing the sharedE.17:5.1bvocabulary when source pointer, source availability or retrieval, source use, source faithfulness, claim-source relation, contradiction, omission, claim widening, added linkage, independent verification, bounded use, forbidden downstream use, or reopen trigger could diverge;- no generic source-relation field; source relation is recorded through
explanationSourceRelationClass; augmentationRelation;addedLinkPolicywhen a non-obviousSourceLinkedExplanationReconstructionconnective points to a derivation from the source claims or to an exact relation occurrence that those source claims already report and whose obtaining is independently established;targetUserModel?when reader-fit materially shapes the rendering;interactionMode?when the explanation is more than one static explanatory paragraph;contrastiveQuestion?when the rendering is answering a user-facing contrast or why-question;boundedReaderUse?when downstream use is bounded by intended reader and task;overreadRisk?when overinterpretation pressure is part of the review load;evidenceRelation?only when a named operative claim or receiving reliance consumes an A.10 evidence/provenance path;noNewBoundaryClaims = trueon explanation faces;compositionRule;reopenCondition.
These fields inherit the E.17:5.1e local-field rule. They classify one explanation-facing rendering for review. They do not create U.Kind, publication-face kind, RelationKind, KindBridge, EvidenceKind, GateDecision, SpeechAct, Commitment, U.Work, authority reference, publication face, or project-side FPF kind and reference named by value unless another FPF pattern explicitly defines or instantiates that object.
The explanationClass value is a local source-relation and bounded-use profile value, not ExplanationKind, not U.Kind, not EvidenceKind, not FaceKind, and not a truth certificate.
When claim content changes, pause EFP until the practitioner uses C.2.1 to identify the target episteme and the applicable source-to-target pattern to identify and test the relation. EFP may then qualify a publication expression of that target only when explanation use remains a distinct question. It never substitutes for that relation or its obtaining test.
Working-model first
Ordinary published expressions do not restate every field or replay the ontic decision. When their exact claim-bearing episteme, MVPK face, any material E.24.PUB occurrence, and already published source references make the branch recoverable, the compact note inherits those conditions by reference.
A source-bearing review record becomes necessary when:
- explanation class differs between expressions constructed under different faces in the same publication bundle;
- the rendering relies on bounded connective prose that is not obvious from the source wording alone;
- didactic or speculative wording creates a credible risk of policy, assurance, or gate misuse;
- source linkage, provenance, or reliability transport would otherwise become unclear;
- the rendering is a fork, adaptation, translation, generated explanation, tutorial, access-format conversion, or another derivative publication that can be mistaken for the source publication, source relation, or source episteme itself.
When one rendering needs its own narrower bounded claim or effect line, blocked downstream claim or effect line, or source-bearing reopen rule because distinctions were deliberately coarsened for reader fit, the issue is no longer only explanation class. Apply A.6.3.CSC Controlled Semantic Coarsening.
What a publication-side reviewer checks first
A publication-side reviewer starts with five questions:
- Which source edition is intended: what ClaimGraph, EntityOfConcern and effective ReferenceScheme identify it?
- What explanation did the reader ask for? Can the reader recover from this text and its accessible source the relation that explains the result over the requested range? If the request includes choosing how to proceed, can the reader also recover how to make that choice? A few claims can supply the full explanation. Retain the qualifications, return to source and omission disclosures that affect this use. An omitted answer-changing condition blocks this use even if the displayed sentences are true.
- Does the text assert changed content or concern another entity or effective scheme? If so, which exact target and independently obtaining source-to-target relation support its expression?
- Which E.24.PUB expression or applicable A.6.3.RT representation expresses the identified episteme, and which explanation class changes the reader's next action?
- Has the expression begun carrying another unsupported claim, relation, reliance or deliberately coarsened use that must return to its direct pattern?
If a prerequisite is missing, return the exact source-identity, source-sufficiency, target-identity or relation gap. Do not manufacture a target from an omission or repair a gap with a class.
Once these prerequisites and the action-changing class distinction are recoverable, the compact note is complete. Open a fuller face-by-face record only when a section 4.2 trigger consumes additional fields.
Interpretant-side block
This profile classifies the explanation use of expressions constructed under existing faces. It does not describe full interactive explanation systems.
When reader fit materially changes the explanation class, bounded use, blocked use, or reopen condition, make only the distinction needed for that change. A familiar audience and static note may need no separate reader-model field. A contrastive or interactive case may need one or more of targetUserModel, interactionMode, contrastiveQuestion, boundedReaderUse, or overreadRisk.
These names are optional prompts, not a five-field publication block. They serve only to show the reader-fit difference that changes the present use.
When the bounded use depends on how much selected structure the reader can recover, use C.2.8 with the relevant preparation, access and budget. Compare recovered structure separately from explanation faithfulness. Keep the exact episteme identified by the first screen: a content change requires its own target, even if the revised explanation is easier to use. The ordinary source-linked note remains sufficient whenever it meets EFP's non-use condition.
Explanation class set
The profile uses four explanation classes:
SourcePinnedExplanationSourceLinkedExplanationReconstructionDidacticRetellingSpeculativeRetelling
In field form, the local assignment is explanationClass = SourcePinnedExplanation | SourceLinkedExplanationReconstruction | DidacticRetelling | SpeculativeRetelling.
Class assignment follows, and never replaces, the ontic first screen.
SourcePinnedExplanationqualifies an expression or representation that expresses enough of the exact source edition for the declared use, with action-changing qualifications, source return and use-relevant omissions recoverable. It claims no whole-source coverage.SourceLinkedExplanationReconstructionqualifies a non-obvious connective only when it remains in the same source ClaimGraph. Either a stated derivation from the exact source claims recovers it, or the source ClaimGraph already reports an exact relation occurrence shown independently to obtain under its defining pattern. An independently true relation that the source does not claim belongs to another target ClaimGraph.DidacticRetellingqualifies teaching or onboarding use of a sufficient source expression, including a partial expression, or of an exact target connected underA.6.3.CR,A.6.3.CSCor another applicable pattern when pedagogy changes claims, concern or effective scheme.SpeculativeRetellingqualifies only the bounded exploratory use of an expression of a separately constituted hypothesis episteme, normally produced underB.5.2. It is not a speculative expression of the original source ClaimGraph.
These values are not U.Kind values, MVPK faces, semantic merit grades, source-to-target relations, or episteme identities. They state how the published expression may be used after those objects and relations have been recovered.
Assign the class to each published expression on its face. A multi-face bundle can therefore contain expressions of different classes. If a PlainView expression stays source-pinned while a TechCard expression expresses a separately related target episteme, the bundle names both exact epistemes and the class difference.
Ordinary class-selection guidance
A practical order is:
- identify the source's ClaimGraph, exact concern and effective scheme;
- test sufficiency of the text for the declared use, including qualifications and omissions;
- identify any changed target and independently obtaining source-to-target relation;
- identify the expression or representation and assign an EFP class only if that explanation-use distinction changes the reader's next action.
Use SourcePinnedExplanation for a sufficient source expression, SourceLinkedExplanationReconstruction for an already justified connective explanation, DidacticRetelling for bounded teaching use of the identified source or target, and SpeculativeRetelling only for a separately constituted hypothesis episteme. A missing prerequisite returns its exact gap. Any narrower reader-help use must itself satisfy the source-expression test.
Do not keep one narrower-use target with declared source-loss mode inside explanation merely because the prose is reader-friendly. When its narrower bounded claim or effect, blocked downstream use, and source-bearing return are primary, use A.6.3.CSC Controlled Semantic Coarsening. EFP may qualify a later publication expression only if explanation use remains a separate live question.
Entailed connective and addedLinkPolicy
Harmless connective wording adds no proposition: conjunction markers, pronoun recovery, and sentence order can make an already explicit source statement readable. No addedLinkPolicy is needed for that case.
SourceLinkedExplanationReconstruction applies to a less obvious connective only when one of two bases is recoverable:
- the exact source claims plus their effective reference scheme make the connective a consequence under a stated derivation; or
- the exact source claims already report the relation occurrence, and that occurrence independently obtains under its defining pattern.
When that basis is material but not visible in the prose, a compact addedLinkPolicy points to it:
addedLinkKind— the connective being exposed;sourceReferenceSet— the exact source claims used;effectiveSchemeOrRuleRefs— the designation, interpretation, ordering, or inference rules used by the derivation;derivationOrRelationRef— the inspectable derivation or the exact relation occurrence already reported by the source claims and independently shown to obtain;claimContentResult = source-recoverable— confirmation that the connective introduces no unsupported target claim;reopenTrigger— a source, scheme, rule, context, or relation change that invalidates the basis.
The policy is an index to the basis, not evidence that the basis exists. boundednessReason, a forbidden-link note, or author intent may help delimit use, but none substitutes for derivationOrRelationRef.
If neither a derivation from the exact source claims nor an exact source-reported relation occurrence that independently obtains can be recovered, the connective is another claim.
Constitute its exact target episteme under C.2.1 and apply the direct relation, bridge, comparison, or B.5.2 hypothesis pattern that fits the new claim. If that result is unavailable, remove the connective or leave EFP. A downgrade label cannot make it source-linked.
Working bounded-use matrix
This matrix assigns no evidence relation. An ordinary EFP result needs no A.10 path. Exact evidence, trace, pin, or provenance details open only when a named claim, dispute, derivative transformation, or receiving reliance consumes them and its applicable pattern or project record requires them.
ExplanationFaithfulnessProfile ordinarily stays on publication face/form. Any appearance on interop publication form remains source-pinned and structure-preserving, and cannot introduce explanation-specific semantics into interop publication. Didactic or speculative restrictions are use-profile restrictions over existing faces, not new face kinds.
Source-pinned explanation on AssuranceLane-facing publication is exceptional. Unless the exact face or source policy permits that use with visible evidence carriers, source pins, and no added semantics, reviewers treat AssuranceLane-facing explanation rendering as blocked.
DidacticRetelling may carry analogy, scaffolding, or reader orientation without asserting a domain fact. Every domain claim it does express belongs either to the exact source ClaimGraph or to an identified target episteme with an obtaining source-to-target relation. Marking prose non-canonical or trace-free does not erase claim content, create its episteme, or establish that relation.
When such analogy or scaffolding sits beside technical content, box or otherwise visibly separate it so readers do not merge it into the technical source. That cue limits likely use but does not establish episteme identity or a source relation.
The compact ordinary result needs only a source locator sufficient to reopen the exact source or target decision. Publish exact claim IDs, pins, trace paths, provenance details, or an A.10 evidence relation only when a named claim, dispute, derivative transformation, or receiving reliance consumes them. A reopenable locator does not by itself establish an evidence path.
When a reader-fit difference changes the bounded or blocked use, state only the relevant audience, interaction, question, use, or overread distinction. Do not publish or inherit all five reader-model fields for ordinary reader help.
Shared explanation rule set
Preservation rule
Every published explanation expression under this profile expresses enough of one exact episteme edition for the declared use. Source identity rests on ClaimGraph, exact EntityOfConcern and effective ReferenceScheme. Sufficiency rests on retained action-changing qualifications, source return and disclosure of use-relevant omissions. A partial expression may satisfy both.
Insufficiency does not constitute a target. Changed content, concern or effective scheme requires its own identity decision and any claimed source-to-target relation. E.24.PUB publication occurrence remains separate, and the EFP class changes neither identity nor relation.
Loss and reliability rule
A published expression states material omission, reordering, simplification or connection. An omission that removes information required by the declared use blocks that use even when the remaining text is true.
When the expression instead asserts changed content, identify its exact target and establish the claimed source-to-target relation under A.6.3 or another applicable pattern. An EFP label supplies neither missing content nor that relation. Reliability is never silently widened by more persuasive prose.
When a concrete reader-fit difference is load-bearing, expose only enough of its bounded use or overread risk to prevent the didactic or contrastive expression from being mistaken for assurance, policy, or gate guidance.
Downstream-use and boundary rule
This profile stays explanation-facing and episteme-facing. It does not decide bridge stance, retargeting, action selection, executable docking, gate-bearing claims or effects, assurance, engineering justification, or work enactment.
If a case starts carrying one bounded comparative review case, rival interpretations, bridge-mediated comparison load, world consequences, work or reliance consequences, gate consequences, assurance, or engineering justification, apply the neighboring FPF pattern, then name the project-side object or record that carries the claim or effect and its FPF kind. Relevant patterns include E.17.ID.CR, F.9 for an obtaining Bridge and its bounded-use claim, F.9.1 for an optional stance note, B.5.2, A.6.4, A.15, A.15.1, A.15.4 only while an appearance hides the needed work or reliance prerequisite, B.3, A.20, and A.21.
Interpretant-side fields do not weaken that boundary rule. They serve only to bound reader use. They cannot authorize unsupported downstream guidance.
If a coarsened explanation-like rendering needs a narrower bounded claim or effect, a blocked downstream use, and a source-bearing reopen condition, apply A.6.3.CSC Controlled Semantic Coarsening rather than keeping the case in ordinary explanation-use discipline.
Composition and reopen rule
Repeated SourcePinnedExplanation over expressions of the same exact source edition can be idempotent for the same bounded use. A changed selection of content reopens sufficiency for that use. Changed assertions, concern or effective scheme reopen C.2.1 identity and any source-to-target relation.
Didactic target expressions reopen when their target edition, relation or use changes. Speculative expressions reopen when their B.5.2 hypothesis edition, prompt relation or exploratory use changes.
Hard boundary rules
A rendering reviewed under this profile keeps the following explicit:
- it does not create a second face family;
- it does not turn faces into a second semantic rule track;
- it does not license new A.6.B boundary claims on explanation faces: law claims, use-boundary claims, deontic or commitment claims, and effect or evidence claims;
- it does not replace bridge discipline, retargeting discipline, or world or gate boundary discipline;
- it does not let
publication face/formandinterop publication formcollapse into one undifferentiated explanation channel.
If explanation text carries a changed ClaimGraph, stop class review, identify the exact target episteme and establish that the direct source-to-target relation obtains. Resume EFP only for a publication expression of that target when bounded explanation use remains separately material.
Archetypal grounding
Source-pinned explanation across multiple faces
Source claim slice. Claim D-14: Cooling loop CL-2 maintains the required temperature margin during standard load. Evidence pins: T-44, E-17.
PlainView rendering. Cooling loop CL-2 keeps the required temperature margin during standard load. Source pins: T-44, E-17.
TechCard rendering. D-14 stays source-pinned to T-44 and E-17; this rendering only shortens and reorders the claim.
This stays within SourcePinnedExplanation because the rendering changes readability, not the semantic load.
Entailed connective
Source claims under thermal scheme RS_plantThermal.
D-14: During standard load, CL-2 outlet temperature is at most 65 °C.D-18: During standard load, inspection criterion IC-7 is satisfied when that same outlet temperature is at most 70 °C.
Published reconstruction. During standard load, CL-2 outlet temperature satisfies the IC-7 upper-bound criterion stated by D-18.
The connective is recoverable because both claims concern the same outlet and load context, RS_plantThermal supplies the Celsius order, and 65 <= 70. The compact addedLinkPolicy points to {D-14,D-18}, RS_plantThermal.order, and that one-step derivation, so a reader can check the implication. This expression may be SourceLinkedExplanationReconstruction while those premises and rules remain current.
Non-entailed link exits the profile
Source claim. D-21: The reserve path remained available during observed overload interval O-7.
Proposed connective. Therefore the reserve-path design is robust against every short overload.
No source premise, effective-scheme rule, or already obtaining robustness relation derives the universal design claim. addedLinkPolicy cannot repair that absence.
To retain the sentence, constitute exact target episteme E_robustnessClaim and apply the direct robustness, comparison, bridge, or B.5.2 hypothesis pattern appropriate to the intended claim. Until that relation obtains, remove the sentence or leave EFP. It is not source-linked reconstruction.
Selected-method explanation with an explicit source relation
Source slice. The method-selection note chooses method M-2 because the material stays below threshold T and resource window W is available. It also says that work plan WP-17 and result measurement RM-4 remain required before and after execution.
Published explanation. M-2 is selected here because the material stays below threshold T and resource window W is available. Work plan WP-17 and result measurement RM-4 remain required before and after execution.
The selection relation and both limits are explicit in the source, so this is ordinary same-ClaimGraph re-expression. It needs no addedLinkPolicy. It is not evidence that work occurred, a gate decision, or engineering justification.
These facts can explain this choice of M-2. Suppose the reader must choose a method for a later job in which the window may change. The account of this selection alone leaves open what to do then. If the source gives a selection rule and the grounds for using it across those conditions, carry them into the explanation, even if the source develops them inside this one case. Otherwise return the supported explanation of this choice and identify what remains to learn about choosing a method.
Selection use concerns exact U.Method M-2. Planning concerns U.WorkPlan WP-17 under A.15.2. Any claim that work occurred requires a dated U.Work under A.15.1. Evidence, engineering-justification, or gate use remains under A.10, B.3, A.20, or A.21 only for a claim of that kind.
Partial source expression and a mixed-face bundle
Source edition. In this example, ReservePathDescription-E4 concerns path R-1 and its overload intervals under effective duration scheme RS_duration. It contains D-31: R-1 is configured to remain available for overload intervals no longer than five minutes and T-8: Observed interval O-7 lasted two minutes.
PlainView expression. R-1 is configured for overload intervals up to five minutes. Source: ReservePathDescription-E4, D-31. Use: orientation to the configured duration; this expression omits the O-7 observation.
For that use, D-31 is sufficient and the expression may be SourcePinnedExplanation of the same source edition. It does not assert T-8, observed availability or a guarantee. The omitted T-8 does not constitute another episteme, and the class claims no whole-source coverage.
TechCard expression. O-7 falls within D-31's configured duration window. Sources: ReservePathDescription-E4, D-31 and T-8.
That connective follows from 2 min <= 5 min under RS_duration. The expression may be SourceLinkedExplanationReconstruction with this derivation pointer. The bundle states the class difference.
The PlainView expression alone cannot answer whether O-7 fits: that use needs the omitted T-8 and must return to it. Neither comparison establishes that the path remained available.
Missing qualification. Suppose an identified source edition also states D-32: This availability configuration applies only while the backup supply is energized. An expression giving only the true five-minute duration is insufficient for deciding whether the configuration applies during loss of that supply. Return D-32 and the supply-state facts before that use.
This insufficiency creates no target episteme. A separate assertion of unconditional availability would be changed content requiring its own target and relation.
Equal words, different subject. The same duration sentence about path R-2 does not identify the R-1 source. Likewise, identical displayed duration words interpreted under another effective scheme do not establish the same source identity. Recover the exact concern and scheme and apply the relevant retargeting or scheme-change pattern before claiming a source-to-target relation.
Didactic retelling
Source episteme claim. The pressure-control condition is satisfied whenever the reserve valve opens within 80 ms.
Didactic publication expression. For onboarding: opening the reserve valve within 80 ms is enough to satisfy the pressure-control condition. The exact condition and threshold remain in the pinned source edition.
This expression conveys the same source ClaimGraph. DidacticRetelling qualifies only its teaching use. If the text instead says that the whole system is safe, that different safety claim requires its own target episteme, an obtaining source-to-target relation, and the applicable safety relation before publication. A didactic label cannot supply them.
Speculative retelling
Observed-source episteme. The pinned source notes record the observed recovery, but they do not explain why the recovery was so rapid.
That observation may frame an abductive prompt. If B.5.2 produces hypothesis episteme E_couplingHypothesis with claim A temporary coupling effect may have accelerated recovery, that claim belongs to the new hypothesis ClaimGraph, not to the observed-source edition.
Speculative publication expression of the hypothesis episteme. Exploratory hypothesis: a temporary coupling effect may have accelerated recovery. This is the separately identified hypothesis, not a claim of the incident source.
SpeculativeRetelling qualifies only this expression's exploratory explanation use. It neither constitutes E_couplingHypothesis nor turns the expression into a passive rendering of the observed source.
Anti-example: explanation that quietly becomes a new claim
Source episteme claim. The reserve path remained available during the observed short overload interval.
Overreaching text. The reserve-path design is robust against short overloads.
The second sentence has a different ClaimGraph. To retain it, constitute an exact target episteme under C.2.1, identify an obtaining source-to-target relation, and establish the wider design-robustness claim under its applicable pattern. Until that relation obtains and the wider claim is established, the sentence is unsupported and receives no EFP class. Reopening the source or calling the text face-local does not make the claim part of the source edition.
Anti-example: reader help that quietly becomes policy-bearing use
Source slice. The onboarding note explains, in simplified prose, that the reserve valve usually opens quickly enough to keep the local pressure condition inside the tolerated window.
Overreaching rendering on an AssuranceLane-facing use. This explanation is sufficient assurance that short overloads stay inside the tolerated window.
This assurance sentence has a different ClaimGraph. It requires an exact target episteme under C.2.1 and the applicable A.10/B.3 relations. Until those obtain it is unsupported and receives no EFP class. The earlier onboarding expression may retain its bounded didactic use, but that class neither carries nor weakens the assurance claim.
Boundary to lighter explanatory note with source-bearing return
Source slice. The technical incident note says the reserve path remained available during the measured load band, but it also keeps one unresolved ambiguity about recovery latency.
Lighter explanatory rendering. In plain terms: the reserve path stayed available during overload recovery.
For a use that needs the measured load band or the unresolved latency qualification, this lighter text is insufficient. Return those source conditions. Omission alone does not constitute another episteme. If the text is instead intended to assert availability throughout overload recovery beyond the measured band, it makes a changed claim. Identify that target under C.2.1 and establish its A.6.3.CSC or other applicable source-to-target relation before EFP classifies a later target expression.
Class-specific reopen cues in the worked slices
SourcePinnedExplanationreopens when the pinned source claim set, source pins, or face-use assumptions change so that the rendering can no longer remain claim-preserving and visibly source-bound.SourceLinkedExplanationReconstructionreopens when any source premise, effective-scheme rule, derivation, context identity, source claim about the exact relation occurrence, or that occurrence's obtaining basis changes or disappears.DidacticRetellingreopens when the exact source or target edition connected under A.6.3 changes, or when teaching use starts functioning as policy-bearing, design-bearing, or gate-bearing guidance.SpeculativeRetellingreopens when its exact B.5.2 hypothesis edition, prompt link, or exploratory use changes. It never falls back to being a passive expression of the observation source.
Boundary to interpretation and world or gate use
If a text carries a new hypothesis or another changed claim, first constitute its exact target episteme and apply B.5.2, A.6.3, or the other direct source-to-target pattern. Comparative review, rival interpretation, bridge, world, gate, assurance, and engineering-justification uses likewise require their own patterns. EFP can only qualify a later published expression's explanation use.
Human-authored and generated task replay against the simpler alternative
These editorial comparisons ask what the displayed text preserves and what its reader would need for the intended use. The author can decide which form to publish by examining the two texts. Observed comprehension, effort and error rates require evidence from the corresponding use.
The first case establishes the simpler-note boundary through unchanged claims, source recovery and a sufficient use limit. The same boundary applies when a generator produces that adequate note. In the second case, the added robustness claim calls for the identity and support check; a human author could make the same addition. The external literature supplies grounds for the generated, model-facing and interactive populations it studies.
Bias-Annotation
Scope: Conditional where explanation-class ambiguity changes use. External source grounding is limited to generated, model-facing, retrieval-facing, or interactive explanation. The simpler-note non-use condition applies whenever that note performs the task. Ordinary human-authored use remains a local design branch.
Conformance Checklist
These checks apply only after EFP's use condition survives the simpler-note comparison. Retain a check only if it changes the next bounded use, blocks a concrete overclaim, or preserves the source or reopen condition needed for that action.
Use core ordinary checks first. Conditional rows open only when reader-fit, bundle-local class difference, bounded explanation class, connective reconstruction, derivative rendering, or downstream reliance use is present.
EFP-Core ordinary checks
- CC-EF-0 — Exact episteme and sufficient expression are recoverable. Identify the source through ClaimGraph, exact EntityOfConcern and effective ReferenceScheme. Compare the declared use with the reader’s original question. For a promised explanation, the expression and its accessible source supply the relation and grounds the reader needs, or identify the part the expression supports and the question still open. Preserve answer-changing qualifications, source return and use-relevant omissions. Whole-source coverage is not required. Insufficiency returns the exact gap, not an invented target. A changed target needs its own identity and an independently obtaining source-to-target relation. A speculative causal or counterfactual claim belongs to a separate B.5.2 hypothesis episteme.
- CC-EF-1 — Explanation class follows identity. The class is named for the publication expression after CC-EF-0. It is not used as episteme identity or source-to-target evidence.
- CC-EF-3 — Source reference and blocked downstream use are explicit. The compact note states source reference, bounded explanation-reader use, blocked downstream use, and reopen or boundary condition.
- CC-EF-5 — No new A.6.B boundary claims on explanation faces. The no-new-boundary-claims rule is explicit on explanation faces. The blocked claims are law claims tested under A.6.B, use-boundary claims, deontic or commitment claims, and effect or evidence claims.
- CC-EF-7 — No second face family. A publication-side reviewer can tell why the case remains explanation-facing rather than becoming a second semantic rule track.
EFP-Conditional checks
- CC-EF-4 — Interpretant-side block is explicit when reader fit changes the classification or use. Only the reader-fit distinctions that change the current class, bounded use, blocked use, or reopen condition are stated. The five optional prompts are not a required block.
- CC-EF-2 — Face and
publication-face kindboundary is explicit when present. State face, pinning, provenance, or reliability details only when the present expression choice, dispute, derivative, or receiving use makes that boundary material and it is not already recoverable by source reference. - CC-EF-6 — Boundary to interpretation, retargeting, coarsening, and world or gate use is explicit.
The boundary is explicit, including
A.6.3.CSC Controlled Semantic Coarseningwhen a narrower bounded claim or effect, blocked downstream claim or effect, or source-bearing reopen condition becomes primary. - CC-EF-8 — Bundle-local class differences are explicit. When one publication bundle carries expressions with different explanation classes under its selected faces, state the difference for those expressions. One bundle-wide label would hide it.
- CC-EF-9 — Source-loss or changed-claim cases retain exact identity and use boundaries. A didactic target names its exact A.6.3 or other relation. A speculative expression names its exact B.5.2 hypothesis episteme. Any material source loss or reliability downgrade states its bounded and forbidden uses without pretending that the EFP class supplies identity or relation evidence.
- CC-EF-10 — Reopen triggers match the class. The published review note makes class-relevant reopen triggers visible when source claim set, pins, provenance, or face-use assumptions change.
- CC-EF-11 — Every non-obvious source-linked connective has a basis in the source.
The exact source claims and effective scheme yield a stated derivation, or those source claims already report an exact relation occurrence whose obtaining is independently established.
addedLinkPolicypoints to that basis. Without the basis, the added claim becomes an exact target episteme under its direct pattern or exits EFP. - CC-EF-12 — Derivative renderings keep source links operative.
A fork, adaptation, translation, generated explanation, tutorial, access-format conversion, or other derivative rendering that will guide work or reliance maps each operative claim to the exact source passage, carrier path, or project record that evidences it and names that record's FPF kind when material. If these requirements are not met, restrict the rendering to reader help or apply
A.6.3.CSCas appropriate. - CC-EF-13 — Generated explanation reliance boundary is explicit. A generated explanation used beyond ordinary reader help states its explanation class, source-finding state, operative claims, the FPF pattern used to test each relied-on claim, the source basis and any exact project record required by that use, and the supported use or specific gap. Evidence, assurance, approval, gate passage, release reliance and Work authority retain their direct grounds.
Common Anti-Patterns and How to Avoid Them
Consequences
- Explanation classes become explicit and reviewable.
- Existing MVPK face discipline stays intact.
- The ordinary result stays compact. Exact pins, provenance, trace, reader-model, and evidence details appear only for the concrete use that consumes them.
- The boundary to interpretation, retargeting, and world or gate work becomes easier to review.
Rationale
EFP keeps the identity and bounded-use check when it changes the permitted use of an explanation. Prefer a simpler source-linked boundary sentence whenever it preserves the needed claims and prevents the same credible overread.
SoTA Alignment and Source-Scope Boundary
Source-use rule. A source supports only claims within the problem population and action it studies. The external sources below concern AI explanations, NLP/model interpretations, LLM-generated explanations, RAG outputs, or interactive XAI systems. They do not establish a universal architecture for ordinary human-authored engineering notes.
Source-grounded branch. The XAI/NLP/RAG sources justify caution about generated or model-facing explanations: fluency, plausibility, retrieved context, or an AI summary label does not establish claim preservation, evidence, or reliance. They support the focused identity and use check only when that population is current.
Local human-authored branch. Use the source-linked note when its contents and use limit complete the intended job. E.17.EFP:5.7 gives the editorial grounds for that choice. To claim that readers understood it more readily, took less time or made fewer errors, compare the outcomes in the relevant work. The same questions apply to a generated note.
Retained result. Keep only the identity-and-bounded-sufficiency screen, an explanation class when it changes the next action, the compact bounded/blocked use, and a reopen condition. Add reader-model, trace, provenance, evidence, RAG, self-consistency, or interactive-system details only when their exact source-scoped situation is present.
Relations
- Uses when current:
C.2.8for the extractable structural amount consumed by a bounded explanation use. Its comparison retains the identified episteme, expression that conveys it and qualified observer, while EFP retains explanation-use classification. - Builds on:
E.17.0,E.17,A.7,E.10.D2,A.6.B,F.9,F.18 - Coordinates with:
ConservativeRetextualization,RepresentationTransition,A.6.3.CSC Controlled Semantic Coarsening,E.17.ID.CR ComparativeReviewUnit,A.6.4,A.15,A.15.4,B.3,A.20,A.21 - Profile basis and main neighboring-pattern boundaries: E.17 supplies face discipline. E.17.0 supplies viewpoint/view conformance only when
U.Viewmembership is material. A shift toward new semantics, a coarsened narrower-use target, or a gate-bearing claim or effect leaves the profile. - Boundary notes: for a bounded comparison over a comparative review unit, use
E.17.ID.CR ComparativeReviewUnit. For explanation-like renderings with a declared source-loss mode, useA.6.3.CSC Controlled Semantic Coarseningwhen their narrower bounded claim or effect, blocked downstream claim or effect, and source-bearing reopen are primary. For retargeting, useA.6.4. For work and reliance consequences, use the patterns that govern those claims, includingA.15for alignment,A.15.1for dated Work, andA.15.4only while an appearance hides the needed prerequisite. For assurance and engineering-justification consequences, useB.3. For gate-bearing consequences, useA.21under the named gate’s applicable profile. UseA.20only for a named internal-constraint check under that pattern's subject and applicability conditions.
C.29 mathematical-lens use relation
When a published explanation expression uses a mathematical lens, EFP still classifies and bounds its explanation use. Cite the applicable
C.29output only for the mathematical-lens claim used. If the applicable C.29 result isMathLensUse.LensCandidateNote, retain its first-candidate recognition use and next lens-use action and output.CandidateMathObject?remains optional. For a load-bearing mathematical-lens claim, cite the exactMathLensUse.OneLine,MathLensUse.MiniCard, orMathLensUse.FullCardresult required by C.29. Keep recoverable the candidate mathematical object, lens mapping mode, preserved and lost structure, exposed invariant or distinction, and stop condition required by that output, plusLensUseBoundaryValueanddeclaredLensUsewhere C.29 requires them. IncludeblockedLensOverread?only when it passes F.19's plausible-reader test. Keep EFP's bounded explanation use and blocked downstream use explicit. Do not copy fields already recoverable through that exact reference. Add source-relation, evidence, face, or forbidden-use detail only when the receiving use makes it material; the mathematical-lens result does not make the explanation faithful, evidential, or admissible downstream by itself.
E.17.EFP:End
Compare Sources Within a Shared Review Frame (ComparativeReviewUnit)
Status: Stable
Plain-name. Bounded source comparison under a shared review frame.
Use this when. A team needs a small note, comparison sheet, or guided aid to compare available source epistemes or source publications. The unit should make one bounded contrast or a small set of contrast rows available for inspection, keep their shared review frame visible, and leave downstream claims and effects to the work that establishes them.
First-minute working moment. A team has two or more source-pinned notes, sheets, views, or review aids on the table. They need one comparison unit for their review: two design options for one release, two methods for one task family, two vendor bulletins for one control scope, two research syntheses for one uncertainty question, or two programme strategies for one initiative. The note compares the sources. Action selection, approval, ontology repair and wider control of the work remain separate. A comparison can settle which option meets an agreed criterion. Use the governing decision rule for a commissioned choice, and establish who is entitled to make or approve it when that matters.
First output. One bounded comparison whose minimum content for ordinary use can be recovered in the unit or its immediate review context. A source-linked paragraph or small comparison table may suffice. The card below helps recover that content. The card can stay in the review context unless a receiving contract requires its separate publication. Include a blocked-use or boundary field only when it changes this comparison’s interpretation or use:
What goes wrong if missed. A comparison unit is either dismissed as harmless prose or overread as equivalence, action selection, gate pressure, release approval, work or reliance guidance, or adjudication authority. The team then argues about hidden authority instead of inspecting the bounded contrast.
What this buys in practice. The team can compare the available sources, inspect one bounded contrast or a small comparison sheet, and identify a claim that goes beyond the comparison and the pattern that governs it when the boundary trigger fires.
Not this pattern when. If the primary question is no longer the bounded comparison unit or its shared review frame, name the crossed claim and apply the governing pattern for that claim: source transformation, bridge, explanation face, prompt or action selection, ontology or EntityOfConcern change, decision, work or reliance, gate, assurance, adjudication, or reduced-use source rendering.
Quick working-fit check.
- Is the comparative review unit my work object?
- Does the comparison preserve the shared review frame and keep distinct alternatives separate?
- Is one bounded contrast or small row set being made visible?
- Is the downstream claim or effect still outside?
If yes, stay here and keep the ordinary minimum recoverable. If no, use the neighboring-work boundary in E.17.ID.CR:4.5.
Problem frame
Engineer-managers, programme leads, and research or cultural reviewers often need to prepare or share a small unit that helps a team read two available source epistemes or source publications together. The aim is to compare them without overstating the downstream claim or effect the unit carries. Typical moments include:
- a design-review note that says one already available option write-up foregrounds coupling risk more than another;
- a release or compliance comparison that says an internal control sheet and a vendor bulletin are not yet equivalent even though they speak to the same review task;
- an operations comparison that says a dashboard view and a maintenance note foreground different operational pressures in the same service episode;
- a research-review note that says one available synthesis foregrounds measurement uncertainty more than another without yet declaring a better method;
- a program or cultural review note that says one available brief foregrounds participation continuity more than another without yet deciding funding, curation, or program direction.
These comparison units make the next review discussion more precise. Their risk is that a reader may treat a comparison as establishing equivalence, root cause, redesign priority, action selection, program choice, or approval.
Problem
Without a named comparative-review-unit discipline:
- a useful comparative review unit is dismissed as harmless prose;
- a cautious review aid is read as a licence for substitution, interoperability, or equivalence;
- a calmly worded comparison can conceal pressure to select an action or conceal work on a hypothesis;
- same-entity viewing, explanation rendering, and bounded comparison are treated as one kind of review;
- ontology-facing target shift or changed EntityOfConcern hides inside comparative wording;
- a review unit written to serve review is mistaken for work or reliance guidance, assurance shorthand, or release authority.
Forces
Solution - comparative review units with bounded comparison, escalation, and boundary rules
Ordinary comparative review-unit move
Make one bounded comparison from the claims the reviewed sources support. Start with the receiving review question, recover the relevant claims and conditions in each source, and derive the contrast on a common basis. A source-linked paragraph or small table can carry the result. Use the construction below when the comparison basis or contrast is not already established.
Obtain the comparison from the available material
Recover the question and its criterion. Identify what the receiving review needs to distinguish and why: for example, the restoration time a release plan can guarantee, the populations covered by two studies, or the conditions under which two Methods can be used. Read the request, applicable requirement and surrounding review context for the object, situation and consequence that make a difference relevant.
Turn that difference into an answerable question about each source. If the purpose is only to compare the accounts, the criterion can concern their coverage or emphasis; state that subject. If a necessary priority or requirement is missing, ask the receiving reviewer which consequence or condition the comparison must resolve. Meanwhile retain any useful source contrast whose basis is already clear.
Recover each source's answer. Locate the passages bearing on that question, including definitions, qualifications and supporting results. For each alternative, recover what is asserted, about what, under which conditions and with what support. Keep the source return beside the extracted claim when it would otherwise be lost. Read enough surrounding argument to distinguish a stated result, an assumption, a proposed result and an unanswered question.
When a source is silent, report the missing coverage. Infer absence only when an explicit negative claim or a sufficiently complete account supports it.
Establish the common basis. Put the relevant claims together and ask whether the proposed predicate applies to both under compatible interpretations. A shorter reported time is comparable only after its event, unit, scope and status are understood: one observed duration, a mean and an upper bound answer different questions. A common topic does not make them interchangeable.
Recover a missing interpretation from the source's definitions or cited basis. Use the subject's comparison Method for the claim being compared: C.16 for measurement conditions, for example, or F.9 when a required semantic correspondence connects distinct local senses. If the correspondence or required fact remains unavailable, split the contrast into what each source establishes and state the exact missing basis for the stronger comparison. Different conditions can themselves be the useful contrast.
If the sources concern different constituents or give models at different levels of abstraction, recover what each establishes for the same review question. To judge whether their results hold together, identify the shared conditions or the relation under which one result constrains the other (C.37.MM:4.1.1). One source can supply sufficient grounds for the whole contrast when it compares both alternatives under that criterion. A claim about one alternative can also settle a request for one eligible option.
Derive and bound the result. State what follows from the aligned claims and show the reason that connects them. For textual emphasis, identify the content or allocation of discussion that warrants the contrast; this alone says nothing about the described alternatives' actual performance.
For a claim about the subject, preserve its applicable comparison rule and evidence. Distinguish a contrary result from insufficient support. Do not force a common ranking when the sources support different dimensions or incomparable conditions. A precise question such as “obtain B's restoration bound at production load L” is a usable return when that answer is necessary for the requested comparison.
Publish for the receiving use. Keep the sources, comparison basis, supported contrast and conditions needed to interpret it recoverable. Use an adjacent pair of passages, a paragraph or a table according to what the reader needs to inspect together. If the wording, arrangement or intended use could reasonably be taken as a stronger claim, state the limit on that stronger claim and the condition for reopening it. Apply the relevant neighboring Method when that stronger claim is requested.
Worked construction. A release brief asks whether a rollback plan can establish restoration within eight minutes at production load L. Note A states a ten-minute upper bound at L. Note B reports one six-minute rollback in staging at 0.2L. The reviewer first locates these passages and recovers the different status of the numbers: A gives a bound under the relevant load; B gives one observation under a smaller load. The shared question is the supported production restoration bound, not which displayed number is smaller.
The resulting note can say: “A establishes an upper bound of ten minutes at L; B reports six minutes once at 0.2L. Neither statement establishes the required eight-minute production bound. A may have a tighter supported bound; B needs support for a bound at L. These notes do not establish which rollback is faster in production.”
The comparison identifies the next inquiry: inspect A's bound derivation for an applicable tighter result, or obtain B's qualified bound at L from its validation or analysis. Select further testing only if the available support cannot answer the receiving question and that answer is still needed.
Now suppose B supplies a qualified seven-minute upper bound at L with the same restoration endpoint and applicable conditions. The comparison can establish that B supplies a bound meeting the eight-minute requirement while A's stated bound does not establish that fit. If the brief asks for one plan with an established bound, this result supports putting B forward. Choosing B under a standing rule to use the first plan with a demonstrated bound meeting that requirement needs that rule, which may already be in the brief. Ranking actual execution times would require different evidence.
If B instead says nothing about restoration time, the result is a source-coverage gap. The same construction applies to non-numerical sources: align the claims and their conditions, then state only the contrast that those claims support.
In plain working terms, this pattern is for a review unit that says something like:
this option write-up foregrounds integration pressure more than that one;these two available source epistemes or source publications are useful together, but they are not yet equivalent;this dashboard view helps triage one contrastive question, but it is not yet a release decision or a root-cause claim;this research synthesis foregrounds uncertainty more than that one, but it is not yet a method choice;this program brief foregrounds continuity risk more than that one, but it is not yet a funding decision.
If your review needs this kind of comparison, keep the ordinary minimum recoverable. If the first move is no longer bounded comparison over pinned sources, name the crossed claim and let its governing pattern carry that claim before this unit is used.
Compact placement
ComparativeReviewUnit governs this comparison within the wider InterpretationDiscipline naming family. The family name helps readers find the interpretation area; it does not govern the local claim. The local object is one comparative review unit carrying one bounded comparison, or a small set of bounded contrast rows, over already available source epistemes or source publications.
Use E.17.ID.CR, ID.CR, or ComparativeReviewUnit when this bounded comparison unit is the current object. Use the neighboring pattern when the crossed claim becomes primary.
Why the comparative-review-unit specialization needs its own discipline
Teams already produce small comparative review units, often as comparison notes, comparison sheets, or guided review aids, that add more interpretive lift than a short F.9.1 stance note about an existing bounded-use claim but still stop below action selection, ontology reframing, retargeting, or approval guidance. Leaving that middle band unnamed creates two opposite failures: one reader dismisses the review unit as harmless prose, while another reads it as carrying substitution, action-selection pressure, or action authority.
This pattern gives teams a narrow way to prepare, share, and inspect that comparative review unit without implying a downstream claim or effect beyond what the source, bridge stance and bounded use support.
Local working vocabulary
This pattern uses a small local vocabulary for review.
- Comparative review unit = a lightweight review unit such as a short comparison note, small comparison sheet, guided review aid, or guided comparative UI whose stated purpose is one bounded comparison or a small set of bounded contrast rows under one shared review frame.
- Base governing case = the primary source relation, pattern-governing case, or project work question that already governs the review use before bounded comparison is added.
- Reviewed source episteme or source publication = an available source episteme or source publication under comparison, pinned or otherwise reviewable.
- Source references =
sourceAnchorSetorsourceRefsthat make the interpreted source episteme or source publication inspectable. - Shared review frame = the review target, described situation, decision situation, release candidate, method family, control scope, problem frame, or source-set reference that remains preserved while the comparison is made.
- Compared alternative = one distinct option, method, bulletin, strategy, note, view, source episteme, source publication, or project-side FPF kind and reference named by value kept separate under the shared review frame.
- Same
EntityOfConcernRefcase = the special case where the compared sources describe the same entity. This is common, but it is not required when distinct alternatives remain under one shared review frame. - Interpretive lift = the bounded contrast derived from available source claims; in a small comparison sheet, each row has its comparison criterion and material conditions under the shared review question.
- Bridge references = required
bridgeOccurrenceRefandboundedUseClaimRefwhen the case depends on bridge-mediated correspondence rather than ordinary source interpretation alone. The use-claim reference resolves a claim whoseEntityOfConcernis that Bridge occurrence and whose proposed use, direction, correspondence rule, tolerated loss, and polarity match this comparative unit. OptionalbridgeCardRefcites reusable packaging, and optionalbridgeStanceRefcites a separate F.9.1 episteme whoseEntityOfConcernis the same bounded-use claim. - Bounded comparative use = what this review unit can be used for while it remains only a bounded comparative review unit.
- Overread risk = how the review unit is most likely to be overread into a bridge, action-selection, ontology, or authority claim that it does not carry.
- Prompt boundary = the point at which an explanatory question with rival explanations still open becomes primary under
B.5.2.0orB.5.2;U.AbductivePromptclassifies the published account of that initiating question underB.5.2.0:4.2.1–4.2.2. - Ordinary minimum block = the smallest ordinary record that preserves what a reviewer needs to use the comparison within its bounds.
- Load-bearing extension = the fuller declaration record used when the case sits close to bridge, explanation, abductive, ontology, or authority boundaries.
These terms are local review fields for completing the comparative review unit. They keep source references, shared review frame, compared alternatives, bounded lift, blocked downstream claim or effect, and boundary trigger readable in the review unit. When one of those fields starts carrying a bridge, evidence, gate, speech-act, commitment, work, authority, publication-face, or project-side FPF claim, name that crossed claim and use the governing pattern for it.
Scope and exclusions
In scope
- bounded comparative asymmetry over already declared reviewed source epistemes or source publications;
- reader-facing interpretive caution that stays source-tethered and preserves the shared review frame;
- comparison of distinct alternatives under one shared review target, described situation, release candidate, method family, control scope, problem frame, or source-set reference;
- comparative review units that answer one explicit contrastive question without creating a rival action-selection search;
- bounded user-fit when that fit only limits use rather than widening authority.
Out of scope
- same-entity restatement, conservative rewrite, or representation shift whose main question stays with
A.6.3,A.6.3.CR, orA.6.3.RT; - a separate F.9.1 stance note that only clarifies an already constituted F.9 bounded-use claim;
- explanation-face use discipline, bounded-use boundary, or added-link review on existing faces (
E.17.EFP); - explanatory questions with rival explanations open (
B.5.2.0orB.5.2), and choice among already available actions, probes, or method options (C.11); - ontology-facing reframing or changed EntityOfConcern (
OntologicalReframingorA.6.4); - policy, gate, adjudication, assurance, or work-facing use (name the claim and use its governing pattern; see
E.17.ID.CR:4.5).
Working-fit test
Use this discipline only when all of the following hold:
- the reviewed source episteme or source publication is already pinned or otherwise reviewable;
- the review unit adds one bounded comparative or interpretive lift, or a small set of bounded contrast rows with row-level comparison criteria;
- the case is still answering a bounded contrastive question rather than selecting an action;
- the shared review frame stays preserved, and compared alternatives remain distinct unless an explicit bridge or substitution source supplies equivalence, substitution, or another named relation between them;
- the main question is not already better described as same-entity viewing, an F.9.1 stance note about an existing bounded-use claim, or explanation-face use discipline.
If any of those fail, handle the current work under the neighboring FPF pattern and project-side FPF kind and reference named by value that govern it.
Nearest neighboring work
Name the base source relation or work question before adding bounded comparison. If the current question is already source transformation, bridge, explanation-face use, prompt or action selection, ontology or changed EntityOfConcern, decision, work or reliance, gate, assurance, adjudication, or reduced-use source rendering, do not stretch ComparativeReviewUnit to carry it. Use the compact boundary map in E.17.ID.CR:4.5 and the governing pattern for the crossed claim.
Working-model first; plain questions first, ordinary minimum second, full declaration third
Most users can start with the questions below.
Following E.14's working-model-first discipline, the questions help an engineer-manager keep the comparison within the work it supports.
The ordinary minimum block states the content needed for ordinary use; the card is one way to recover and arrange it.
The fuller declaration block remains available as a reviewable declaration extension that carries source, boundary, and downstream-claim fields by value. If an assurance or B.3 threshold applies, cite the separately constituted B.3 claim or record. The declaration extension cannot serve as that assurance record.
Five plain working questions
The near-top working-fit check selects the comparison task. Use these questions to recover its result:
- Which available sources and alternatives answer the receiving review question?
- Which claims, conditions and supporting passages in each source bear on it?
- On what criterion can those claims be compared, and what does the comparison establish?
- What necessary interpretation or fact remains unavailable, and where can it be obtained?
- Does the present wording or receiving use imply a stronger claim whose grounds must be supplied?
Use :4.1.1 to obtain a missing comparison basis or contrast. Repair an ambiguous local expression under E.17.AUD.LHR when wording is the obstacle; use E.17.AUD.OOTD when the unit's primary subject or move remains unstable. Missing subject evidence remains a substantive gap.
Ordinary minimum block
For an ordinary bounded comparison, keep recoverable:
- the reviewed sources and passages carrying the relevant claims;
- the shared question, distinct alternatives and comparison criterion;
- the supported contrast and conditions needed to interpret it; and
- any unresolved basis that limits the requested result.
Add a downstream-use limit and its governing pattern when the wording, representation, receiving use or applicable authority rule gives rise to a stronger reading than the comparison supports. Preserve a required contractual form and the grounds of any actual equivalence, causal, choice, gate or authority claim.
Stop if the same note, sheet or review aid cannot keep those minimum answers stable because it confuses reviewed source episteme or source publication, bounded comparative review unit, bounded lift, and outside work. Repair local lexical-head kind pressure through E.17.AUD.LHR (Local Head Restoration). If the whole review unit still has unstable primary-subject identification (including the carried EntityOfConcern when applicable) or carried-move identification after that repair, apply E.17.AUD.OOTD (PublicationUnit Primary-Subject Discipline) before adding more declaration weight.
Ordinary working card
An ordinary comparative review unit keeps the applicable content below recoverable. Use these rows when their arrangement helps:
The working card may appear inline or in the immediate review context. Retain a card or table when its arrangement supports comparison or independent revision of contrast rows. Keep any form a receiving contract requires. When the unit travels separately, carry the interpretation and source return needed for that use. Use these questions as the ordinary recovery reference for the near-top working-fit check:
- if rows 1-4 are still unstable because one local head word or qualifier has to carry too much of the meaning, stop and repair the wording through
E.17.AUD.LHR(Local Head Restoration) before continuing to build the comparative review unit here; - if the applicable content in rows 3-7 cannot stay stable because the same review unit still has unstable reviewed-source, comparative-move identification, or outside-work boundary after a local wording repair, apply
E.17.AUD.OOTD(PublicationUnit Primary-Subject Discipline); - if the applicable rows stay recoverable over one pinned source slice or source pair, one preserved shared review frame, distinct alternatives where present, and one bounded contrast or small row set,
ComparativeReviewUnitremains the primary governing pattern.
The nearest stay-here worked slices for this pattern are E.17.ID.CR:5.4.5 through E.17.ID.CR:5.4.6.b.
The nearest stop-and-reopen worked slice is E.17.ID.CR:5.4.6.c.
Use the fuller declaration extension only when one of the boundary, reader-fit, or misuse conditions in E.17.ID.CR:4.3.c becomes true.
ComparativeReviewUnit remains primary only while the ordinary minimum stays recoverable and the same review unit is still mainly about one bounded comparison, or a small set of bounded contrast rows, over already pinned source epistemes or source publications.
If the first question is what the review unit is about, what move it carries, and what wider work remains outside, use E.17.AUD.OOTD (PublicationUnit Primary-Subject Discipline) to stabilize that PublicationUnit question before extending the declaration.
Fuller Declaration Extension Guidance
A fuller declaration record becomes warranted only when a local condition changes the first action: reader fit changes how the comparison can be used, overread risk is high, a mixed case depends on A.6.3.* or E.17.EFP, or a bridge-mediated relation is live. Resolve remaining subject or move instability through E.17.AUD.OOTD before choosing the declaration weight, as below.
The fuller declaration extension can inherit already-declared case ids, source pins, and provenance references instead of restating them inline. When recorded as a claim-bearing review unit, that extension normally captures the ordinary minimum block plus only the neighboring-pattern fields that govern the mixed case.
Do not answer PublicationUnit instability by stacking more local fields onto the fuller declaration extension. If E.17.AUD.LHR (Local Head Restoration) has already repaired the local lexical-head pressure and the same review unit still has unstable reviewed-source, publication-unit, comparative-move identification, or outside-work boundary, stabilize that PublicationUnit question with E.17.AUD.OOTD (PublicationUnit Primary-Subject Discipline) before deciding how much declaration weight stays here.
Fuller Declaration Block
When the fuller declaration is warranted, make at least these fields recoverable:
sourceRelationClassusing the sharedE.17:5.1bvocabulary when the comparison depends on source pointer, source availability or retrieval, source use, source faithfulness, claim recoverability, contradiction, omission, claim widening, added linkage, independent verification, bounded use, forbidden downstream use, or reopen trigger;sourceAnchorSetorsourceRefs;comparativeRelationClass = sameEntityComparisonClass | sharedFrameDistinctAlternativeClass | readerFitComparativeClass;comparisonBasis;addedClaimPolicy;- required
bridgeOccurrenceRefandboundedUseClaimRefwhen the case depends on bridge-mediated comparative relation; the use-claim reference resolves a claim whoseEntityOfConcernis that Bridge occurrence and whose proposed use, direction, correspondence rule, tolerated loss, and polarity match the current comparative unit; - optional
bridgeCardRefwhen a reusable Card exists; - optional
bridgeStanceRefwhen it resolves the separate F.9.1 episteme whoseEntityOfConcernis that use claim; targetUserModelwhen reader-fit is materially shaping the comparison unit;interactionModewhen the review unit is not limited to a single static comparative sentence;contrastiveQuestionwhen the case answers a contrastive question;boundedComparativeUse;overreadRisk;promptWorthinessThreshold;ontologyBoundaryTrigger;worldContactPolicy;downstreamAuthorityLimit;baseCasePatternwhen the review unit is a mixed case layered overA.6.3.*orE.17.EFP.
sourceRelationClass classifies only the source relation or bounded claim for the local claim or use. comparativeRelationClass classifies only the comparative relation of this review unit. Neither field is a neighboring object or claim such as a relation kind, Bridge occurrence, bounded-use claim, Card, stance note, semantic identity, evidence relation, gate, assurance, work relation, speech act, commitment, authority reference, or decision record.
The sameEntityComparisonClass value is a special case for comparisons where the compared sources describe the same entity; it does not assert semantic identity. When the unit compares distinct alternatives, use sharedFrameDistinctAlternativeClass plus distinct alternative refs. Equivalence or substitution requires its applicable direct relation. When that use depends on correspondence between local senses from different semantic contexts, also establish the F.9 Bridge and its required bounded-use claim.
readerFitComparativeClass by itself does not create an interpretation claim. When bounded correspondence wording implies a cross-context Bridge, first apply F.9. The boundedUseClaimRef must resolve a claim whose EntityOfConcern is the bridgeOccurrenceRef, and its proposed use, direction, correspondence rule, tolerated loss, and polarity must match this comparative unit.
A positive proposed use requires affirmative polarity; when A.10 or B.3 is triggered, current reliance must support that use. A degraded reliance result narrows the use. Negative, abstaining, reopened, evidence-needed, blocked, or mismatched results stop this bridge-mediated use.
The pattern that directly constrains the proposed comparison decides authorization, and evidence of the comparative-review Work says whether it occurred.
A bridgeCardRef remains optional packaging. A bridgeStanceRef is also optional and is admissible only when it resolves a separate F.9.1 episteme whose EntityOfConcern is that same bounded-use claim. None of these references can substitute for another.
Use the main comparison question and the neighboring patterns’ boundaries to select the FPF pattern or project-side FPF kind and reference by value.
Interpretant-side block
The interpretant-side fields describe how reader fit affects this comparison. They do not specify a full interactive explanation or dialog-management system. Use them to:
- identify how the comparison depends on its audience when it does;
- make the contrastive question, guided review mode, and bounded use visible;
- prevent comparison prose from implying prompt-bearing guidance, assurance or policy pressure.
Static note versus interactive aid
Use two comparison-relation forms.
- Static comparative review note. A static note needs its sources, comparison criterion, supported contrast and material conditions. Use the
E.17:5.1bsource-relation class when that relation is disputed, and an explicit downstream-use limit or boundary trigger when the current use requires it. - Interactive comparative aid. Add
targetUserModel,interactionMode, state or history needed for the comparison claim,overreadRisk, and bounded-use boundary only when the aid is interactive, stateful, adaptive, or user-model-bearing. These fields keep the interactive comparative aid from being mistaken for audience-neutral static prose; they do not carry a crossed claim.
A comparative review unit can expose or cite the source epistemes, source publications, or project-side FPF references being compared, but layout, fluent contrast, side-by-side placement, or guided-review reuse does not change the kind of the unit or create a stronger source relation.
If the required source relation is missing, the repair request or source-gap note is prospective only; it does not backdate a source relation into the earlier comparison.
Comparative-review-unit identity over revision. A revised comparison table, regenerated comparison note, or updated guided review aid is not the same bounded comparison merely because the layout, title, or compared-source family stayed familiar.
If new source input, revised source references, changed comparison criterion, changed shared review frame, or a change in the blocked downstream claim or effect changes the comparison identity or downstream use, publish the preserved comparative frame and the changed claims, or treat the result as a new comparative review unit before using it for a stronger crossed claim.
Representation ontology and modeling lens (informative)
This pattern does not model interpretation in general.
It models one comparative review unit governed by ComparativeReviewUnit within the broader InterpretationDiscipline family.
In each case, the object under review is the comparison unit.
That unit can appear as a comparison note, comparison sheet, or guided review aid, but it is not the whole review process or the source system.
The bounded comparison is the interpretive lift carried by that review unit.
The minimum typed lens is a compact record of:
- source references and source relation;
- one declared source-relation class;
- one declared comparison criterion and added-claim policy;
- one bounded-use boundary, one overread-risk line, and one
worldContactPolicythat remains subordinate to the pattern governing a gate or adjudication claim;A.21governs a named gate decision under its applicable profile, whileA.20applies only to a named internal-constraint check in a transformation-flow case; - the relevant prompt, ontology, and authority boundary triggers;
- and which neighboring pattern still governs the base case when this remains a mixed overlay.
The lens keeps attention on the review unit and the problem-owning review domain, while the pattern governing the base case retains responsibility for source, continuity and boundary conditions.
Working read-out
A working reader can usually say, in one short paragraph:
- what reviewed source episteme or source publication is being comparatively read;
- what bounded interpretive lift is being added;
- what shared review frame remains preserved, and, in the special same-EntityOfConcern case, why the same
EntityOfConcernRefremains preserved; - which crossed claim is still outside this pattern and which neighboring pattern would govern that claim if it became primary;
- and which boundary condition shows that the primary claim is no longer a bounded
ComparativeReviewUnitclaim.
If the reader cannot tell what the comparison establishes, return to the receiving question. Clarify the source claim or the inference that bears on it. If the review unit supports only a narrower comparison, say what that comparison supplies and pursue the missing ground for the original question. The person who commissioned the work can agree to a different question.
Branch-discipline summary
This section is the compact governing-rule summary for ComparativeReviewUnit inside the Core. Use the fuller solution, boundary table, worked slices, and relations section here only when you need the wording of a clause, the full field set, or the full reopen conditions.
-
Preserve the shared review frame. Keep the reviewed source episteme or source publication set, source references, declared comparison criterion, and distinct alternative identities visible. State
contrastiveQuestionwhen the review depends on it. -
Keep the lift bounded and comparative. The review unit can add one bounded comparative or asymmetry-bearing lift. It stops when that lift starts carrying a stronger crossed claim.
-
Name the crossed claim instead of repeating exclusions. When the case stops being bounded comparison, name the claim that crossed the boundary and apply the pattern that governs it; use
E.17.ID.CR:4.5to identify the relevant boundary. -
Keep neighboring-pattern authority explicit. Bridge-mediated comparison requires a
bridgeOccurrenceRefand a tuple-matchedboundedUseClaimRefwhoseEntityOfConcernis that Bridge.Positive use requires affirmative polarity and, when A.10 or B.3 is triggered, current reliance for that use. Degraded reliance narrows the use; a negative, abstaining, reopened, evidence-needed, blocked, or mismatched result stops it.
A Card and F.9.1 stance note remain optional and separate. Authorization and evidence that comparative-review Work occurred remain with their own patterns and records.
-
Keep reader-fit bounded.
targetUserModel,interactionMode,contrastiveQuestion,boundedComparativeUse, andoverreadRiskcan be stated when they change how the comparison is used, but these fields cannot confer authority.
Neighboring-work boundary glance
Use this table to distinguish the comparison from neighboring project work and identify the source requirements for each claim. For a fuller reading of a mixed case, read the applicable neighboring pattern with its row.
For first-minute use, the following four boundaries summarize the rows around the comparison case. Use them with the near-top working-fit check and the ordinary working card:
- pressured local lexical head ->
E.17.AUD.LHR(Local Head Restoration); - stable same-object comparative review unit -> stay with
ComparativeReviewUnit; - same unit still unstable after local repair ->
E.17.AUD.OOTD(PublicationUnit Primary-Subject Discipline); - any stronger crossed claim already primary -> the governing pattern for that claim is primary.
If the comparison unit is already carrying neighboring work, use the boundary rows first and then read E.17.ID.CR:5.4.7 through E.17.ID.CR:5.4.10 as the nearest worked boundary examples.
Ordinary working order
The ordinary working order is to recover the receiving question, inspect the relevant source claims, establish their comparison basis and derive the bounded contrast as explained in :4.1.1. Publish that result with its source return and material conditions. Add a downstream limit or neighboring Method only when the current claim or receiving use requires it.
If the result cannot yet be obtained, return the missing source, interpretation, criterion or fact and the attainable inquiry that could settle it. Use E.17.AUD.LHR or E.17.AUD.OOTD when the obstacle is respectively local wording or unstable primary-subject or move identification; those repairs do not supply missing subject evidence.
Archetypal grounding
Worked-slice note. The system case, episteme case and worked boundary examples cover different situations. Choose the example that fits the case; their order does not prescribe a progression. They show different bounded outcomes for the same governing pattern: some comparisons stay small and stop, some also need a neighboring pattern, and others reopen or need another governing pattern when new observations from outside the review, environmental changes or downstream constraints change what the comparison can support.
Make the ordinary minimum recoverable for the current comparison unit; if the boundary trigger fires, stop here or apply the governing pattern for the crossed claim.
One comparison unit
ComparativeReviewUnit applies when a team prepares one stated, bounded comparison over source epistemes or source publications with already declared references, while any stronger crossed claim remains outside until the governing pattern for that claim is named.
Keep the unit small enough for a reviewer to recover the shared review frame, the same EntityOfConcernRef when the compared sources describe the same entity, the declared comparison criterion, the blocked downstream claim or effect, and the boundary trigger that would end interpretation.
Same service episode
Source slice. Two pinned operating notes describe the same service episode from different operational responsibilities.
One note has its source reference in the maintenance log, the other in the continuity dashboard for the same declared episode and the same EntityOfConcernRef.
Comparative review unit. Under the declared comparison criterion, the maintenance note foregrounds operator-induced variance, while the continuity note foregrounds buffer-sensitive drift; each view exposes a blind spot in the other without granting direct substitution.
Why this stays here.
- source relation and source references are explicit;
- the same
EntityOfConcernRefremains preserved; - one bounded comparative lift is added;
- no substitution licence is added;
- no rival action-selection question is yet being asked.
Two analytic renderings
Source slice. Two pinned analytic renderings over the same evidence set are already available for review.
One rendering is a SourceLinkedExplanationReconstruction on a TechCard face; the other is a compact comparison sheet that preserves the same evidence set and the same described operational episode.
Comparative review unit. For maintenance reviewers, the reconstruction foregrounds operator load more than the comparison sheet, while the comparison sheet foregrounds recovery sequencing more than the reconstruction; this difference is useful for review, but it is not yet a design recommendation or an action-selection claim.
Why this stays here.
- the base-case governing patterns remain identifiable;
- the comparative lift is explicit and bounded to one reviewer task;
- explanation-face governance and same-entity transform discipline remain with their neighboring patterns;
- explanatory questions, action selection, and authority-bearing use remain governed by their respective neighboring patterns.
Worked boundary examples
Lower-boundary stance-note case
Stance-note unit. F.9 already records an obtaining Bridge and a bounded-use claim for reading the local maintenance-pressure term alongside the partner continuity term. This separate note says that, for that use, the relation is best read as asymmetry-explicating rather than substitution-friendly.
Why it stays under F.9.1:
- the Bridge and bounded-use claim already exist;
- the note only makes the claim easier to read; and
- no bounded comparative lift beyond that stance note is added.
Mixed primary-pattern composition with A.6.3.RT
Base-case rendering. A same-entity comparison sheet retabulates one pinned incident note into columns for trigger, pressure, and recovery.
Comparative review unit. In the retabulated view, the recovery column makes the operator-induced asymmetry easier to inspect than the trigger column, but the table is not treated as establishing a new causal hierarchy.
Why the case uses both patterns:
A.6.3.RTstill governs the base representation shift;- bounded comparison is secondary and adds only the bounded contrast;
- most restrictive forbidden-use constraint wins, so no new ontology or gate claim is licensed.
Mixed primary-pattern composition with E.17.EFP
Base-case rendering. A TechCard-face explanation rendering is already classified as SourceLinkedExplanationReconstruction and publishes a bounded connective policy.
Comparative review unit. For maintenance reviewers, this rendering foregrounds the difference between operator load and throughput pressure more than the original prose, but it is not treated as a design-level recommendation.
Why the case uses both patterns:
E.17.EFPstill governs explanation class and face bounded use;- bounded comparison adds only bounded comparative use for one reviewer task;
E.17.EFPstill governs explanation-face use; any authority-bearing use must name its governing pattern.
Guided review aid with bounded interaction mode
Source slice. A reviewer UI presents two pinned source notes side by side for the same described operational episode.
Guided comparative review unit. Question: which note foregrounds variance introduced by operator timing rather than environmental drift? Bounded comparative use: bounded comparative triage only. Misuse risk: do not treat this aid as action selection or release guidance.
Why it stays here:
- the interaction mode is explicit but still bounded;
- the review unit answers one contrastive question without initiating work on an explanatory prompt or action choice;
- bounded use and overread risk are stated in the aid instead of being implied by its tone.
Product and design-review comparison case
Source slice. Two available design-review notes concern the same integration boundary and planned release. The release meeting needs to know what each note establishes about rollback and implementation effort.
Note A states: Rollback uses the existing adapter, which must first be modified. No implementation-effort estimate is available. Note B states: Implementation uses one interface and is estimated at two engineer-days. Rollback has not been assessed.
Comparative review unit. For release review, A describes a rollback path conditional on modifying the adapter, while B leaves rollback unassessed. B supplies a two-engineer-day implementation estimate; A supplies no estimate. These source statements do not establish which option has lower total effort or risk. If the next decision needs that comparison, obtain the missing estimates and rollback assessment on a compatible basis.
Working-boundary use. The shared review frame, bounded contrast and boundary with outside work are all explicit. The comparison therefore remains primary.
Why it stays here:
- the same planned release remains the
EntityOfConcernRef; - one bounded comparative lift is made explicit for a declared review task;
- the unit helps design discussion without selecting an action or granting approval.
Compliance and release-review comparison case
Source slice. An internal control checklist and a vendor compliance bulletin are already available for the same release candidate and the same declared control scope.
Comparative review unit. For release review, the vendor bulletin foregrounds protocol conformance more than rollback evidence, while the internal checklist foregrounds rollback evidence more than protocol conformance; this comparison helps frame the review, but it is not yet a release gate or equivalence claim.
Working-boundary use. Under release or compliance pressure the comparison unit remains stable. The review concerns its bounded contrast, with no need for local wording repair or PublicationUnit stabilization.
Why it stays here:
- the comparison criterion is explicit and bounded to one review task;
- the source references remain visible and the same release candidate stays in view;
- the review unit helps an engineer-manager see the difference in emphasis without giving the comparison authority over the release gate.
Research-review comparison case
Source slice. Two already available research syntheses discuss the same measured phenomenon and the same declared evidence slice. One synthesis foregrounds variance decomposition limits more; the other foregrounds protocol repeatability more.
Comparative review unit. For method review, the first synthesis foregrounds variance decomposition limits more than the second, while the second foregrounds protocol repeatability more than the first; this asymmetry helps frame the discussion, but it is not yet a method choice or a claim that one synthesis is globally better.
Why it stays here:
- the same measured phenomenon remains the
EntityOfConcernRef; - the comparative lift is bounded to one review task;
- the unit helps research discussion without selecting an action or reframing the ontology.
Program and cultural-review comparison case
Source slice. Two already available programme briefs discuss the same continuing initiative and the same declared participation scope. One foregrounds continuity of community engagement more; the other foregrounds short-term event visibility more.
Comparative review unit. For programme review, the first brief foregrounds participation continuity more than the second, while the second foregrounds short-term visibility more than the first; this comparison helps frame the discussion, but it is not yet a funding, curation, or programme-direction decision.
Why it stays here:
- the same initiative remains the
EntityOfConcernRef; - the comparison criterion is explicit for one declared review task;
- the unit helps programme discussion without giving the comparison authority to decide.
Exogenous-change stop-and-reopen case
Source slice. An internal release-review comparison sheet already compares one control checklist and one vendor bulletin for the same declared release candidate and the same control scope. Mid-review, an external incident bulletin arrives and changes the rollback assumptions governing use of that same candidate.
Initial comparative review unit. Before the new bulletin, the vendor bulletin foregrounds protocol conformance more than rollback evidence, while the internal checklist foregrounds rollback evidence more than protocol conformance; this comparison frames the review, but it is not yet a release gate or equivalence claim.
Working-boundary follow-through. The initial comparison fits E.17.ID.CR:5.4.6, but the new observation changes its declared criterion. Stop and reopen the unit.
Why this stops and reopens.
- the new outside observation changes the declared comparison criterion;
- the previous bounded comparison can remain traceable, but it cannot be used unchanged under the new review conditions;
- the bounded next use is either to restate a fresh comparative review unit over the new declared criterion or to apply a neighboring governing pattern if downstream gate or authority question has now become primary.
Lighter comparison note with source-return discipline
Source episteme and source publication set. A release team already has the full internal rollback worksheet, vendor bulletin, and incident-note bundle for one release candidate. A short comparison note is then prepared for the daily review stand-up.
Comparative review unit. For today's review, the vendor bulletin foregrounds protocol conformance more than rollback evidence, while the internal rollback worksheet foregrounds rollback evidence more than protocol conformance; this short note is only a review aid over the same release candidate, and the full source episteme or source publication set remains primary for any bridge, release, coarsening, or work or reliance claim.
Why it still stays here:
- the comparison unit is still one bounded comparative review unit, not a replacement for the source episteme or source publication set;
- the note remains source-pinned through the already available source episteme or source publication set, and bridge, gate, or work or reliance use remains with the governing pattern for that claim;
- any attempt to treat the short note as enough for equivalence, release approval, execution claim or effect, or a reduced-use source substitute requires source-bearing return,
A.6.3.CSC Controlled Semantic Coarsening, or another neighboring pattern.
Functional versus constructive-description comparison
Source slice. A functional-description publication and a constructive publication or product-description publication describe the same pumping skid. The functional description foregrounds flow relation and method-selection relation; the constructive description foregrounds module composition and installed equipment.
Comparative review unit. For design review, the functional description foregrounds what the skid is supposed to do in the declared flow relation, while the constructive description foregrounds what parts are present. This contrast helps the engineer keep function and construction separate, but it is not a module-equivalence claim, a performed-work record, or a gate decision.
Why it stays here:
- both source publications keep source references and remain inspectable;
- the same pumping skid remains the
EntityOfConcernRef; - the bounded comparative lift is the function-versus-construction contrast for one review task;
- module equivalence, work occurrence, evidence, and gate claims remain blocked downstream uses.
Method-option comparison without method choice
Source slice. Two method descriptions are already pinned for the same fabrication task. One foregrounds lower setup cost; the other foregrounds tighter result-measurement discipline.
Comparative review unit. For method review, the description of method M-1 foregrounds lower setup cost more, while the description of method M-2 foregrounds tighter result-measurement discipline more. This comparison helps prepare method discussion; method choice, a work plan, and evidence that either method has been performed remain outside it.
Why it stays here:
- the reviewed source epistemes are pinned;
- the comparison criterion is explicit;
- no selected-method, work-plan, performed-work, evidence, or engineering-justification claim is added;
- if the team chooses a method or prepares a work plan, record the selected method as project
U.Method, record the work plan asU.WorkPlanunderA.15.2, and useA.15.1only when a datedU.Workoccurrence is the governed occurrence claim.
Nearest neighboring-work examples. Use the next four cases to recognize prompt pressure, same-entity viewing, ontology shift, and gate or authority misuse when the boundary rows near the beginning point beyond the comparison.
Upper-boundary prompt-bearing case
Prompt-bearing review unit. "This contrast raises the question whether both systems are being constrained by the same hidden gating variable, so we normally publish a U.AbductivePrompt around that shared control possibility."
Why ComparativeReviewUnit no longer governs:
- an explanatory question with rival explanations still open governs the next action;
- the review unit is now prompt-bearing rather than only interpretive;
- the selected governing pattern is
B.5.2.0orB.5.2through explicitU.AbductivePromptpublication.
Same-entity viewing boundary case
Viewing rendering. The source note is retabulated into a compact comparison sheet that preserves the same claims and entity but makes pressure, trigger, and recovery fields easier to inspect.
Why it does not enter interpretation:
- the main question is representational reshaping rather than bounded comparison;
- no bounded asymmetry or interpretive claim is added;
- the more precise governing pattern is
A.6.3.RT.
Ontology-boundary anti-case
Ontology-pressuring review unit. The older maintenance note and the new field-observation note are best treated as two observational cuts over the same latent failure mode, so we normally recast both under a new operational kind and treat the source labels as legacy labels.
Why ComparativeReviewUnit no longer governs:
- the case is now asking for a same-referent interpretation with continuity-witness demand or retargeted-EntityOfConcernRef interpretation;
- continuity witnesses would now be needed;
- bounded comparative interpretation is no longer enough, so the case applies
OntologicalReframing.
Authority and gate misuse anti-case
Authority-pressuring review unit. Because this comparison consistently foregrounds the safer operating condition, reviewers can use the review unit as a release gate and do not need the underlying source episteme or source publication during triage.
Why ComparativeReviewUnit no longer governs:
- the review unit is being overread as gate-facing authority;
- the bounded comparison has become a substitute for the source episteme or source publication;
- the release-gate claim is governed by
A.21and the applicable release policy; useA.20only for an internal-constraint check that meets its transformation-flow conditions.
Invalid publication and repair example
Source slice. Source note SN-12 and receiving note RN-12 use maintenance terms from different local semantic contexts.
Invalid review unit. These two views describe the same entity for current operations, so the team can use whichever wording is easier.
Why it is invalid here:
- no source references are visible;
- bridge-mediated comparison is being implied without identifying an obtaining Bridge and its bounded-use claim;
- the invitation to use either wording implies substitution and authority that the comparison does not establish.
Minimal repair. Under Bridge B-12 between the local senses pinned by SN-12 and RN-12, bounded-use claim UC-12 has B-12 as its EntityOfConcern and says, with affirmative polarity, that the SN-12-to-RN-12 direction is suitable for comparing only the operator-timing concern in this review task; the remaining source distinctions are tolerated loss, not substitution. A current A.10 result supports relying on UC-12 for that use. Both notes foreground the concern, but they are not substitution-equivalent and the source episteme or source publication set remains primary. The pattern for the proposed review decides authorization, and evidence of the review Work says whether it occurred. Card BC-12 may be cited when that optional package is useful.
What the repair does:
- restores the source references and verifies
bridgeOccurrenceRef, the bounded-use claim'sEntityOfConcern, its use/direction/rule/loss/polarity tuple, and current A.10 reliance for that positive use, while keeping anybridgeCardRefoptional; - narrows the claim back to bounded comparison and leaves authorization and the occurrence of comparative-review Work to their own patterns and evidence;
- reasserts the blocked downstream claim or effect.
Bias-Annotation
Scope: bounded comparative review units governed under ComparativeReviewUnit inside InterpretationDiscipline, not all review, all publication, or all decision work.
This pattern focuses on a bounded comparison over available sources. Recover the source claims, shared question, comparison basis and supported contrast. When the receiving work asks for a stronger claim, use its governing Method and retain the grounds already established by the comparison.
Fluent comparison can hide a stronger use. Make reader fit, interaction mode and source relation visible only when they change the unit’s use. Authority-bearing claims remain with their governing patterns and project-side FPF kinds and references named by value.
Conformance Checklist
A conformance check is retained only if it changes the next bounded use of the comparative review unit, blocks a concrete overclaim, or preserves a source reference or reopen condition needed for the declared bounded use.
Use ID.CR-Core for ordinary comparison notes. Conditional rows apply only when the note touches neighboring-pattern relation, bridge declaration, or reader-fit fields. For a fuller reading of a mixed case, use this checklist with the applicable neighboring pattern and the boundary conditions below.
ID.CR-Core ordinary checks
- CC-ID-1 - Bounded comparative review unit is explicit. The pattern distinguishes the bounded comparative review unit from the surrounding review or decision work.
- CC-ID-2 - Source references and comparison criterion are explicit. A reviewer can see which fixed source episteme or source publication is being interpreted and which declared criterion or contrast supports the comparison.
- CC-ID-3 - The lift stays bounded. The unit or its immediate review context keeps the supported contrast, comparison basis and material conditions recoverable. State a blocked use or reopen condition when it changes this comparison’s interpretation or receiving use.
- CC-ID-6 - Neighboring-pattern boundaries stay visible. When the boundary trigger fires, the neighboring FPF pattern carries that prompt, ontology, action, gate, authority, or downstream claim instead of leaving it hidden inside comparative prose.
- CC-ID-8 - The review unit does not over-claim authority. Any stronger authority claim requires its governing basis. Require an explicit review-only or non-executive limit when the unit’s wording, arrangement or receiving use could otherwise imply that authority.
ID.CR-Conditional checks
-
CC-ID-4 - Base-case governing-pattern relation is explicit. A reviewer can tell why the case does not belong to
A.6.3.*, an F.9 Bridge or bounded-use branch, an F.9.1 stance-note branch,E.17.EFP,B.5.2(.0),OntologicalReframing, orA.6.4. -
CC-ID-5 - Bridge declaration does not hide. If the case depends on bridge-mediated comparison,
bridgeOccurrenceRefandboundedUseClaimRefare required. The latter resolves a claim whoseEntityOfConcernis that Bridge and whose use/direction/rule/loss/polarity tuple matches the comparative unit.Positive use requires affirmative polarity and, when A.10 or B.3 is triggered, current reliance for that use; degraded reliance narrows it, while a negative, abstaining, reopened, evidence-needed, blocked, or mismatched result stops it.
Authorization and the occurrence of comparative-review Work remain separate. Optional
bridgeCardRefremains packaging; optionalbridgeStanceRefresolves a separate F.9.1 episteme whoseEntityOfConcernis that claim. -
CC-ID-7 - Reader-fit stays bounded.
targetUserModel,interactionMode,contrastiveQuestion,boundedComparativeUse, andoverreadRiskare visible when needed, but they do not create an authority claim that the unit does not carry.
Checklist recovery map. To recover a checklist row by value for assurance, use its ordinary card row and worked example below. Consult these sources before relying on the checklist alone:
Common Anti-Patterns and How to Avoid Them
Consequences
- The middle band between a short F.9.1 stance note about an existing bounded-use claim and prompt-bearing abduction becomes reviewable rather than rhetorical.
- Reviewers get a cleaner way to distinguish comparative interpretation from the first crossed claim that would make another governing pattern primary.
- Authors have a little more to declare. The comparison unit is more stable, and fewer mistakes at the boundaries with neighboring patterns go unnoticed.
- Guided comparative review units become easier to prepare: authors can state bounded use, overread risk and world-contact limits without implying a broader guidance claim.
- Users get a bounded way to keep comparative review units modest while the boundary trigger remains below the first crossed claim.
Rationale
Teams already write small comparative review units, often as comparison notes or sheets, to move a review forward. What they usually lack is a disciplined way to keep that unit useful without letting it imply equivalence, an unstated hypothesis, pressure to redesign, or a release decision.
The pattern protects that everyday use of bounded comparisons. It keeps a comparative review unit usable by making five entries visible enough to inspect: the bounded comparative review unit, the source references, the bounded comparative lift, the blocked downstream claim or effect, and the boundary trigger that would end interpretation. A team can compare available source epistemes or source publications while keeping the authority of the comparison within its bounds.
SoTA-Echoing: Adopted and Adapted Invariants and Rejected Shortcuts
Row 1. The inspected ISO abstract supplies architecture-description lineage, not a rule for comparative review units. This pattern itself requires visible source references and comparison criteria so the reviewer can inspect the contrast. The Reviewed source, Source references and Bounded lift rows carry that local requirement; applying them makes no claim of conformity to the complete standard.
Row 4. Jacovi and Goldberg examine whether an interpretation reflects the reasoning behind an NLP model's prediction, separately from whether people find it convincing. A comparative review unit here instead compares claims in available sources under an explicit criterion. Its source-text fidelity follows from comparing what the note says with those sources, keeping their qualifiers, omissions and added contrasts visible under E.17:5.1b and this pattern's ordinary working card.
A note may reproduce a source accurately while the source's explanation misrepresents a model; a faithful model explanation may also be misquoted in a note. The evaluation question therefore selects what must be checked. Human review of source claims remains useful; this source analogy imposes no NLP-model intervention on an ordinary static comparison.
The Blocked downstream claim or effect and World-contact limit rows keep further bridge, prompt and authority claims subject to their own grounds. This lets an engineer-manager inspect what the comparison warrants without giving polished prose authority over the source episteme or publication set. Case linkage: see E.17.ID.CR:5.4.6, E.17.ID.CR:5.4.9, E.17.ID.CR:5.4.10, and E.17.ID.CR:5.4.11.
Relations
- Citation. Cite this pattern as
E.17.ID.CR,ID.CR, orComparativeReviewUnitwhen the current object is one bounded comparative review unit. UseA.6.3.CRwhen the intended pattern isConservativeRetextualization. - Placement.
ComparativeReviewUnitsits inside the widerInterpretationDisciplinenaming family, but the local governing object is the comparative review unit and its bounded comparison. The wider review or decision work remains outside until a crossed claim becomes primary. - Builds on:
C.2.2a,A.16.0,F.9,F.9.1, Part F,A.6.9, andE.14. - Coordinates with:
A.6.P,A.6.3,A.6.3.CR,A.6.3.RT,A.6.3.CSC,F.9,F.9.1, Part F,A.6.9,E.17.EFP,B.5.2.0,B.5.2,OntologicalReframing,A.6.4,C.11,A.15,A.15.2,A.15.4,A.20,A.21. - Boundary map. Use
E.17.ID.CR:4.5when the comparison starts carrying relation precision, bridge or sameness, prompt or action selection, ontology or changed target, decision, work or reliance, gate, assurance, adjudication, or reduced-source-use claims. The neighboring pattern governs only the crossed claim it names. - Local repair neighbors. Use
E.17.AUD.LHRonly when a local lexical head or qualifier still destabilizes the same unit; useE.17.AUD.OOTDonly when the reviewed-source, publication-unit, comparative-move, or outside-work boundary is still unstable after local repair.
C.29 mathematical-lens use relation
When a bounded comparative review unit uses a mathematical comparison criterion, rival lens, invariant, obstruction, or structural similarity,
E.17.ID.CRstill works over comparison unit, viewpoint, comparison criterion, review-unit boundary, and bounded-use boundary.The applicable
C.29output for the stated use can be cited only for the mathematical-lens use. If that output isMathLensUse.LensCandidateNote, retain its first-candidate recognition use and next lens-use action and output;CandidateMathObject?remains optional.When citing an output for a claim-bearing mathematical-lens use, use the exact
MathLensUse.OneLine,MathLensUse.MiniCard, orMathLensUse.FullCardresult required by C.29 and keep recoverable its candidate mathematical object, lens mapping mode, preserved and lost structure, and stop condition, plusLensUseBoundaryValueanddeclaredLensUsewhere C.29 requires them; includeblockedLensOverread?only when it passes F.19's plausible-reader test.Keep the review unit's bounded use and blocked downstream claim or effect explicit. The C.29 output does not create the comparison record, adjudicate rival publications, or authorize bridge, evidence, selector, or benchmark claims outside the comparative-review-unit record.
E.17.ID.CR:End
Keep a Publication Unit Stable (PublicationUnit Stability Discipline)
Status: Stable
Plain name. Keep one publication unit stable enough for readers to identify its subject, publication move and outside boundary.
Problem frame
Use this pattern when a person or an agent is meant to interpret a note, memo, sheet, table, screen or short section as one unit, and a change of subject, publication move or boundary with a decision, gate, work or reliance claim goes unmarked. A validator asked only whether the material meets a syntax rule can finish with that finding. To report what the intended reader understood, use evidence showing how that reader interpreted the content.
A typical case starts with one bounded architecture or status question and ends by sounding like rollout, approval, assignment, or assurance. One reviewer wants to repair a vague word, another wants to rewrite the whole unit, and a third sees a comparison or explanation problem. Before they patch different defects, identify the bounded publication unit and its current interpretation.
Name the unit’s primary subject or the subjects whose connection the reader needs. When it carries or exposes a claim-bearing U.Episteme or episteme-side U.View, use that item's EntityOfConcern value only when it is also the unit's primary subject. Otherwise name the ordinary topic or subject and do not invent an EntityOfConcernRef. Keep the publication unit distinct from the episteme, publication occurrence, form, face, carrier, and any downstream project claim.
The primary reader is an author or reviewer who needs one usable repair choice. Architects, managers, and program leads are secondary readers when the same unit is being over-read as architecture, approval, or work guidance.
If this check is missed, teams repair one word when the whole interpretation has shifted, rebuild a whole unit when one local head was enough, or polish a comparison, explanation, or status note until it looks like evidence or approval.
The check helps the team choose one response early: keep the unit as it is, repair one local head, stabilize the whole unit, treat it as a bounded comparison, or leave this pattern for the applicable neighboring pattern and project record.
Use E.17.AUD.LHR when an overloaded local head is the only defect. A stable unit presenting a bounded comparison calls for E.17.ID.CR; an explanation-use problem calls for E.17.EFP. If the text is already being used to approve, direct, assign, adjudicate, or support reliance, use the applicable decision, gate, work, evidence, or reliance pattern. In those cases the narrower or neighboring pattern governs the work.
The first useful result is one of those five repair choices. If the reader can already identify the unit, its subject and move or its connected subjects and moves, and the boundary with outside work, return stable for current use and stop. Use the checks and examples as needed. Their order is optional.
Problem
Without a named publication-unit stability discipline:
- teams repair local wording when the interpretation of the whole unit is unstable;
- teams open whole-unit stabilization when the defect is one overloaded local lexical head;
- teams extend a publication-unit repair when the task already calls for a bounded comparison;
- teams mistake note, sheet, table, or screen labels for different kinds of publication unit when they name different presentation forms of one unit;
- teams infer an engineering-process, approval, or rollout claim or effect that the text does not support.
Forces
Solution
Recover the question the unit was meant to answer. Read what it actually asks the reader to understand or do. If the two questions differ, decide what the passage contributes to the original one (C.39.QR:4.1). Name its subject, the move it carries and any boundary that changes the reader’s use. A request confined to identifying what a passage says can end with that reading.
A joint decision may require two subjects and the relation between them. Keep that connection readable. A note assessing agreement between the catalog and checkout must let the reader compare the responses the catalog can return with those checkout can handle. The connected claims can share one short passage or occupy marked sections that the reader can use together. For a scope notice, simply identifying the subjects and the question can suffice.
Plain working terms
publication unit under review= one note, memo, sheet, table, screen, or short section that readers inspect as one unit;primary subject= what the unit is mainly about. UsepublicationUnitPrimaryEntityOfConcernonly when the unit’s primary subject is also the exactEntityOfConcernof the claim-bearing episteme or episteme-side view it carries. A unit can use claims about several entities to explain a relation between them. For other subjects, use the non-claim-bearing kind named by value or an ordinary topic or subject without inventing anEntityOfConcernRef. An ordinary topic or subject value is not a claim about a C.2.1 entity participant;carried publication move= the claim, interpretation, comparison, or explanation move the unit makes about that primary subject;outside boundary= a decision, gate,U.Work, work planning, reliance claim, or continuing engineering work that remains outside the unit;local lexical head= one word or phrase such asreview,interpretation,note, ortextwhose meaning is unstable inside an otherwise stable unit;repair choice= stable for current use, local-head repair, whole-unit stabilization, bounded comparison, or leave publication-unit stability for explanation classification, bridge or hypothesis work, representation change, controlled coarsening, a changed primary EntityOfConcern, or a downstream action, authority, adjudication, decision, gate, work, or reliance claim;applicable pattern and project reference= the FPF pattern to apply plus, when the live claim needs it, the exact evidence, gate, decision, work-plan, work-occurrence, method, action-invitation, or relation record, selectedU.Episteme, or exactEpistemePublicationRelationoccurrence when availability matters;publication-unit stability family=E.17.AUD,E.17.AUD.LHR, andE.17.AUD.OOTDtogether with their comparison and explanation neighbors, grouped here for selecting the applicable repair;presentation-form label=note,memo,sheet,table,screen, or a similar description of form. The label alone does not establish the unit’s kind.
Use the terms above only when their distinctions change the repair choice.
Minimum admissible interpretation
A locally admissible interpretation keeps four entries available for inspection by value:
- one publication unit under review;
- the primary subject, or the subjects whose relation matters to the reading;
- the move about that subject, or the moves whose connection the reader needs for one question;
- one outside boundary to work, work planning, decision, gate, or reliance claim, with one light boundary type when that distinction matters: neighboring pattern application, downstream claim or effect, or ongoing engineering-process continuation.
If the publication unit changes any of those four without saying so, its interpretation has shifted, however polished the sentences look.
Publication-unit stability vs whole-unit requirement
Ordinary result. Choose one of the following responses.
stable for current use: the four-part interpretation is explicit enough and none of the neighboring questions named above is live;local lexical-head repair: applyE.17.AUD.LHRto the overloaded head;whole-unit stabilization: applyE.17.AUD.OOTDto the unit;bounded comparison: if the unit is stable, applyE.17.ID.CR;leave publication-unit stability: the live question concerns work, work planning, decision, gate, evidence, explanation, reliance, carrier or front-end work, or another claim that this pattern does not test; apply the relevant pattern and name the exact project object or record.
Leave the narrower whole-unit check to PublicationUnit Primary-Subject Discipline. Its question is whether this unit can still keep a stable primary subject or connected subjects needed for one decision, a recognizable publication move, and an outside boundary to work, work planning, decision, gate, or reliance claim.
Inherited dynamic frame
When the unit describes a lineage or move, use C.2.2a or A.16.0 for that subject and inspect how the unit expresses it.
Kind and boundary
Treat one publication unit as a readable unit. Do not identify it automatically with:
- the
U.Epistemeor episteme species whose claims the unit carries, quotes, or describes; - an
EpistemePublicationRelationoccurrence, publication form, or carrier involved in making that selected episteme available; - the primary subject, including the exact EntityOfConcern when applicable;
- a generic publication face or MVPK face under E.17 constraints;
- a carrier or evidence carrier;
- proof, evidence record, assurance claim, or release admissibility;
- a view or viewpoint;
- an engineering-process stage;
- a downstream decision, gate, work, or reliance publication.
Those objects may matter, but mentioning them in the same note, sheet, or screen does not make them the current publication-unit problem.
Publication-unit boundary choice. One subject, one move and a visible boundary with outside work suffice for an ordinary short note. A joint decision can require several subjects and moves in one readable unit. Keep their relation visible in the passage or in marked sections. For example, one section can specify which responses a catalog returns and another can test which of them checkout accepts. Readers need both to judge the interface. Mark the boundary with work, work planning, decisions, gates, reliance claims and neighboring pattern applications. If the unit undertakes one of them, use its governing rule.
Choose the bounded item that carries the claim being made or effect being repaired. Do not choose a smaller boundary merely to hide a downstream overclaim, or a larger one merely to absorb several primary subjects into one unit.
A table row may be the unit when that row carries the claim; the whole table may be the unit when the table-level caption or comparison frame carries the claim. A dashboard tile, note, card, sheet, or screen block may be the unit only when that bounded item, not the whole carrier or interface, carries the live publication move.
Publication-unit snapshot identity. A PublicationUnit may remain the same bounded unit while its carrier rendering, export format, screenshot, or layout changes. It does not remain the same stabilized interpretation by visual or file continuity alone.
If a revision, refresh, translation, regeneration, or dashboard update changes the primary subject, carried publication move, outside boundary, source pins, or admissible use, rerun the four-part interpretation for the new snapshot before the unit is used for comparison, explanation, evidence, gate, decision, work, or reliance claims.
Ordinary working card
When a wider repair needs these distinctions, use the seven questions to recover them:
Choose the next pattern
- If row 5 points to one overloaded local lexical head, apply
Local Head Restoration. - If row 5 shows that readers cannot follow the subjects and moves the unit connects or the boundary with work, planning, decision, gate or reliance, apply
PublicationUnit Primary-Subject Discipline. - If the publication unit is stable enough and its move is bounded comparison over available source publications, apply
E.17.ID.CR ComparativeReviewUnit. - If the main problem is explanation classification over an existing face, apply the neighboring explanation pattern.
- If claim content, representation, coarsening, or the primary EntityOfConcern changes, apply the relevant
A.6.3orA.6.4pattern before checking a later publication form here. - If wording alone suggests approval, repair the unit to say what its source warrants. If an authorized decision exists, identify its issuer, scope and outcome. If the reader needs a decision that is still owed, put the choice and grounds before the responsible authority. Check which act the governing rule makes decisive. An authorized publication can constitute the decision itself or bring an earlier decision into effect. Establish the issuer, act and form required for the particular claim.
- If the active problem concerns publication form, bridge or hypothesis work, or another downstream action, apply its direct pattern and identify the project object or record the work needs.
Local naming rule
Treat ordinary labels such as note, memo, sheet, table, screen, review, and status as clues to presentation form. A label alone does not establish the kind of publication unit.
Working rule:
- if one overloaded local lexical head is doing most of the semantic work, repair that local lexical head first through
Local Head Restoration; - if the local lexical head does not explain the instability, keep the publication unit stable in the whole-unit stabilization pattern. Adding another qualifier could hide the shift in interpretation;
- cleaner or more formal wording cannot make an inadmissible downstream claim, effect, or comparison source relation admissible.
Keep a needed model or rationale visible
If the primary subject or the carried publication move depends on a modeling substrate or rationale, publish that substrate or rationale briefly in the unit or move the case to a fuller publication form or neighboring pattern that can carry it. When the interpretation depends on a formal model or rationale, wording repair alone is insufficient.
Keep stronger claims separate
When explanation, comparison, or a downstream claim is load-bearing, keep five facts recoverable to preserve the repair choice:
- evidence status and source-pin status when the unit leans on already available source publications;
- which reliance or interpretation of work is currently admissible on the basis of this unit, and which decision, work or gate claims cannot be made on that basis;
- whether this unit is the primary publication unit or a derivative helper publication;
- any claim-bearing modeling substrate or rationale;
- how any assurance finding corrects or bounds the opening claim according to its result. If the revision also proposes a downstream claim or effect, distinguish that proposal from what the finding establishes.
Worked slices
Local-head case
A semio note keeps saying this review and this interpretation, but nobody can tell which FPF kind or locally declared head those lexical heads name here. Once those terms are clarified, the rest of the unit is stable. Use Local Head Restoration; a whole-unit repair is unnecessary.
Whole-unit interpretation-shift case
A memo starts about one bounded architecture question over an inherited lineage or move, then shifts into wider rollout or approval language without declaring the transition. Repairing one sentence does not stabilize the publication unit under review because its concern and carried publication move have widened; that widening does not by itself establish a changed EntityOfConcern. Use PublicationUnit Primary-Subject Discipline.
Stable-unit comparison case
A comparison sheet already keeps one stable primary subject and one clear outside boundary to work, work planning, decision, gate, or reliance claim, but the team is using publication-unit instability language because the comparison is contentious. Use E.17.ID.CR ComparativeReviewUnit. The team can address the contested comparison without further stabilization.
Explanation-laundering case
An onboarding explainer starts from one stable source-pinned note, but then the simplified prose begins to sound like canonical assurance or policy. The publication unit may still be readable, yet its main problem has shifted from unit stability to explanation use. Use E.17.EFP's simpler source-linked boundary sentence first; instantiate ExplanationFaithfulnessProfile only when explanation-class ambiguity still changes the next action.
Downstream decision and reliance case
A progress card headed “Ready to roll out” reports that the tests passed. If its job is to communicate progress, a heading that states which tests passed can finish the repair. To report a release decision already made, cite the authorized outcome, including rejection or deferral, and the scope it covers. If release approval is owed and still open, send the test results and the outstanding choice to the responsible authority. Suppose the owner approved release on Monday and the rule makes that approval effective when the signed notice is issued. Monday’s decision establishes adoption. Issuing the notice on Tuesday establishes when that decision took effect. Under a different rule the owner decides by issuing the signed notice. There, the authorized issuance itself constitutes adoption. Check the rule and the act it names for the conclusion the reader needs.
Quick contrasting cases
Use this quick contrast set when the initial interpretation is unclear:
Common Anti-Patterns and How to Avoid Them
Consequences
- You pause to name the problem with the publication unit before patching the draft.
- You reduce unnecessary escalation from one overloaded local lexical head into a whole-unit rewrite.
- You reduce the opposite failure too: trying to solve whole-unit interpretation instability with one more qualifier on the same local lexical head.
- You keep neighboring publication-unit repair patterns and neighboring non-publication-unit patterns explicit so that a broad stability label does not obscure which pattern applies.
- You make it harder for clearer prose, official-looking formatting, or wider circulation to be mistaken for a downstream claim or effect.
Rationale
Local lexical-head repair and whole-unit primary-subject stabilization address different problems. Authors and reviewers need one check that distinguishes those cases, recognizes a bounded comparison, and shows when the work has moved outside the publication-unit stability family.
The pattern has a narrow scope. It does not turn every publication-unit problem into publication design or downstream decision, gate, work, or reliance work. It helps readers identify the publication unit’s primary subject, the move it makes about that subject, and which downstream U.Work, work planning, decision, gate, or reliance claim remains outside.
SoTA-Echoing
Claim 1. FPF’s current treatment of EntityOfConcern and descriptions keeps the entity of concern distinct from the claim-bearing episteme or publication that describes it. This supplies the exact-entity case of the publication-unit subject distinction.
Practice, source, alignment, and adoption. C.2.1, A.7, E.17, and the description patterns keep an EntityOfConcern, a description episteme, its publication occurrence, form, and carrier distinct. ISO/IEC/IEEE 42010:2022 is standards lineage for the narrower architecture/architecture-description distinction, not the source of this general publication-unit ontology.
PublicationUnit Stability Discipline adapts that FPF distinction to one readable unit and requires a primary-subject shift to be explicit; an EntityOfConcern change requires the exact C.2.1 case. For a reviewer or architect, this distinction guides worked slices 5.2 and 5.3.
Claim 2. Information-for-use practice treats user-facing units as purpose-bound, structured information rather than as loose bundles that can mix explanation, instruction, warning, and decision or reliance effect by convenience.
Practice, source, alignment, and adoption. Joint IEC and IEEE 82079-1:2019 requires information for use to be purpose-directed, structured, and evaluated for usability.
PublicationUnit Stability Discipline adopts purpose-bound publication units and explicit outside boundaries to work, work planning, decision, gate, or reliance claim, adapts that discipline from information-for-use to notes, memos, sheets, tables, and screens, and rejects the shortcut where a clearer or official-looking unit is treated as approval, policy, gate, work, or reliance text.
For a manager or operator, this matters in worked slices 5.4 and 5.5: better explanatory form does not by itself establish downstream claim or effect.
Claim 3. Pattern-writing and pattern-validation practice keeps patterns tied to recognisable situations, explicit problem, solution, and consequence structure, and reviewable rationale rather than elegant internal naming alone.
Practice, source, alignment, and adoption. Iba (2021), How to Write Patterns, and Riehle, Harutyunyan, and Barcomb (2025), Pattern Discovery and Validation Using Scientific Research Methods, with the authors' 2021 preprint, treat pattern writing and validation as requiring recognisable situations, explicit structure, and reviewable reasoning rather than only elegant naming.
PublicationUnit Stability Discipline adopts worked slices, recognisable entry cues, and an explicit next-pattern and project-reference boundary, adapts those expectations to publication-unit stability work, and rejects a well-labeled pattern text that offers too little domain content or too little guidance for its reader.
For a reader using the pattern, this matters in the Problem frame and slices 5.1 through 5.5: the pattern should be usable before one has to reconstruct the surrounding rationale from the beginning.
Local stance. The unit is sufficiently clear for its declared use when readers can recover its primary subject, carried publication move and outside boundary, and apply the neighboring pattern needed by a different problem. This clarity judgement does not itself establish the truth or warrant of its claims.
Conformance Checklist
- CC-AUD-1 — One publication unit under review is explicit. The case names one note, memo, sheet, table, screen, or short section as the unit under review. Its presentation-form label alone does not identify that unit.
- CC-AUD-2 - Primary subject and carried publication move are explicit enough to identify the applicable pattern. The case identifies what the unit does for the reader’s question. When a passage turns to a constituent, another model or a worked example, the connection back to that question remains recoverable. A local move can supply one premise, settle the whole question or leave a gap.
- CC-AUD-3 — Outside-work boundary is explicit.
The case states what downstream
U.Work, work planning, decision, gate, or reliance claim remains outside the publication unit under review, including neighboring pattern application, downstream claim or effect, or ongoing engineering-process continuation when that distinction matters. - CC-AUD-4 — The active repair choice is named.
The case returns
stable for current usewhen the unit, primary subject, carried publication move, and outside boundary already suffice. Otherwise it names local lexical-head repair, whole-unit primary-subject stabilization, bounded comparison, or the applicable neighboring pattern rather than patching several problem situations at once. - CC-AUD-5 - The next pattern and project-reference boundary is explicit.
When the problem calls for
Local Head Restoration,PublicationUnit Primary-Subject Discipline,E.17.ID.CR ComparativeReviewUnit, an explanation-faithfulness pattern, or a downstream decision, gate, work, or reliance pattern, name the pattern to apply and the exact project object or record when one is needed. - CC-AUD-6 — Presentation-form labels do not launder publication-unit kind or downstream claim or effect.
note,memo,sheet,table,screen, and similar labels are clues to presentation form. They do not silently change the unit under review, create proof or evidence, make a release admissible, or establish a downstream claim or effect. - CC-AUD-7 - A claim-bearing modeling substrate or rationale remains visible. If the primary subject or carried publication move depends on a modeling substrate or rationale, publish that substrate or rationale briefly enough for review or handle the case by a fuller publication form or neighboring pattern that can carry the needed basis.
- CC-AUD-8 — Clearer prose does not silently widen downstream claim or effect. Readability, formatting, and wider circulation may improve the unit, but they do not by themselves turn the unit into approval, policy, assignment, gate, work, or reliance text.
Relations
-
Builds on:
A.7,E.10,F.18,E.14,E.19,E.17, andC.2.1. -
Coordinates with:
E.17.AUD.LHR Local Head Restoration,E.17.AUD.OOTD PublicationUnit Primary-Subject Discipline,E.17.ID.CR ComparativeReviewUnit,E.17.EFP ExplanationFaithfulnessProfile, andE.21when a pattern-quality card, table, status line, or generated summary is published as a bounded publication unit. UseE.17.AUDto test whether the publication unit keeps its subject, move and outside boundary clear andE.21to evaluate the underlying pattern-quality claim.Also use project-side patterns such as
C.11,A.10,A.15,A.15.4,B.3,A.20, andA.21when decision, evidence, gate, assurance, engineering-justification, work, or reliance claims become primary. -
Boundary consequence: when the publication unit can no longer be interpreted within the scope of this pattern, apply the neighboring FPF pattern and name the exact project object or record when one is needed. Publication-unit stability does not govern explanation, comparison, decision, gate, work, or reliance in general.
E.17.AUD:End
Restore the Local Meaning of an Overloaded Word or Phrase (Local Head Restoration)
Status: Stable
Placement. Local lexical-head repair within the PublicationUnit Stability Discipline.
Builds on. A.6.P, A.7, E.10, F.18, E.14.
Coordinates with. E.17.ID.CR, E.17.AUD.OOTD, E.17.EFP, A.6.3, A.6.3.CR, A.6.3.RT, A.10, A.15, A.15.4, B.3, A.20, A.21.
Plain-name. Clarify an overloaded word or phrase before its ambiguity spreads through the publication unit.
One-line summary. Use Local Head Restoration when the sentence leaves it unclear what a familiar word such as text, document, surface, review, or interpretation means there.
What to repair. The local lexical head under repair is a word or phrase in one publication unit. Its sentence no longer tells the reader what kind the head names. Restore its kind and local reading, the primary entity or relation when one is active, the action or question the phrase carries, and the boundary with the nearest outside work. Do this before later sentences inherit the ambiguity.
Use this when. A familiar word has an unclear FPF kind or locally declared meaning in a note, memo, review unit or table, or in a paragraph chiefly about episteme publication. Start here when that word is the source of the ambiguity and the need to stabilize the primary subject of the whole unit has yet to be established.
First example. A draft says this review, this text, this document, this publication, or this interpretation. People in the room read different FPF kinds or locally declared meanings into the same word. Repair the word before relying on the rest of the unit. At this point the example calls for no whole-unit rule check.
What goes wrong. The next three sentences inherit the ambiguity of one word. Reviewers argue about taste because the unit leaves them to guess whether it names a description, a carrier, a publication unit, a carried move, a governing pattern, or wider work.
Practical gain. Start with the smallest unit that can be stabilized. Repair the overloaded head and show its local reading and the question under repair. The team can then decide whether the publication unit needs a wider review.
Naming boundary. A repaired head can become a durable, reusable naming task under F.18. If one head repair clarifies the sentence and introduces no durable cross-context name, UTS row, Core-facing name, reusable FPF head, or high-risk label, skip the full Name Card.
The local result is the repaired head, its kind and reading, the primary entity or relation when one is active, the action or question under repair, and the boundary with outside work.
Success condition. LHR succeeds when a careful reader can identify the local lexical-head kind, the primary entity or relation when one is active, carried action or question under repair, and outside-work boundary for this sentence or small unit. If that is enough and the publication unit no longer shifts, stop. Apply E.17.AUD.OOTD only when readers still cannot follow which subjects and moves the unit needs to connect for their question, or where its claim ends.
When another approach applies. Use another approach when:
- the same publication unit still has unstable primary-subject or carried-move reading after local repair and now needs one stable answer to what it is about, what move it carries, and what remains outside;
- the question under repair is already one bounded comparative review move over an otherwise stable source episteme or publication;
- the main issue is view, face, carrier, publication architecture, or downstream approval, gate, adjudication, or execution work rather than an overloaded local lexical head;
- the text is already clear locally, and the unresolved problem is wider strategy, rollout sequencing, or architecture framing.
Who uses it. This pattern helps an author, reviewer, architect, or manager who needs a quick way to repair an overloaded local lexical head before the whole text overclaims.
Use the five questions in E.17.AUD.LHR:3.2 to recover the local repair.
Local-repair threshold. One local repair should restore the meaning of the overloaded head, its kind and local reading, the primary entity or relation when one is active, and the action or question the sentence carries. If the evidence leaves two readings open, retain the content both support. Resolve the difference only if it affects the question at hand. If the next sentence borrows a different kind, reading, or outside-work boundary from the same head, the problem extends beyond that repair.
When to use a neighboring pattern. If one local repair stabilizes the unit and the remaining question is a bounded comparison over pinned source epistemes or publications, use E.17.ID.CR (ComparativeReviewUnit). If its subjects, carried moves or outside-work boundary remain unclear after the word is repaired, use E.17.AUD.OOTD (PublicationUnit Primary-Subject Discipline). Adding more qualifiers to the same head will not settle that question.
Problem frame
The quick checks, ordinary card, worked cases, and rules for choosing a governing pattern or a project reference help repair one publication unit. They prescribe no canonical transformation flow or fixed order for all cases. One case may stabilize after one head repair. In another case, an outside observation may change the question and reopen the case. A unit whose primary subject or carried move still shifts may need E.17.AUD.OOTD.
The recurring defect is small but expensive:
- one broad familiar word enters early;
- the word is never restored to one kind or local work position;
- later sentences inherit its ambiguity.
Typical load-bearing local heads include:
documenttextartifactnotesheetpublicationsurfacefaceviewreviewinterpretationreading
These words can be appropriate. A word becomes risky when readers must infer the primary entity, active relation, local work position, move, or governing-pattern boundary before anyone has clarified what it names.
Problem
Without a named local restoration move:
- teams keep asking qualifiers to rescue an unstable local lexical head;
- one sentence names one FPF kind or locally declared head while the next sentence names the move over it;
- readers over-infer publication-unit meaning from one under-restored broad-family word;
- teams begin a publication-unit review too early for a problem that is still local;
- or an unresolved local head hides an unstable reading of the whole publication unit.
Solution
Local Head Restorationrepairs an overloaded word or phrase before later sentences in the publication unit inherit its ambiguity.It restores lexical-head kind, active local reading, carried action or question under repair, and any family, governing pattern, primary entity, and active relation on which the sentence relies.
Pairwise plain glosses
- Pressured local lexical head = a word doing more work than the sentence makes clear.
- Lexical-head kind = what FPF kind or locally declared head that word names here: for example description, carrier, publication unit, relation record, face, or view. When the word names an EntityOfConcern, recover that entity's kind.
- Active local work position = what work is at issue: for example review, publication, comparison, process, or authority.
- Active primary entity or relation = what the local sentence or publication unit is about, when such an object or relation is active.
- Move or question under repair = what the sentence is doing with the active primary entity, active relation, or local lexical-head repair object, if anything.
- Family, governing pattern, primary entity, and active relation set = when a broader family or governing pattern is active, name the family, governing pattern, primary entity, active relation, carried action or question under repair, and outside work separately rather than letting one familiar local lexical head carry them by implication.
Local reading. Ask what kind the word names, which work position is primary, which primary entity or relation is active, what action or question the sentence carries, and what remains outside. If these five answers no longer stabilize the same unit, apply its governing FPF pattern or use the project-side FPF kind and reference named by value.
Ordinary working card
Use these five questions to recover an ordinary local repair:
Use the card to recognize a local repair. Its rows need not be performed in order in every case. Use it while one overloaded local lexical head remains the main defect.
When family or governing-pattern language is load-bearing, make the repair disposition, governing pattern, primary entity or relation, carried move and outside-work boundary recoverable in the repaired text or its immediate context. A separate compact line can record them when the receiving review needs it.
Read the card as a three-way recovery aid:
- if rows 1-5 stabilize around one repaired local lexical head, one restored local work position, one active primary entity or relation, and one local question that the unit supports, stay here;
- if rows 1-5 stabilize locally and the remaining question is one bounded comparative review move over already pinned source epistemes or publications, apply
E.17.ID.CR; - if rows 2-5 still cannot stay stable because the same publication unit keeps borrowing a different object, move, or outside-work boundary from the same local lexical head, apply
E.17.AUD.OOTD. An extra qualifier will not stabilize that unit.
The nearest worked slices for those three repair dispositions are:
- ordinary stay-local:
E.17.AUD.LHR:5.2; - bounded comparison:
E.17.AUD.LHR:5.4; - whole-unit stabilization:
E.17.AUD.LHR:5.5.
Load-bearing extension
If the local case is close to a neighbouring-pattern boundary and the ordinary card already stabilizes the unit, add these checks:
- overloaded local lexical head;
- restored lexical-head kind;
- the supported local reading, or what is established across the remaining readings;
- restored active primary entity or relation;
- restored carried action or question under repair;
- restored outside-work boundary;
- any family, governing pattern, primary entity, and active relation distinction now made explicit;
- governing-pattern and project-side-reference decision.
Use that extension as the assurance section only when ordinary repair is already holding and the remaining risk is misuse at a neighboring-pattern boundary.
It is for the stay-local repair disposition, not for deciding again whether the case belongs in E.17.ID.CR or E.17.AUD.OOTD.
- If the ordinary card now shows one stable local repair plus one bounded comparative review question, apply
E.17.ID.CRbefore opening the extension. - If the ordinary card still shows publication-unit reading instability after local repair, apply
E.17.AUD.OOTDbefore adding declarations here.
Do not use it to rescue a unit whose publication-unit reading still shifts, and do not turn it into a second rule sheet.
Ordinary repair order
Use this order when one local lexical head is carrying too much:
- name the overloaded word;
- restore the lexical-head kind;
- recover the supported reading. If several remain, state what still holds and check a distinguishing fact only when the difference matters to the reader’s work;
- restore the active entity or relation, including the relevant whole or part when the same word can name either. For instance, both an entire review Method and one constituent can be called
the method. Read the sentence in light of the question the reader needs it to answer. - restore the carried action or question under repair, if any;
- restore any family, governing pattern, primary entity, active relation, and nearest outside-work boundary the sentence is relying on;
- choose the disposition the repaired unit supports: stay with local repair, apply bounded comparison, or apply publication-unit discipline.
A narrowing qualifier alone does not count as restoration. Treat this order as one local repair aid, not as a canonical flow. Steps 1-6 restore the overloaded local lexical head; step 7 classifies what the repaired unit supports as the next move.
- If step 6 keeps reopening because the reader still cannot follow which subjects and moves the unit connects or what remains outside, apply
E.17.AUD.OOTD. - If the local lexical head now has a clear reading and the only remaining question is one bounded contrast over already available source epistemes or publications, apply
E.17.ID.CR. The comparison alone does not require a fuller local repair record. - If the head is clear and no neighboring reading has become primary, stop without adding the extension.
Quick worked-slice starter
When you need a starting sentence, choose an example that fits the working situation:
Use these starters only as local examples. If outside observations or downstream constraints change what the sentence can claim, reopen the question using the governing FPF pattern or the project-side FPF kind and reference named by value. The starter does not set the first step of a fixed flow.
Worked slices
The following cases illustrate local repair, transfer to comparison or whole-unit repair, and reopening after changed observations. They are alternative outcomes, not a recommended sequence.
Worked-slice aid. The cases below display the overloaded local lexical head, its kind and active reading, the primary entity or relation, the carried action, outside work and repair disposition. Use this arrangement when it helps recover the repair; an adequate repaired sentence needs no separate output schema.
review carries two jobs
A note says:
This review establishes the release boundary for the service.
Two sentences later it says:
The review should therefore assign rollout responsibility to platform.
Local repair first:
- overloaded local lexical head =
review; - restored lexical-head kind = review publication unit;
- active local reading = boundary review, not responsibility assignment;
- primary entity/relation = the release boundary as made visible in this review unit;
- carried move = make one boundary visible;
- outside work = responsibility assignment.
If the request was to report the service boundary, the note can end with that boundary. If the same commission also calls for assigning responsibility, identify who can make the assignment and how the boundary bears on that choice. The authorized assignment has its own grounds. A proposal to change the job belongs with the person who commissioned it.
A published text is mistaken for the policy it describes
A paragraph says:
This text is the policy.
The intended claim is that this published text describes the policy.
Local repair:
- overloaded local lexical head =
text; - restored lexical-head kind = publication expression;
- active local reading = this expression of a policy description;
- primary entity/relation = the described policy;
- carried move = describe the policy rather than claim authority for it;
- outside work = approval, rollout, release, gate, policy, assurance, or adjudication status.
Repaired sentence. This published text describes the policy.
This is the ordinary stay-local case. One repaired local lexical head keeps later sentences from borrowing authority from the wrong local reading without forcing publication-unit stabilization.
Recovery reading. Stay in E.17.AUD.LHR: the local lexical head now identifies the intended expression, the same local unit no longer shifts, and no neighboring reading has become primary.
Semio-heavy family name does too much work
A semio note says:
This interpretation clarifies the package.
But the same paragraph is about one bounded comparison over one review unit, not about InterpretationDiscipline as a whole and not about the whole package.
Local repair:
- overloaded local lexical head =
interpretation; - restored lexical-head kind = bounded comparative review unit inside one episteme-publication-heavy paragraph;
- active local comparison = bounded comparison, not wider-family package explanation;
- primary entity/relation = comparative review unit;
- restored relation set = family
InterpretationDiscipline, governing patternComparativeReviewUnit; - move = bounded comparison;
- outside work = wider architecture strategy.
The repaired paragraph can answer a request for that comparison. If the reader asked how the whole package works, retain the comparison as a contribution and supply the connection between the compared parts and the organization of the package (C.39.GA:4).
Local repair selects bounded comparison
A comparison note says:
This review shows option A is safer than option B.
But the unit is one comparative review note over already pinned source epistemes or publications, not a publication-unit reading-instability case and not yet a publication-unit stability case.
Local repair:
- overloaded local lexical head =
review; - restored lexical-head kind = comparative review unit;
- active local comparison = compare the pinned options within this review unit; the whole release process remains outside;
- primary entity/relation = the proposed safety contrast between the options described by the pinned sources;
- carried move = make one bounded contrast visible over already available source epistemes or publications;
- outside work = rollout choice or approval.
Recover the declared safety criterion and source-backed contrast through E.17.ID.CR. If either is missing, leave the contrast unresolved and return to its sources; the repaired head alone establishes no comparison result. Once both are recoverable, the repaired sentence can say:
This comparison note shows how the pinned descriptions differ on the declared safety criterion; it does not by itself establish overall safety or approve rollout.
Recovery reading. Finish the local repair here. Use E.17.ID.CR for the remaining bounded contrast over the now-stable unit.
Local repair exposes publication-unit reading instability and must apply whole-unit discipline
A release note says:
This document records the release decision for the candidate.
After one sentence, the same unit starts talking as if it were:
- the review publication unit that compares evidence;
- the decision object;
- and the rollout work that follows if approval is recorded.
First identify what the entry actually contains. The word document can be replaced by release note on the quoted sentence alone. Its heading leaves open whether the entry gives a reviewer’s recommendation, an authorized outcome, or both. If all the reader needs is the note about the release question, the local repair can finish there.
If the entry attributes “defer until the seal test passes” to the authorized release owner and the provenance confirms that decision for this candidate, the editor can report the deferral. If the reviewer only recommended deferral, report the recommendation. An entry that contains both can identify the reasoning and the decision. If the distinction matters and the source leaves it open, state which outcome remains unknown and what would settle it.
In this example the following paragraphs still move from weighing evidence to reporting a decision and then to carrying out rollout. Mark the change of move and the boundary of the work that follows. A heading alone cannot grant rollout authority.
Recovery reading. Keep the restored head as the local result and stop expanding the local repair. Apply E.17.AUD.OOTD because the unit’s reading still shifts after that repair.
Outside observation changes what the same head can support
A status note says:
This note captures the current rollback state for the candidate.
Mid-review, a new vendor bulletin changes the reported failure boundary and pushes the surrounding conversation toward approval pressure.
Local repair can still clarify the current sentence:
- overloaded local lexical head =
note; - restored lexical-head kind = review publication unit;
- active local reading = current review publication unit, not downstream approval record;
- carried move = capture the rollback state visible on the current evidence slice;
- outside work = any new approval, adjudication, or widened authority step.
Outside observation has changed what this local unit can claim. Appending the new pressure to the old repair would conceal that change. Stop, reopen with a newly declared question, or use the governing pattern if approval, rollout, release, gate, policy, assurance, adjudication, or publication-unit stabilization has become primary.
Recovery reading. The outside observation has changed what the unit can claim. Stop and reopen or apply the neighboring governing pattern. Another local qualifier will not settle the new question.
Boundary dispositions
Use a different governing pattern when:
- the repaired local lexical head is no longer the main problem and the publication unit still has unstable primary-subject or carried-move reading;
- the same unit is already stable enough and the remaining question is one bounded comparative review move over already pinned source epistemes or publications;
- the problem is primarily about view, face, or carrier architecture;
- the unit is being used to communicate approval or a gate or adjudication decision, or to direct execution work;
- outside observation or environmental change has changed what the same local unit can claim, so the case now needs stop-and-reopen or application of the neighboring governing pattern rather than one more local qualifier.
Governing-pattern boundary recovery map.
The comparison-side neighbor is E.17.ID.CR ComparativeReviewUnit: use that governing pattern when the local lexical head now has a clear reading, the unit already stays about the same primary subject, and the remaining question is one bounded comparison over already available source epistemes or publications.
The main publication-unit neighbor is E.17.AUD.OOTD PublicationUnit Primary-Subject Discipline: use that governing pattern when local lexical-head repair is no longer enough and the reader still cannot follow the unit’s subjects, moves or outside-work boundary.
Treat those as neighboring recoveries, not as a required sequence. Some cases will stop after one local repair, some will apply bounded comparison under E.17.ID.CR, and some will apply publication-unit stabilization under E.17.AUD.OOTD once the question the unit supports changes.
Consequences
Used well, this pattern:
- prevents one vague local lexical head from governing a whole section by accident;
- keeps local repair cheap instead of escalating too early;
- makes later publication-unit stability review cleaner because the local head’s meaning has already been restored;
- gives authors and reviewers one common language for saying
the problem is still local.
Used badly, it can become one more vocabulary exercise. If the publication unit still has unstable primary-subject or carried-move reading after local repair, stop polishing the overloaded local lexical head. Apply the governing pattern for the remaining problem situation.
SoTA-Echoing
Relations
Builds on
A.6.P Relational Precision Restoration (RPR)E.10 Unified Lexical Rules for FPFF.18 Local-First Unification Naming ProtocolA.7 Strict Distinction
Nearest neighbors
E.17.AUD.OOTD PublicationUnit Primary-Subject DisciplineE.17.ID.CR ComparativeReviewUnit
E.17.AUD.LHR:End
Clarify the Primary Subject of a Publication Unit (PublicationUnit Primary-Subject Discipline)
Status: Stable
Placement. A pattern in the PublicationUnit Stability Discipline for stabilizing one publication unit.
Builds on. A.6.P, A.7, E.10, F.18, E.14, E.19, C.2.2a, A.16.0.
Coordinates with. E.17.AUD.LHR, E.17.ID.CR, E.17.EFP, A.6.3, A.6.3.CR, A.6.3.RT, A.10, A.2.8.PER, A.2.9, A.15, A.15.4, B.3, C.11, A.20, A.21.
Plain-name. Keep one publication unit explicit about its primary subject.
One-line summary. For one bounded publication unit at a time, PublicationUnit Primary-Subject Discipline makes clear what the unit is mainly about, what claim or communicative move it carries, and which wider work, downstream use, decision, or reliance claim remains outside.
Primary subject. In this pattern, publicationUnitPrimarySubject means what this bounded publication unit is mainly about for the current reading. It may be a named entity, boundary, episode, question, proposal, pattern section, or another plainly named subject. This is a publication aid, not a new U. kind or a C.2.1 participant by default.
Exact C.2.1 projection. Only when the unit carries one identified claim-bearing episteme E, and its primary subject is the entity that the claims of E concern, may the author state publicationUnitPrimarySubject = EntityOfConcern(E). Otherwise do not infer an EntityOfConcernRef from a topic label or interpretation alone, or use a primary-subject transition as evidence that the C.2.1 participant changed.
Publication unit. A bounded note, memo, sheet, review aid, screen, table or short section intended for a person or an agent to interpret as one unit. A machine-only operating note qualifies when its recipient needs to interpret the whole note.
Use this when. A note, memo, sheet, screen, table, comparison aid, or other publication unit can sound continuous even as it shifts what it is mainly about, which question it foregrounds, what it claims or asks the reader to do, or which wider process it appears to license. Use this pattern when local word repair is no longer enough and the unit needs stable answers about its subject, the move it makes, its permitted use, and the work that remains outside.
What goes wrong. A publication unit changes its subject, concern, claim, communicative move, or downstream use without marking the change. Reviewers argue over sentences while the reading of the unit as a whole shifts. Readers then give it decision weight or scope it never claimed.
Practical gain. A team can prevent an unmarked change of reading before a memo, note, or review unit comes to stand for rollout, approval, wider architecture strategy, or another concern beyond the unit without saying so. Reviewers can identify what needs to be stabilized sooner, keep downstream work outside the unit, and decide more quickly whether its interpretation is stable enough for continued use.
When to use another approach. Use another approach when:
- the problem is still local lexical-head kind or qualifier repair and
E.17.AUD.LHR(Local Head Restoration) is enough; - the same publication unit is already stable enough, and the question under repair is one bounded comparative review move over already available source epistemes or publications under
E.17.ID.CR; - the question under repair is still same-entity rewrite, representation shift, explanation-face work, bridge-explication, or another neighboring pattern whose move is already primary;
- the question under repair is view, face, carrier, or publication architecture rather than publication-unit interpretation instability;
- the unit is already being used to approve, assign, adjudicate, or direct work and should use the downstream decision, work, or reliance publication appropriate to that use.
Quick recovery. If this situation fits, write the ordinary natural-language declaration in E.17.AUD.OOTD:4.3 and compare it with the nearest worked slice in E.17.AUD.OOTD:5.1 through E.17.AUD.OOTD:5.6. Use the six diagnostic prompts only if it is hard to state the declaration accurately. For a request to clarify scope, one or two sentences can finish the work. If the unit owes its reader an assessment, include the criterion, relevant facts and resulting judgment.
What this pattern does. PublicationUnit Stability Discipline names the broader family. PublicationUnit Primary-Subject Discipline is the local writing-and-review pattern for making one unit's primary subject, carried move, downstream-use boundary, and outside-work boundary clear together.
Reader. The primary reader is an engineer-manager, architect, reviewer, or programme lead who needs to stop one publication unit from changing its subject without saying so.
Problem frame
Use the examples as alternatives. The quick checks, ordinary declaration, optional diagnostic, heavier extension, and worked slices are alternative aids for one publication unit, not a required sequence. One case may stop after the declaration; another may reopen when outside observations change the concern, claim, or downstream use the unit can support; another may use the applicable neighboring pattern once approval, work, reliance, or another question becomes primary.
Teams repeatedly write one publication unit that begins with one primary subject and ends with another subject, concern, carried move, or downstream use while still sounding like one unchanged text.
Typical moments include:
- an architecture note that starts about a system boundary and ends by directing rollout work;
- an operations review note that starts about an incident episode and ends as an action approval;
- a requirements or policy note that starts about an identified entity and ends about its carrier or document status;
- an episteme-publication-heavy note that starts about one pattern section or publication form and ends about wider architecture strategy;
- a comparison sheet that starts about one subject and shifts without warning into engineering-process, approval, work, or reliance pressure.
That interpretation instability is usually not caused by one bad sentence alone. It arises because the unit as a whole no longer gives a stable answer to what it is about, which concern it foregrounds, what move it carries, how readers may use it, and what wider work remains outside.
Problem
Without a named publication-unit discipline:
- authors repair one vague phrase at a time but still leave the unit unstable as a whole;
- reviewers argue about wording while missing that the unit has already shifted subject, concern, claim, communicative move, or downstream use;
- teams quietly read one note as if it licensed a downstream use the unit never declared;
- local lexical discipline (
A.6.P,E.10,F.18) gets blamed for publication-unit interpretation instability it was never meant to solve alone; - unit-form confusion is mistaken for view, face, carrier, or publication architecture even when the immediate problem is how to read the unit.
Forces
Solution - make the subject, move and scope of a publication unit clear
Manager-first entry
PublicationUnit Primary-Subject Disciplinekeeps one publication unit explicit about what it is mainly about, what claim or communicative move it carries, and what wider work remains outside.It becomes necessary when local repair is no longer enough and the publication unit still shifts among subject, concern, description, carrier, process, or downstream use while sounding unchanged.
First recover what the reader needed to understand, choose or do before this text was written. Ask where the property or relation in that question could arise. The reader may need an account of how a practice repeatedly produces a result, while the draft explains one participant’s choice. Trace what that choice establishes about the practice. Keep the incident as the subject when the request concerns that incident. When the wider question remains, recover the organization or rule that connects the pertinent cases and show the reader how to use it (C.39.QR:4.1, C.39.SA:4).
Examples of statements that clarify a unit’s scope include:
this memo is about the architecture boundary, not yet about the rollout plan;this review note is about the incident episode and the observed contrast, not yet a production-action recommendation;this comparison sheet is about the options under review, not yet about approval or the downstream decision;this semio note is about one pattern section or publication form, not the wider architecture policy around it.
If that is the clarification you need, start here.
If the problem is still confined to one vague local lexical head, start with E.17.AUD.LHR (Local Head Restoration).
Plain working terms
- Publication unit = a bounded piece of writing or display that the intended person or agent is meant to interpret as a whole, such as a note, memo, sheet, table or guided screen. A validator asked only to check syntax can finish with that result.
- Primary subject = what that bounded unit is mainly about for the current reading. It is an ordinary local publication term, not a new ontological kind.
- Concern = the question, aspect, or issue foregrounded about that subject. The concern can change while the subject remains the same.
EntityOfConcern= the oneU.Entityoccupying the entity-participant position in theC.2.1constitution of one identified claim-bearing episteme. Neither a topic label nor a writing question identifies that participant by itself.- Carried move = what the unit asserts, compares, explains, recommends, or otherwise communicates about its subject; it may also say that it only stabilizes the reading without adding a new claim.
- Downstream use = what a reader is invited or permitted to do with the unit, such as understand, compare, approve, rely, assign, or act.
- Outside-work boundary = what wider review, execution work, non-admissible downstream decision, or reliance claim stays outside the current unit.
- Explicit transition = the unit shows which subject, concern, move or use changes and why the new passage bears on the reader’s question. A move to one part, a model at another level of abstraction or a worked case may contribute a premise. Return from it to the conclusion its grounds support.
What can change
Treat the publication unit as the bounded material the reader is meant to interpret together. One primary subject and move suffice for a simple reading. Where a decision depends on linked sections, identify each section’s subject and move and show how their connection supports the judgment. Identifying the unit this way leaves the source episteme, its C.2.1 EntityOfConcern, the publication form, carrier and E.24.PUB occurrence under their own identity rules.
Keep five change types distinct:
- a subject change changes what the unit is mainly about;
- a concern change foregrounds another question or aspect while the subject may remain fixed;
- a claim or carried-move change changes what the unit asserts, compares, explains, or recommends;
- a downstream-use change changes what the reader is invited or allowed to do; and
- an
EntityOfConcernchange occurs only when the claim-bearing episteme being carried has changed in the entity participant that its claims concern underC.2.1.
Use the optional prompts in 4.3 only when these distinctions are hard to recover from the unit itself.
Once the carried episteme E is identified, the author may add the conditional projection publicationUnitPrimarySubject = EntityOfConcern(E), and only when both sides name the same entity. If the lens cannot stay stable after local repair, do not patch over the shift with a heavier declaration; reopen the unit or use the neighboring pattern that addresses the remaining question.
Scope and exclusions
In scope
- one publication unit with an unstable primary subject;
- one unit mixing concern, carried move, downstream use, and outside work;
- one unit quietly shifting between subject, description, carrier, publication unit, process, or downstream decision use;
- episteme-publication-heavy texts where repair disposition, the applicable boundary rule, primary subject, carried move, and outside work must stay explicit across one publication unit;
- a conditional
C.2.1projection when one carried episteme and its entity participant are already identified.
Out of scope
- local lexical-head repair only;
- pure view, face, or carrier architecture work;
- entityOfConcernRef-preserving transform, explanation, bridge, ontology, or comparative-review questions for which a neighboring pattern already supplies the needed method or test;
- downstream gate, approval, execution, or decision pressure;
- invention of a publication-wide
EntityOfConcernwhen no identified claim-bearing episteme supplies one.
Ordinary stop rule. If the commission is to clarify what a unit covers, a clear declaration can finish it. When the reader needs a judgment about the subject, state the judgment and the grounds it needs. A transition is required only when one occurred, and a neighboring-pattern reference only when a concrete unresolved question remains. A unit that has established one primary subject, one carried move, and one outside-work boundary needs no further assurance solely to establish them again. Ordinary use requires no diagnostic card, ClaimGraph, evidence dossier, assurance result, work record, or C.2.1 projection unless a separate receiving use independently needs one.
Choose the least-cost suitable unit architecture
“One primary subject” is a local default for a short unit meant to carry one readily recognizable move. It is not an ontological law and does not forbid a deliberately structured document that readers need as one unit.
Compare four repairs before deciding to split:
Choose the least-cost option that preserves the meaning of the claims and the reader’s intended use. When the decision turns on a relation between two subjects, keep the comparison and its grounds available together in one passage or in connected, marked sections. Where there is no truthful umbrella subject or joint use, treat the material as separate units or split it. Give connected subjects their own headings when that helps the reader follow the relation. Expected savings in reading effort can guide a design choice. To report what particular readers actually understood, use evidence from their reading.
Ordinary declaration and optional diagnostic
A declaration of scope is complete when it gives the reader the following information.
- the bounded publication unit;
- its primary subject or the subjects whose connection matters to this use;
- the claim or communicative move it carries; and
- the wider work or use that remains outside.
One or two sentences can state that scope.
This review note compares the interface-boundary options under the current incident evidence. Rollout responsibility and approval remain outside this note.
Do not require a separate card, record, identifier, table, or field set when that declaration is already clear. Add an explicit transition only when the unit changes subject, concern, carried move, or downstream use. Name a neighboring pattern only when one concrete unresolved question remains and that pattern supplies the needed instruction. Leave out an empty transition row or a search for a merely possible neighboring pattern.
When the sentence is hard to write or a reviewer suspects a hidden shift, use these six prompts privately as an optional diagnostic:
Use the six prompts to guide diagnosis. Once they yield a clear ordinary declaration, discard the diagnostic. The prompts need no six-row publication.
- If local repair is still enough, use
E.17.AUD.LHR(Local Head Restoration) without adding more structure to the unit. - If the unit remains one publication unit but a claim or downstream use depends on the neighboring-boundary claim kind, misuse risk, or cross-interpretation ambiguity, use the heavier extension as the assurance section.
- If the same unit is already stable as one primary subject, one carried move, and one outside-work boundary, and the remaining question is one bounded comparative review move over already available source epistemes or publications, apply
E.17.ID.CRwithout expanding the declaration. - If the unit cannot stay stable even after local repair, reopen the unit or apply the neighboring pattern that answers the remaining question. Adding fields to the declaration will not stabilize the unit.
Claim-bearing extension and quick boundary summary
Use the heavier extension only after the ordinary declaration is stable and a concrete neighboring claim or downstream use needs more detail. The extension cannot stabilize a unit that still shifts between primary subjects, carried moves, or outside-work boundaries.
Then add only the fields needed by the current claim or downstream use:
publicationUnitFormCue;primaryInterpretation;transitionPolicy;modelingLensPolicy;downstreamDecisionPolicy;entityOfConcernProjection, only for theC.2.1case stated in 4.1.b.
These fields supplement the ordinary declaration without establishing a separate set of rules. publicationUnitFormCue records words such as note, sheet, screen, and table as form clues only; it does not make those clues subjects, entity kinds, or claim kinds. entityOfConcernProjection records an already justified equality with the entity participant of one identified episteme; it neither creates that participant nor turns a topic into an entity. The remaining fields clarify the relevant boundary only when the ordinary declaration is not enough for a named later use.
Quick boundary to neighboring patterns and project records
- use
E.17.AUD.LHR(Local Head Restoration) when the instability is still local to one lexical head, qualifier, or interpretation word; - use
E.17.ID.CRwhen the same publication unit already holds one stable primary subject, one carried move, and one outside-work boundary, and the question under repair is one bounded comparative review move over already available source epistemes or publications; - use this pattern when the unit's unstable subject, concern, carried move, downstream use, or outside-work interpretation extends beyond local lexical repair;
- use the neighboring pattern that addresses the view, face, carrier, entityOfConcernRef-preserving transform, explanation, bridge, ontology, gate, approval, or execution question; keep any required project record with that question.
Boundary-rule summary
Use this summary to decide whether to stay with this pattern or move to a neighboring one.
- choose the least-cost suitable unit architecture under 4.2.a: retain one primary subject, declare a transition, keep an explicitly sectioned unit, or split;
- do not collapse primary subject, concern,
EntityOfConcern, description, carrier, publication unit, carried move, process, and downstream use into one unchanged interpretation; - keep the carried move and permitted downstream use distinct from the wider work around them;
- use
E.17.AUD.LHRwhen one local head or qualifier is the defect; use this pattern when that repair cannot settle the unit's subject, move, use, or outside boundary; - apply
E.17.ID.CRwhen publication-unit stability already holds and the remaining question is one bounded comparative review move over already available source epistemes or publications; - move out when the unit starts carrying downstream decision pressure or another neighboring-pattern question.
Archetypal grounding
The cases below show alternative outcomes in architecture, operations, episteme publication, a return to comparison, and a changed concern. They do not prescribe a progression.
Architecture note shifting into rollout work
A short architecture memo begins with:
This note is about the proposed service boundary between catalog and checkout.
Three paragraphs later it says:
We should therefore assign rollout responsibility to platform and stage migration in two sprints.
The memo moved from the interface to a proposal to assign and schedule rollout. Mark the transition if the reader needs both, or put independent rollout work in its own unit. Moving from describing to assessing the same interface need not change the C.2.1 EntityOfConcern of the episteme.
For an introduction, “This note introduces the service boundary between catalog and checkout. Rollout is outside its scope” can finish the job. For an assessment, keep the two sides together and give the finding. Suppose checkout must handle every agreed catalog response. The catalog can return “item unavailable” and checkout accepts only a priced item. The current interface therefore fails that requirement. One such response settles this failure. To establish compatibility for that requirement, show that checkout handles every agreed response under the stated conditions.
Either finding can guide an interface decision. Where a broader assessment also depends on an unknown deployment condition, identify what remains open. A rollout assignment requires the responsible authority’s action.
Operations note shifting into approval
An incident note begins as a comparative review of timing variance and operator context. It ends as if it already recommends a production action.
The incident may stay the primary subject while the move changes from comparison to recommendation. Mark the change if the reader needs both. To communicate an existing approval, cite its authoritative outcome and scope. When approval is needed and still pending, give the decision-maker the proposal and its grounds.
Use C.11 if the new text chooses among already available actions. If an actual approving communication and an instituted permission matter, keep the A.2.9 communicative Work and the A.2.8.PER grant relation separate. Use A.21 only when a current OperationalGate(profile) publishes a gate decision.
Semio-heavy text mixing one local section and wider architecture strategy
A semio note starts about one selected pattern section and ends as if it had decided the packaging strategy for the whole overlay.
Here the primary subject broadens from the selected section to the whole overlay, and the carried move broadens from local interpretation to strategy. The unit should state:
- what the note is about now;
- what concern and move it carries over that subject;
- and what wider architecture strategy remains outside the current unit.
If the commissioned question is the packaging strategy itself, inspect what the section establishes for it. For example, a distribution option that requires every section to stand alone can be ruled out by one section whose essential definition is available only in another. That finding can settle a request to assess this option. To design a workable distribution, establish which definitions readers need together and how each reader will obtain them when needed. The list of required terms from this one section can contribute to that design.
Unit stabilizes and bounded comparison becomes primary
A review note first shifts between the selected interface boundary, the move it is making over the current evidence, and the rollout implications around that boundary.
After one publication-unit repair it now says:
This review unit is about the interface-boundary options and the contrast they make visible under the current incident evidence; rollout responsibility and approval remain outside this note.
At that point the same unit already holds one stable primary subject, one carried comparison move, and one outside-work boundary.
PublicationUnit Primary-Subject Discipline has done its job.
If the remaining question is now one bounded comparison between the already pinned options over the same evidence, use E.17.ID.CR for that contrast. The unit needs no further stabilization for this purpose.
Outside observation changes the live concern or carried claim
A release-readiness note is already explicit that it is about one candidate publication or view and the risk state visible from the current evidence. Mid-review, an external vendor bulletin and a new field observation change the reported failure boundary for that same candidate.
The candidate may remain the primary subject. What changed first is the evidence-facing concern and the claim the note can support; a later approval or execution question may also change the downstream use. Do not report an EntityOfConcern change unless one identified claim-bearing episteme has a different entity participant under C.2.1.
Repair the note in one of three ways:
- stop the current unit at the originally declared evidence boundary and open a new downstream record for the changed question;
- explicitly reopen the same unit with the revised concern, claim or carried move, permitted use, and outside-work boundary;
- or use the downstream decision or work pattern whose instructions now fit once approval, execution, or another downstream decision publication becomes the primary question.
The bulletin and field observation remain sources until a support, evidence, or currentness claim makes A.10 relevant. Use C.11 for a later choice, A.2.9 and A.2.8.PER for an approving act and its permission effect, A.15 for a work claim, and A.21 only for an actual gate decision.
Deliberately sectioned multi-subject review packet
A release-readiness group needs one packet for one meeting. The packet contains three sections with distinct headings:
- Interface-boundary options — compares two architecture alternatives.
- Incident evidence — summarizes the observations that discriminate between those alternatives.
- Rollout constraints — states constraints the later approval decision must respect, without assigning work or granting approval.
The local one-subject heuristic does not force three documents. The packet has one umbrella subject that fits all three sections—the evidence and constraints needed to review the interface-boundary choice—and one bounded use: inform the review, not approve rollout. Keeping the three section-level subjects together avoids navigation and synchronization cost, and the headings distinguish each section’s move from the others.
An unsectioned version is rejected because readers cannot see the subject and move changes. A short narrative with only one small shift may instead declare that transition. Separate documents become the least-cost choice when the rollout section starts assigning responsibility, serves another audience, needs independent reuse, or becomes an approval input with its own reliance boundary.
Bias-Annotation
This pattern favors making the unit’s reading stable and visible. It discourages letting the unit take on wider work or decision pressure without marking that change.
To counter that bias, name the primary subject, concern, carried move, downstream use, and outside work. Use EntityOfConcern only when C.2.1 warrants that identification. Return promptly to E.17.ID.CR when the unit is stable and a bounded comparison remains, and state which boundary applies once a downstream claim becomes primary.
Conformance Checklist
Checklist scope. Use this checklist when checking a claimed application of this pattern, not as nine required authoring steps. A short declaration can settle scope. When the unit is commissioned to assess its subject, check that the reader also gets the judgment and its grounds. Inspect only the rows implicated by the unit, transition, EntityOfConcern projection, neighboring claim, or unit-architecture choice, and do not publish a nine-row record by default.
- CC-OOTD-1 - One publication unit is explicit. The text identifies the publication unit under review as one note, memo, sheet, screen, table, or section meant to be read as one unit.
- CC-OOTD-2 - Primary subject is explicit. The unit states what it is mainly about in ordinary language rather than asking readers to infer it from tone.
- CC-OOTD-3 - Any
EntityOfConcernprojection is exact and conditional. The unit usesEntityOfConcernonly for the entity participant of one identified claim-bearing episteme underC.2.1. A topic label alone does not identify that participant. An independently identified entity, including an admitted relation kind, may do so when the claims concern it. - CC-OOTD-4 - Concern, carried move, downstream use, and outside work are distinct. The unit states which question it foregrounds, what it asserts or communicates, how readers may use it, and which wider work, approval, execution, decision, or reliance remains outside.
- CC-OOTD-5 - Any transition is typed and explicit. For a change of subject, concern, carried move, downstream use or entity participant, the reader can tell why the passage shifts and what conclusion returns to the original question. A wider conclusion needs grounds across its stated range.
- CC-OOTD-6 - Choose between local and publication-unit repair.
Use
E.17.AUD.LHRwhen one local head or qualifier is the defect. Apply this pattern when that repair cannot settle the unit's subject, move, use, or outside boundary. - CC-OOTD-7 - Neighboring-pattern boundary is explicit. If an entityOfConcernRef-preserving transform, explanation, bridge, comparative-review, ontology, gate, approval, or execution claim becomes primary, use the neighboring pattern that defines or constrains that claim.
- CC-OOTD-8 - Claim-bearing lens is stated when needed.
If a claim or downstream use materially depends on a minimal modeling lens,
C.2.1projection, or downstream-decision policy, state that lens, projection, or policy. - CC-OOTD-9 - Choose the least-cost suitable unit architecture. Retaining one unit, declaring a transition, keeping a sectioned multi-subject unit, or splitting is chosen from the current reader, use, reuse, dependency, and overread costs. The author does not split to satisfy a count and does not retain a vague umbrella to avoid a necessary split.
Common Anti-Patterns
- Local-repair inflation. Opening publication-unit discipline when one overloaded local lexical head or qualifier is the defect that remains.
EntityOfConcerninflation. Calling every topic, kind, interpretation, question, or writing transition anEntityOfConcernorEntityOfConcernchange.- Work-process smuggling. Letting a note begin as architecture, incident review, or comparison work and end as rollout, approval, or execution guidance without naming the transition.
- Admissibility-pattern replacement. Treating this pattern as if it replaced view, face, or carrier architecture, entityOfConcernRef-preserving transform rules, explanation-face rules, bridge rules, or downstream decision texts.
- Overgrowth by declaration. Stacking fields onto a unit while readers still cannot tell which subjects and moves it connects or what remains outside.
Consequences
Using the pattern well has three main benefits:
- authors stop smuggling wider work into one unit by accident;
- reviewers can name whether subject, concern, carried move, or downstream use changed instead of only arguing about wording;
- neighboring patterns and downstream decision texts stop getting blamed for confusion created one layer earlier.
The cost is that some notes must become shorter, split earlier, or reopen with the change stated when their subject, concern, carried move, or downstream use changes.
Rationale
This pattern leaves the architecture of views, faces, carriers, epistemes, and downstream decision texts to their respective methods. It addresses a smaller problem. A publication unit can mislead even when each sentence appears acceptable on its own.
A.6.P, A.7, E.10, and F.18 already keep kinds, distinctions, and naming precise. C.2.1 already identifies the entity participant of one claim-bearing episteme. This pattern adds a discipline for the bounded publication unit: choose the least-cost suitable architecture for that unit, make its primary subject or section-level subjects and moves visible, and keep downstream use and outside work explicit. It borrows EntityOfConcern only through the conditional projection in 4.1.b and does not extend that ontology.
The pattern also draws on E.14 and E.19.
Recognition comes first through a manager-usable entry block and one ordinary natural-language declaration; the six prompts remain an optional diagnostic.
Heavier declaration comes only after the ordinary declaration already holds and a named receiving use consumes the added fields.
SoTA-Echoing
Source boundary. These sources support topic focus, scope/non-scope, reader-need organization, and explicit document structure. None establishes a universal ontological rule that every publication unit has one subject, and none supplies a C.2.1 entity participant. OOTD therefore keeps the one-primary-subject rule as a defeasible local heuristic and compares it with transition, sectioning, and splitting.
Relations
Builds on
A.6.PA.7E.10F.18E.14E.19C.2.2aA.16.0
Nearest neighbors
E.17.AUD.LHRfor local lexical-head kind or qualifier repair;E.17.ID.CRwhen the same unit is already stable and the remaining question is one bounded comparative review move;E.17.EFPwhen explanation-face use or faithfulness on existing faces is primary;A.6.3,A.6.3.CR, andA.6.3.RTwhen the question under repair is same-entity rewrite or representation change;A.10when evidence or provenance becomes primary;A.15andA.15.4when work, reliance, or execution claim becomes primary;B.3when assurance or engineering justification becomes primary;C.11when choosing among already available options becomes primary;A.2.9when an actual approval is communicative Work, andA.2.8.PERwhen the question is the permission or grant relation it institutes, its exercise, or its conflict;A.20only when step-localFlowConstraintValiditybecomes primary, andA.21only when a currentOperationalGate(profile)publishes the gate decision.
E.17.AUD.OOTD:End
Transformation Flow Structure
Tech-name: TransformationFlowStructure (pattern label) Plain-name: Transformation flow structure Type: Structural pattern for ontic relations (E) Status: Stable Normativity: Normative unless marked informative
Intent
E.18 helps a practitioner understand the organization of a situation in which a result is produced or used. Ask which independently identified things participate, which relations connect them in this case, which values the use in question assigns, and which dependent decisions must be reconsidered after a change. A TransformationFlowStructure is one selected A.22 structure under the flow-use conditions in this pattern. Its constituents and relations retain their own kinds and identity rules.
Use E.18 when a decision needs that selected organization, several valuations of it, a revisitable internal portion, or a path, crossing or refresh scope whose connection to the real participants matters. Begin by naming the practical question and the actual source and receiving use. A readable note can be sufficient: “This check used the result of that measurement and this calibration; reconsider both dependent checks and their joint decision when the calibration changes.” Use A.6.1 to establish an actual operation binding and A.21 for a gate decision when those are the questions. A note need not first become a TFS or a complete record.
First useful structure result. When the organization itself is needed, identify its constituents and obtaining relation occurrences, the constraints applied in selecting them, and the use question with its action and stop condition. Then locate the participant or connection the current question uses. Section 5.1 shows how to resolve the position. Add a valuation and path under 5.2 only when needed. The result lets the practitioner distinguish another view or valuation of the same organization from a changed selected occurrence or another independently identified structure.
Several valuations remain in E.18 only while the same A.22 identity is retained. A detailed portion remains a parent-relative SubflowRef while its positions and included relation occurrences resolve inside that parent. Independently identified TFS values connected across their boundaries use E.18.NET.
Actual Work, Transformation, production, source use and physical transfer each require the predicate and facts supplied by the pattern or declaration that defines them; membership in the organization does not establish them.
GateCrossings account for changed CtxState bindings at an OperationalGate. Publication, comparison and refresh retain the pins and conditions consumed by their actual use. An F.9 Bridge is added for an independently established correspondence between distinct recovered local senses. E.18.2 describes the selected subject mathematically; C.29 governs a claim that the mathematical lens is adequate for the proposed use. Neither the drawing nor its path creates the represented relations.
Structure ontology. E.18 keeps these distinctions primary:
Result-claim assurance. Use this fuller account only after the Plain test above identifies what the value is a result of or for. The direct basis must be exactly one of these three, according to its category:
- an obtaining relation occurrence, with its predicate and occurrence-identity rule plus the participants of that occurrence, applicability, and case facts;
- an
A.6.1operation-application binding, with operation, application, and argument or result binding; or - an
A.6.RCDlocalC.2.1claim, with polarity, substrate or constructor, base predicates and the patterns or declarations that define them, participants, case facts, and any support required by the receiving use.
When a sentence says that a system performs an actual functional transformation at one point in a flow, E.18 carries only the selected flow structure, locus, path, slice, crossing, valuation, and pins. The independently identified bounded transformation, transformer or candidate bearer, affected referent, input and output boundary, functional-port boundary, functioning relation, method or algorithm, mechanism, and performed work are recovered through A.3.4, A.6.F, C.30.ASV, A.6.M, A.6.1, and the A.15 family as applicable.
A desired state, method, MethodDescription, WorkPlan, architecture selection, model, description, evaluation result, publication, or transfer does not ground the actual transformation.
When dated work is claimed to cause or realize the change, use the current predicate and case facts under A.6.RCD disposition 1, or—only when no direct predicate expresses the compound claim and admitted base-predicate semantics support it—one local C.2.1 claim under disposition 2. Keep the Work, Transformation, and claim separate.
When production-work participation, entity-identity inception, or production completion is claimed, cite the separate local A.15.PROD claim; E.18 does not derive it from structure membership. A computational algorithm may fill MethodRef? or MethodDescriptionRef?; a physical-world way of transforming may fill U.Method; neither is inferred from E.18 structure membership.
Not this pattern when. Use A.20 for internal step validity, A.21 for gate decisions, E.20 for mechanism-governing-definition placement, A.3.4 for bounded transformation under conditions, E.18.2 for mathematical descriptions of the selected structure, C.27.TA for temporal aspects, C.27 for temporal-claim adequacy or supported-use claims, the A.15 family for work planning, performed work, or work-entry readiness (A.15.5), E.17 for publication faces, and E.10 for wording-use repair when the current EntityOfConcern is not the selected structure, path, crossing, or flow valuation.
What goes wrong. A practitioner may treat a reference flow, a wording-use cue such as transition, or a tool pipeline as a new graph kind or a hidden prescribed procedure, then lose comparability and crossing evidence and lose track of which uses the change reopens and which it leaves unaffected.
Practical gain. E.18 keeps selected structure, publication pins, crossings, the separation of internal constraint results from GateFit results, and refresh locality in one structure pattern. This avoids turning every path into its own flow doctrine or every mathematical graph description into the selected structure.
Problem frame
One selected TransformationFlowStructure can carry many well-typed flow valuations only while every valuation resolves to that same structure, its identified positions, and the same selected relation occurrences.
Under one function-oriented viewpoint P selected through a U.ViewpointRef, those valuations may concern transformations of one already identified target holon, for example in a qualified holder-ability claim under A.2.2 or a transformation claim; VP.Functional, when used, is only P's ordinary designator. That target remains distinct from the selected structure and does not become a context object merely because an engineering description concerns it; the E.18 EntityOfConcern is the selected structure over transformations and adjacent identified positions.
E.18.1 P2W Problem-to-Work Carry-Through begins with an accepted ProblemCard@Context claim and carries it into whichever method, plan, dated Work, transformation, evaluation, decision, entity, relation occurrence, interpretation, stop, branch, or local return becomes current. For each continuation, state the question now at issue and apply the pattern whose Solution answers it. Before calling one of those values a result, say what it is a result of or for and cite the fact, relation, or binding that makes that reading true; otherwise stop.
Apply the adjacent result-claim assurance check only when a named reliance use needs it, and never mistake the flow position for assurance.
A first-principles specialization may traverse a path such as U.Signature(profile=FormalSubstrate) -> U.PrincipleFrame -> U.Mechanism -> U.ContextNormalization (UNM) -> selector relation -> one U.WorkPlan, optionally with declaration-local A.15.3 planned-filling rows -> one Work occurrence admitted under U.Work -> evaluation or currentness relation. That is one possible transformation-flow path, not the definition or prescribed order of P2W: a P2W use may skip, branch, split, stop, return, or reopen.
Without a common structure discipline:
- flows look ad-hoc and non-comparable;
- structural crossings fail to name the changed state binding, its from/to values and establishing basis, any applicable rule and current application, or the separate gate decision;
- MVPK faces carry hidden arithmetic or present their own version of the governing input and output declarations;
- set‑returning selection is silently replaced by single scores;
- cycles run without a budget and refresh happens out of band.
MVPK corrects publication drift at the single-arrow scope; E.18 applies those publication and comparability rules to the selected transformation-flow structure as a whole.
Problem
- Confusing a mathematical description with the selected structure. A catalog of morphism-scoped, transformation-scoped, mechanism-scoped, work-scoped, or refresh-scoped patterns does not, by itself, explain how the whole selected structure is built, constrained, and audited.
- Flow proliferation. Multiple “reference flows” can be declared; practitioners need one structure discipline that keeps their flow relations typed and comparable without privileging any single flow.
- Unsafe publication. A face can silently change a signature's terms, hide scalarization, omit required edition or plane pins, or present a Bridge Card,
CLvalue, UTS row or policy id as evidence that a GateCrossing or gate decision is current. - Cycles without norms. Selection↔Planning loops run without an explicit budget (Γ_time), an identified stale-measurement finding and any separately identified refresh plan it triggers, or slice-scoped refresh. A pre-run gate decision is mistaken for actual launch bindings, or a
FinalizeLaunchValuesrecord is written before a Work occurrence and its independently obtaining bindings exist.
Forces
Solution - Transformation-flow structure model and relation disciplines
Ordinary uses. Five uses are central in E.18: name one selected transformation-flow structure; distinguish the selected structure from a flow valuation and from its mathematical descriptions; place gates only on crossings or on a pre-run work-entry claim; preserve the comparison’s required common basis, normalize when needed, and retain declared set returns; and keep cycles under budget plus PathSlice refresh.
A gate may authorize or block an intended entry, but it neither creates a future Work occurrence nor fills fields in one. Use S12 when engineering or publication work calls for a viewpoint mapping.
S1 - Selected Structure (conceptual)
Start with the result the recipient must obtain or explain. Trace each needed contribution to the participant that makes or uses it. If the scope is open, use C.39.SA to select the parts and connections the use requires. Follow them only as far as a different arrangement could change the answer or its grounds. For example, a Method can call for a check, a technician can perform it during a job, and a report can state its outcome. To explain what happened, establish which performance produced the outcome used in this case. To design a repeatable arrangement, specify how the required check will be provided under the allowed conditions. A restriction shared by two contributions belongs in the account whenever their separate descriptions leave their joint use undecided.
Select S under A.22 by four discriminators:
- O: the independently identified constituents;
- R: the selected obtaining relation occurrences, with their direct predicates, participant meanings and occurrence-identity rules;
- C: the semantic constraint claims applied in selecting the organization; and
- F: the named use question, admissible action and stop or return condition.
This selection identifies an organization; it does not create its participants or make a proposed relation obtain. Include the complete participants of each selected occurrence, including participants that a compact drawing leaves off its main line.
An occurrence may itself be a constituent participating in another relation. An exact scope or budget contributes to S through C when it is an applied selection constraint; a record, diagram, selecting system or valuation is not an extra identity discriminator.
A.6.1:4.5 supplies actual application and declaration-local argument/result bindings. Use the subject's direct rules for physical transfer, Work, actual Transformation, production, causation, source use or other participation. A local compound claim can state what follows from established base relations without becoming an additional relation occurrence in R.
If the required governor is absent, return missing-governor; if a deciding fact is unavailable, return missing-information; if sufficient facts fail the positive test, return factually unsupported. A negative conclusion requires its own applicable non-obtaining criterion. A proposed arrangement can remain useful as a description or model while the stronger selected-structure claim is unresolved.
Locate the selected participants and relations. Let Pos_S be the set of participant positions, with beta(p) resolving each position to its constituent x in O. A localFlowPositionId designates that position within S and must not silently be rebound to another constituent. Several visible copies may designate the same position. Participant roles are distinguished by their incidences.
Let Conn_S contain exactly one connection position q = <S, r> for each selected occurrence r in R. The resolution rho(q) = r is one-to-one. An incidence <q, participantDesignator> resolves to the position of that participant of r, preserving the role meaning, required order and repeated participation of the same object under different meanings. Where the predicate uses an unordered participant distinction, retain that distinction; a display order does not add semantics. Even if a drawing shows three lines for one ternary relation, it represents one q and one r.
Pos_S, Conn_S and incidences are local addresses in a representation of the selected organization, not new FPF kinds or additional subject relations. In the complete incidence representation, use a participant node for each position and a relation node for each q. If r is itself in O, its participant position and its relation node resolve to the same occurrence and are identified in the path representation. This connection lets a path pass from an operation to its result-binding occurrence and on to an operation that actually uses that occurrence. It creates no extra transfer relation.
Keep three bindings distinct: beta resolves a structural position; an A.6.1 operation-application binding establishes actual participation or return; a CtxState binding assigns the independently established values for one position-use. A stored reference or compatible type does not establish the actual operation binding.
S remains the same only while O/R/C/F remain the same. A new valuation, alias, path, description or publication can preserve that identity. A new actual application or selected relation occurrence must first pass its own identity rule and then A.22; calling it another valuation does not preserve S. A changed use annotation or policy does not change C unless its content is applied in selecting the structure.
Scope and use. Gamma_time supplies the budget or horizon used by the relevant application, PublicationScope constrains the selected publication use under E.17/A.2.6, and PathSlice identifies portions used in telemetry and refresh. Section 5.13 defines how all dependent portions are covered.
CtxState (bindings for one use; closed slots): CtxState = ⟨L, P, E⃗, D⟩ contains independently established bindings for one position-use. P here denotes the plane/units declaration; Pos_S above denotes participant positions. PublicationScope constrains applicability and publication of that use; it does not choose these bindings. A scope containing two slices chooses neither as L, and even a singleton scope supplies no unique plane, edition vector or design/run tag. Resolve each binding under the rules below, then check the applicable A.2.6/E.17 membership and publication conditions.
Slot definitions and changed-binding account boundary (normative):
L := Locus— oneU.ContextSlicevalue identified underA.2.6; any scope-membership or translated-scope claim remains with A.2.6 and its current F.9/C.2.1/A.10-or-B.3 premises when semantic translation is required.P := ReferencePlane— a ref-only binding to the plane and units declaration used by the current case. E.18 supplies no generic plane conversion. Cite the current declaration and applicable conversion rule by value. Returnmissing-governoronly when no current conversion predicate or rule can state the attempted crossing; returnmissing-informationwhen the needed declaration or case values are unavailable; when the rule and facts are current, state its positive, negative, or inapplicable result rather than a generic blocker.E⃗ := Edition vector— a partial mapedition_key ↦ EditionIdwhose members cite each versioned value, its exact edition, and the registry or declaration that assigns that edition;G.11defines the edition-bump and refresh records, whileE.17defines publication of the refs.D := DesignRunTag—design(T^D)orrun(T^R)only as consumed by theA.21gate and, at work entry, theA.15.5readiness claim; the tag does not identify or create Work.
Invariants. A raw connection step preserves CtxState (⟨L,P,E⃗,D⟩). A changed use binding has an explicit from/to account and the basis required for that change. When an applicable A.21 profile governs the transition, represent it as a GateCrossing at OperationalGate(profile). Retain every required check and action consequence; the gate decision remains separate from the facts identifying either binding, a Work occurrence or an obtaining relation.
Extension discipline. A conforming use registers any extra slot beyond ⟨L,P,E⃗,D⟩ in the E.17 publication discipline and the E.18 LEX “CtxState Extension Registry” with slot‑id, intent, partial‑order rule (neutral or absorbing), and SquareLaw compatibility; unregistered extensions are non‑conformant.
Data-shape location. E.18 defines the connection/cut valuation domain, PathId, PathSliceId, Gamma pins and crossing references. A.20 supplies current internal-constraint claims, G.6 evidence-provenance path visibility and G.11 the currentness and dependency rules used by 5.13. Their references retain their own subject meanings.
Locus markers. Transformation, Signature, Mechanism, WorkPlanning, Work, Check and StructuralReinterpretation name common uses of independently identified values. They are not a closed taxonomy of permitted constituents. An additional position states its direct kind, definition and identity basis; an actual application, binding or result value uses A.6.1 and its own declaration. The following common bindings use the definitions of their respective patterns. E.18 introduces no local types for them:
-
Transformation≡ A.3.4U.Transformationonly when the structure locus binds one independently identified actual bounded change with its exact changed referent, extent or ordering boundary, boundary conditions, actual change facts, and continuity or reidentification rule. Desired, intended, planned, modeled, selected, described, evaluated, published, or transferred change content remains under the definition or test for that exact claim; it is not aTransformationbinding merely because it occupies the selected structure.At the current resolution, identifying the change establishes neither finer parts nor partlessness. A positive transformation-composition,
TransformationPartOfRelation, composite-transformation identity, or transformation-holonhood claim stops under D14.16 with the exact A.6.RCD result:TC-MWH missing-governoronly when no current predicate, applicability condition, or occurrence rule states the required contribution, compatibility, parthood, or whole-identity claim;TC-MWH factually unsupportedwhen the governor exists and the available case basis is sufficient to apply its positive test but that test fails; andTC-MWH missing-informationwhen a fact needed to decide the test is unavailable.A negative conclusion needs its own applicable non-obtaining criterion or complete closure basis and satisfying facts. E.18 retains the independently identified transformations and supplies no provisional contribution, compatibility, parthood, or whole-change architecture; it does not preselect whether a later settlement uses a generic derived relation, subject-specific relations, local compound claims, or non-admission.
-
Signature≡ A.6.0U.Signature(universal, law-governed declaration). -
Mechanism≡ A.6.1U.Mechanism(law-governed application over a SubjectKind and RangedValueKind), with placement and stabilization relations inE.20when current. -
WorkPlanning≡ one A.15.2U.WorkPlanwhen that plan occupies the structure position. Declaration-local A.15.3SlotFillingsPlanItemrows remain content inside that WorkPlan and do not occupy a locus or identify a relation independently. -
Work≡ a dated Work individual admitted under A.15.1U.Work. A structure locus may point to that occurrence after it exists; before execution it points only to aU.WorkPlan, A.15.5 readiness relation, or another identified work-entry claim. No second enactment kind is introduced. -
Check≡OperationalGate(profile)when a gate/check locus is present. A.20 supplies exact internal-constraint results when those constraints are current; A.21 defines the gate profile, independent check retention, result mapping, aggregate decision, and publication minima when a gate decision is current. -
StructuralReinterpretationis only the E.18 position of an independently identified A.6.4 arrow r, bounded-use assertion q, and current-case judgement; it is not a new retargeting kind. E.18 records r and q, the case basis and judgement result needed by this placement, and path-slice locality. q's ClaimGraph carries the invariant, visible loss, named receiving use, conditions, and affirmative or negative polarity; the judgement separately reportssatisfies,fails, orcannot decide. Acannot decideresult names the missing fact and reopen condition. F.9 is additional only when the same case asserts a semantic relation between two exact F.17 local senses and its predicate obtains; its bounded-use claim, optionalCL, evidence, and reliance remain separate.
OperationalGate is the E.18 check locus when a gate or check position is present. A.20 supplies an internal-constraint result when the claim about that constraint is at issue. When a gate decision is current, A.21 supplies the profile application, independently identified check-application results, GateDecisionResult, and rationale. A DecisionLog is added only for a current audit, history, replay, or reuse need.
For structural reinterpretation, E.18 adds a structure-local placement rule: retain the independently identified A.6.4 arrow r, affirmative q, current-case judgement of satisfies, their basis and PathSliceId. With unchanged CtxState, this placement is not a GateCrossing. For a changed binding, state its from/to values, establishing basis and any applicable declaration, rule and current application; add a GateCrossing when an applicable gate profile governs that transition. A Bridge, card, UTS row, optional CL, witness publication, gate decision or permission claim supplies neither r, q’s polarity nor the case judgement.
MVPK integration (import). Every locus with an external publication face is published via MVPK faces (
PlainView,TechCard,AssuranceLane,InteropCard) under a declared PublicationScope (E.17). E.18 reuses MVPK's source-reference, pin, declared-order, and no-new-claim rules. The optional morphism profile keeps the signature authoritative and checks the claimed publication composition under E.17:6. E.18 adds the structure-scope constraints in S3 and CC-E18-09 and CC-E18-10; it does not define a second publication semantics.
GateCrossing (normative)
Definition. A GateCrossing is E.18's structure-local transition from an identified source <FlowPositionRef, CtxState> binding to an identified receiving binding at one identified OperationalGate(profile). It is selected when at least one CtxState binding changes and an applicable gate profile governs that transition. It is not a U.Relation, an F.9 Bridge, a gate decision, a plane conversion, an A.6.4 arrow or use assertion, a penalty, or a publication occurrence.
Per-binding account. For an ordinary binding change, one sentence or table row is enough: name the changed binding, its from and to values, the facts or claims that establish those values for this case, and any declaration or rule whose application is current. When a named downstream use needs replay, the same distinctions may be packaged in this local E.18 block:
Facts or claims establish the case values. A declaration or rule supplies only the meaning, admissibility condition, or constraint it states; ruleApplicationRefs is present only when the current case depends on that rule applying to these values.
A gate decision evaluates the crossing under A.21 and does not establish the underlying facts or apply a rule by itself. A permission claim is separate under A.2.8.PER and is cited only when authorization is current. None of those items entails another.
For example, an analytical use can change its unit binding from metres to centimetres using the applicable conversion and the actual source values. Retain those from/to values and the conversion basis. A gate is needed only if a current profile governs that change. A regulated calibration change before launch retains its applicable gate, required checks and any separate authorization; an unavailable case value remains missing information. If no governing rule states the conversion, return missing-governor.
A.20 may supply an exact current constraint-validity result and witness or reason; A.21 supplies the gate profile, retained check results, mapping, aggregate decision, and decision log. Neither supplies a changed locality, plane, edition, tag, retargeting fact, rule application, or permission claim.
Canonical reference. CrossingRef := ⟨TFSRef, GateId, FromPositionRef, ToPositionRef, FromCtxStateRef, ToCtxStateRef, ChangedBindingIds, PathSliceId⟩. A DecisionLog or downstream use that depends on the crossing cites this ref and the required per-binding accounts.
CrossingBundle publication block. Materialize a CrossingBundle only when a named selector, acceptance, audit, replay, or other downstream use relies on durable crossing evidence. The bundle is publication packaging under E.17, not a constituent of the crossing or gate decision.
The bundle contains the CrossingRef, ChangedBindingAccountRefs[], GateId, the current profileApplicationRef and GateDecisionResultRef when a gate decision exists, an optional current DecisionLogRef, optional separately current PermissionClaimEpistemeRefs[], PublicationScopeId, PathSliceId, and any current witness refs.
When that downstream use also relies on cross-semantic correspondence, add a separate F.9 block: the two SchemeSenseCell endpoints, the obtaining Bridge and its profile, the C.2.1 claim that says whether the Bridge suits this named structural use in the named direction under its rule and tolerance, and the current A.10 or B.3 reliance branch if reliance is claimed. A Bridge Card remains optional packaging and CL remains optional evidence shorthand; neither makes the structural crossing obtain, makes the gate pass, or grants the use.
A penalty appears only when one current policy applies to this crossing and its rule application to the crossing facts supports that penalty. Cite the policy and PolicyIdRef; when the claim also depends on who may issue or enforce it, cite the separately obtaining direct authority relation and its actual participants.
E.18 derives no penalty from CL, plane difference, edition difference, or Bridge publication. If the policy, applicability, rule application, or any separately required authority fact is absent, make no penalty claim and infer no default.
Term separation. An internal connection is the position q of a selected relation occurrence. A subject-specific transfer has its own predicate and episode rule when that claim is needed. Here transport refers to the named Phi conversion policies and registries (TransportRegistry^Phi under UNM); citing them does not establish an additional subject relation.
S2 - Flows as valuations, paths and admissible uses
A flow valuation assigns declared values to connection positions and named cuts of one selected S. Its exact domain is:
For each cut k, name S, the path family being cut, its entry and exit boundaries, and the crossing incidences with their boundary sides. Retain every full q to which an incidence belongs.
The cut must intersect every entry-to-exit path of that family. An empty family permits an empty cut; an incidence cut cannot interrupt a zero-length path between coincident boundaries. Minimality is required only when the declared use asks for it. These are cuts of a representation, not additional world-side relations or wholes.
nu is a typed partial mapping from this tagged domain to declared value channels under a named regime. State each consumed channel's ValueKind, units and Scale when applicable, any observation or assessment window required by that channel or receiving use, and interpretation basis.
Distinguish actual observation, model estimate and counterfactual. A returned value, a rate, an uncertainty assessment and a source reference need different channel meanings. An unknown value belongs only to a result domain that declares it; unresolved input, out-of-range value, inapplicability, notRun and error keep their different conditions.
A value on connection(q) concerns rho(q). It can report the value returned by a result binding or assess the present suitability of relying on that binding. It does not replace any participant of the occurrence. A counterfactual 42 assigned for a model calculation does not change a historical return of 40. A new assessment can be another valuation of the same S; replacing a selected actual result binding reopens R and S.
A cut value has its own rule. A role-indexed tuple may preserve all incident values and their multiplicity. A sum, minimum, probability or total physical flow requires its applicable aggregation law, including shared participants, dependence, duplicate counting and the empty case. Several drawn arms of one q are not several independent inputs.
Construct a path. A path is a finite sequence of compatible oriented incidence steps through the complete representation in 5.1. Use the direct predicate or use rule for a dependency direction. At every visited q retain the full occurrence and all required participants, including side inputs. A relation used as a participant retains the node identification stated in 5.1. Repeating a visit to q records another visit to the same place, not another actual episode.
Mathematical path composition concatenates sequences with matching endpoints and compatible interface designations. The empty path is an identity and concatenation is associative. These laws do not establish causal transitivity, actual Work order or composition of Transformations. E.18.2 may use another formalism with an explicit mapping and its own laws.
Decide whether the path is admissible for this use. Resolve and type the complete path. Check all required n-ary inputs, compatible channels, regimes, units and source editions, the applicable comparison and aggregation conditions, and the budget or cycle rule.
The conditions of a concatenation must be checked even if its two paths were individually admissible. A join that requires A and B cannot be enabled by a path from A with B omitted. Their applicable joint rule, rather than two binary arrows, determines the continuation.
For a joint performance, check the shared capacity, commitments and conditions at the times the connected actions need them. A sufficient upper bound on their joint demand can establish compliance with a limit. To promise a particular configuration, establish one compatible assignment of modes, resources and times that the intended performers can use. Retain common evidence as a shared premise wherever two branches depend on it. E.18.2 gives a worked case where combining the separate descriptions preserves a limit and introduces an unattainable mode.
For each relevant position-use, recover CtxState sigma = <L, P, Evec, D> under 5.1. A raw connection step preserves these use bindings; it does not assert equality of the physical states of all participants. Account explicitly for each changed binding. Where an applicable profile governs the transition, represent it at one OperationalGate by its GateCrossing. Retain SquareLaw evidence when the crossing rule requires it, Gamma pins on compare and launch faces, and any applicable LaunchGate condition for design/run changes. StructuralReinterpretation retains the separately established r, q and current-case judgement under CC-E18-06-EX.
A FlowValuation use therefore names S, the declared domain and value channels, nu and its regime, the relevant position-use bindings and any admissible path used by the claim. PublicationScopeId and PathSliceId are included when publication or slice-addressed use requires them.
A PathSlice is a selected portion of a path; it does not become performed Work or an execution interval. Published slice uses pin PathSliceId; a changed relied-on premise or applicable sentinel trigger reopens the affected uses under 5.13.
Re-emit faces whose current claim or required publication status changes; retain supported historical and unaffected uses. Section 5.13 extends the set of addressed slices to cover the complete actual dependency closure.
Several valuations, a subflow or a network. State, regime, path, slice and local tag differences alone do not reidentify S. Verify its four A.22 discriminators before treating nominal-load and emergency-load values as valuations of one organization. A parent-relative internal portion uses 5.2b. When separately identified TFS members and an obtaining cross-boundary relation are needed, preserve their identities and use E.18.NET; a connection position q is not a universal cross-member relation.
What “flow = valuation” contributes. This Eulerian account assigns a quantity or object to a declared place in an identified organization under a given regime. Revaluation makes changed assessments and values comparable without implying that the description creates a new actual event.
When the question instead depends on a particular delivery, receipt, displacement, possession or causal episode, retain the domain's complete transfer predicate, participants, applicability, extent and recurrence rule in R. Equal payloads can belong to different episodes; two descriptions can concern one episode. For an operation's actual use of an earlier result, the A.6.1 bindings can already provide the needed distinction. The arrow style establishes neither a transfer episode nor an operation’s use of a result.
Split-and-join structure discipline
Use split and join only as selected-structure relations inside one TransformationFlowStructure. A split separates one source locus, variant set, problem-side cue, or candidate family into several identified loci or flow valuations. A join relates several identified loci, selected sets, gates, measurements, or refresh returns back to one current structure position. Neither operation creates a new FPF kind, a new pattern, or a prescribed work procedure.
Minimum split-and-join use names the selected TransformationFlowStructure, the actual split or join predicate and its complete participants, any applicable membership-change policy, the selector-facing outcome when selection is at issue and the actual result binding when a selector has returned a value, the selected-set result declaration when current, the publication relation when that value is published, and all slices needed for the actual dependency closure when currentness changes (5.13).
Apply the definitions and tests in A.19.CPM, A.19.SelectorMechanism, C.18, C.19, and G.5 when comparator, selector, archive, pool, or result-declaration claims are current; use E.17 for a source-backed publication face and return to source, E.24.PUB for the publication occurrence, form, carrier, audience, bounded use, and availability, A.21 for a gate claim, and G.11 for a refresh claim.
For evolutionary-engineering work, the same selected structure may contain, for example, loci for variant generation, retention, archive or front treatment, comparison, selected-set result declaration, actual publication, architecture-candidate movement, planning, performed work, effect measurement, residual triage, and refresh.
E.18 defines the selected organization, its position and occurrence resolution, crossings, valuations, pins and dependency-addressed refresh.
Apply the definitions and tests in C.18, C.19, and G.5 when archive, pool, or selected-set result-declaration claims are current; use E.17 for a source-backed publication face and return to source, E.24.PUB for the publication occurrence, form, carrier, audience, bounded use, and availability, C.11 and C.30 for their decision and architecture-candidate claims, the A.15 family for planning and performed Work, and G.11 for refresh.
Position and parent-relative subflow references
Use a FlowPositionRef to point to one structural position inside one TFS:
The pair is the complete position-reference identity. If the TFS is reidentified, the same local id resolves to a different position. A FlowValuation, PathId, PathSliceId, actual filling, DesignRunTag, value kind, and reference mode may qualify or bind a use of that position; none of them enters its identity. These use annotations do not override A.22: replacing a selected constituent or relation occurrence requires reidentification before reusing the reference.
Use a SubflowRef when the practitioner needs to select and revisit a detailed internal portion of one parent TFS without treating the portion as a separately identified structure:
Every included and boundary position must resolve through FlowPositionRef in the same parent S. Each included occurrence is selected in that parent's R and resolves through its q without losing participant meanings or arity. A participant outside the included internal portion remains explicitly resolved at its parent boundary. If the required participants cannot resolve in that parent, the proposed portion is not an internal subflow. This parent-relative selection adds no world-side parthood, containment or membership relation.
The tuple is the complete SubflowRef identity. Replacing the parent, an included position, an included internal relation occurrence, or a boundary position gives another reference; reidentifying the parent invalidates the old resolution. Changing only a valuation, path or slice, tag, actual filling, graph, mathematical description, publication, or demonstrative view leaves the reference unchanged while the tuple still resolves. Branching, joining, or cycling inside the portion does not make it a network.
Quick discriminator. Grinding, dosing, and wetting may be shown as a coffee-preparation subflow while their positions, internal relation occurrences, entry, and exit all remain in one coffee-brewing TFS. If heating instead has its own TFS identity and boundary and an identified relation connects it to preparation, stop using SubflowRef and apply E.18.NET.
S3 - Publication discipline (faces)
E.18 imports E.17 in full and associates MVPK faces with PublicationScope (USM).
MVPK remains the source for:
- the set of face designators (
PlainView,TechCard,InteropCard,AssuranceLane), - pin discipline and Publication Characteristics (PC),
- source fidelity, one authoritative signature for inputs and outputs in the optional morphism profile, and separation of publication morphisms from the resource use of performed Work.
E.18 adds only the following structure-scope obligations to those rules for faces published over transformation-flow paths:
- Crossings on faces. When a face publishes a GateCrossing, it cites the
CrossingRef,ChangedBindingAccountRefs[],GateId, and any currentGateDecisionResult, optionalDecisionLog, policy-application, or permission-claim refs. An F.9 Bridge block appears only for a separately established cross-semantic use; its optional card andCLdo not replace those refs. - Edition refs on faces. A face that cites
CG-Spec,ComparatorSet,UNM.TransportRegistryPhi, or another versioned value cites that value and its edition. Edition citation alone requires no Bridge Card, UTS row, or semantic Bridge. - Comparison and selection on a face. Identify the editions of any
ComparatorSetandSetSemanticsRefthe path uses. For a source contrast, report what the comparison establishes under its criteria. If a selection or ordering is published, preserve its actual result, declared order and any supported scalarization. Reconsider affected uses when a relied-on premise changes and re-emit a face when the result or an applicable publication rule requires it. - Gamma_time on compare and launch faces. Every current compare or launch publication face on an E.18 path pins
Gamma_time; implicit latest is not admissible. A.21 cites the current profile application and qualification window. CHR avoids acceptance thresholds (NoThresholdsInCHR); gate and threshold claims are carried by A.21 and Part G, while actual performed facts are established through independently obtaining relations involving exact Work occurrences under A.15.1. A sourceunknown,notRun, or error remains explicit before the current profile rule maps it to a gate decision.
MVPK supplies the "signature" naming rule, the optional morphism profile's input-output rule, arithmetic-visibility rules, and material numeric-pin requirements (E.17 §5.4-5.5). E.18 does not weaken those rules;
CC-E18-09states the additional constraints on faces published along transformation-flow paths.
The recipient may read the needed input or output declarations in the identified signature edition through a timely authorized route, including a checked offline copy. When no such route is available and the face is to support that decision, include the operative declarations on the face, attribute them to the signature edition and retain the material pins. Check the accessible expression against the source.
Lean publish-mode (AssuranceLane-Lite). Lean changes publication faces only, not policy or checks. A current face cites the profileApplicationRef, identified GateCheckApplicationResult refs, and GateDecisionResultRef; it cites a DecisionLogRef only when an audit, history, replay, or reuse record is current. The underlying check-application results remain unchanged.
Decision stability and idempotency (gate-local). A gate decision is recomputed when an input named by A.21 changes. An equivalence witness is needed only for a current reuse, cacheability, or stability claim covering the inputs whose equality that claim relies on; an optional DecisionLog may cite it. Use G.6 for evidence-provenance path visibility and G.11 for refresh implications. E.18 does not prescribe storage formats, key shapes, or hashing schemes.
Retargeting and semantic-Bridge boundary.
An EntityOfConcernRef change is not established by a UTS row, mapping label, card, CL, or GateCrossing; a kind change alone only reopens the C.2.1 identity test.
First recover the A.6.4 arrow r from its endpoints, arrow rule or designator, and formal equivalence. Separately recover q, whose ClaimGraph states the invariant, visible loss, named receiving use, conditions, and affirmative or negative polarity. Compare the current case facts with q and keep the current-case judgement separate: satisfies, fails, or cannot decide; for cannot decide, name the missing fact and reopen condition.
Any application occurrence and Work remain separate. If the use also needs a semantic relation between two exact local senses, apply F.9 separately and keep its own bounded-use claim, optional CL, evidence, and reliance separate.
S4 - Assurance operations on connection positions
The four declarative assurance operations act on the annotation of connection(q) for a single flow use. rho resolves q to the selected occurrence. When the assertion concerns a participant, name that subject as well. The operations preserve CtxState and the identity of r.
For a cut, identify the affected q values or state a separately governed aggregate-claim rule. The cut identifies part of the representation. It does not by itself establish another subject relation. Declaration-change labels from A.6.6 may describe an edited assertion; they neither replace these operations nor establish their subject relations.
CalibrateTo may cite an edition already compatible with the current Evec binding. A change from C3 to C4 that changes that binding needs its from/to basis and a GateCrossing when an applicable profile governs the transition. The same rule applies to a changed plane, units declaration or locality; the assurance operation preserves the explicit changed-binding account. If the conversion governor is absent, return missing-governor; if required declarations or case values are unavailable, return missing-information; otherwise report the applicable rule’s positive, negative or inapplicable result.
A penalty requires an applicable current policy and its rule application. Cite PolicyIdRef and any separately needed authority relation and participants, and publish the result only in the assurance lane named by that policy. Calibration, a registry entry or CL alone supplies no penalty or gate decision.
S5 - Comparability and aggregation
Use the following conditions to decide whether the comparison is admissible:
- Compare on an already adequate common basis. Apply A.19.UNM before comparison when the CN-Spec, applicable protocol or receiving use requires re-expression. UNM admits different normalization methods. Retain the method used for this case, its invariants and edition, and any transport rule used. The comparison and aggregation still require their applicable rule, admissible operands and evidence. A missing compatible basis remains a comparison gap; a normalization label does not supply it.
- A comparison can settle a relation between two candidates under the stated criteria. Preserve a partial or total order when the comparator actually defines it. For a shortlist, joint-use set or archive, report the result returned by the selected procedure under its declared semantics. Identify any scalarization and its basis.
- If a claim is ordinal-only, publish comparison results under its declared order. Arithmetic transforms such as means and z-scores require a separately governed scoring method and Scale that support them under C.16; naming a comparator or downstream policy does not supply that justification.
Edition-aware publication records for sets or archives (e.g., QD archives) pin DescriptorMapRef.edition, DistanceDefRef.edition, and the CharacteristicSpaceRef with its separate characteristicSpaceDescriptionRef: U.EpistemeRef under A.19:5.1.1.1 when applicable; refresh covers all actual dependent uses under 5.13.
For current selector, archive, pool, selected-set result-declaration, comparator, or refresh claims, apply the definitions and tests in A.19.SelectorMechanism, C.18, C.19, G.5, G.9, and G.11. For actual publication, use E.17 for a source-backed face and return to source and E.24.PUB for the occurrence, form, carrier, audience, bounded use, and availability.
S6 - Cycle discipline (Selection ↔ Planning)
-
The selected structure may center a loop between the
SelectionAndTuninglocus, whose relation satisfies the named selector and comparator definitions or tests, and theWorkPlanninglocus, which binds oneA.15.2 U.WorkPlan. Any A.15.3 planned-filling row remains declaration-local content inside that WorkPlan. -
Declare the local budget and max_iter in
Γ_time. At expiry report the outcome supported under G.5:4.4b–4.4e. An admitted selection may have returned a set, including an empty shortlist under its criteria. Missing required grounds can justify abstention, and the declared handoff or escalation requires its own grounds. A search stopped by the budget establishes optimality only with a proof or adequate coverage of the stated domain. For a result over explored candidates, give that extent. If further tuning is needed, identify a separateU.WorkPlanwith any A.15.3 declaration-local planned-filling rows, or a configuration or policy that passes its applicable rule. Carry it into the nextPathSlicethrough its planning, configuration, policy or refresh relation. -
Establish the comparison basis before the loop. Use it directly when adequate; apply UNM when normalization is required. If the applicable measurement, currentness or UNM check finds missing or stale measurements, retain that finding. A request for fresh measurements does not by itself establish a refresh plan.
If the receiving use plans measurement refresh, A.15.2 identifies the WorkPlan; when a reusable declaration member must be pinned, A.15.3 adds only a declaration-local row inside that WorkPlan. For later dated refresh Work, recover each actual performer through A.13 and let A.15.1 independently admit the occurrence. Add F.6 only when the receiving use also consumes precise assignment-bound attribution; F.6 neither discovers the performer nor supplies classification, and its failure alone leaves independently admitted Work intact. Keep the later measurement and calibration separate. A
RefreshReport@Contextis likewise separate from the request, plan, Work, measurement, and calibration.A publication that states a calibration target cites the calibration reference and any applicable transport-conversion rule. A penalty requires its own current policy, applicability, rule application, and any authority relation actually used; calibration, conversion, registry publication, or a report supplies no penalty by itself.
-
Work-entry claim and actual Work stay distinct.
workEntryClaimRefdesignates oneU.WorkPlan, A.15.5 readiness relation, or other prospective claim consumed byLaunchGate. If Work later occurs, each actual launch value is established only through an independently obtaining direct relation or A.6.1 application binding of that Work individual. A separateFinalizeLaunchValuesepisteme may then designate the Work occurrence and those facts; it neither performs Work nor fills slots in the occurrence.
Refresh orchestration. Telemetry records and publications that designate an identified Work occurrence are slice-scoped, editions re-pinned, and faces re-emitted. Telemetry remains a separate episteme and does not constitute the occurrence.
S7 - Selector semantics (G.5) and parity harness (G.9)
For a selector-facing outcome, keep the selection, its result binding when one exists, the declared outcome and any exploration archive identifiable along the path. Use A.19.SelectorMechanism, C.18, C.19, G.5, G.9 and G.11 for their selector, archive, dominance, comparison and refresh rules.
- Apply set semantics to set returns. For an admitted Select, preserve the actual
CandidateSetand its binding. For abstention, preserve the failed or unknown condition. The default DominanceRegime isParetoOnly. IlluminationSummary, coverage and regret remain report-only telemetry unless a CAL policy identifies them as dominance inputs (policy id in SCR).
PortfolioMode=Archive retains exploration evidence under G.5:4.4e and CC-G5.23. It neither activates QD nor names the emitted selector outcome. When an actual selector relation constructs or returns a QD archive, retain that archive under its C.18 and ArchiveConfig basis; a G.5 outcome is declared separately under G.5:4.4b–4.4e.
When generation is in scope, pairs {environment, method} are managed under declared EnvironmentValidityRegion and TransferRulesRef; parity records and PathSliceId are pinned on publication.
For current selector, archive, pool, selected-set result-declaration, comparator, or refresh claims, apply the definitions and tests in A.19.SelectorMechanism, C.18, C.19, G.5, G.9, and G.11. For actual publication, use E.17 for a source-backed face and return to source and E.24.PUB for the occurrence, form, carrier, audience, bounded use, and availability.
S8 - Guard aggregation assignment and handling (USM §1.2)
- USM.CompareGuard and USM.LaunchGuard publish the guard-gate aggregation assignment field
GuardOwnerGateId. The legacy field designates the gate assigned to aggregate failures. It does not establish an owner relation. Guard failures are events aggregated by the declared gate (not GateChecks). - Aggregation-assignment rules: (i)
USM.LaunchGuard.aggregationGate = LaunchGateId(workEntryClaimRef), where the ref resolves to the prospective claim consumed by the gate and never to a not-yet-existing Work occurrence; (ii) inside a Subflow,USM.CompareGuard.aggregationGate = OperationalGate(InSentinel); join loci cannot be assigned as guard-pin aggregation gates.
Profile-application boundary (cross-reference). A.21 distinguishes a GateProfile description from the current fact that applies it to one gate, subject, action, scope, and window. E.18 cites that application only where a current gate or crossing needs it; a profile name, matrix, branch, or PathSlice supplies no application or authority by itself.
Scope-translation guards (cross-reference). A.2.6 defines and tests exact slice and scope membership and any actual translated-scope application. When that translation relies on different local senses, it additionally requires an obtaining F.9 Bridge, a separate affirmative C.2.1 bounded-use claim, and current A.10 or B.3 reliance. Use A.21 for gate aggregation; no CL value or Bridge Card decides the guard.
Error, timeout, or unknown (profile-bound). Keep each source error, timeout, unknown, and notRun result explicit. The current profile application cites the rule and edition that maps that result to abstain, pass, degrade, or block. A profile name alone supplies no fixed fold, and no missing or unrun required result maps to pass or neutral abstain. The GateDecisionResult retains the mapping and rationale.
S9 - Transport and crossings
- A GateCrossing records one selected-structure transition between identified source and receiving positions with their
CtxStatebindings at one gate whose profile and decision test come from A.21. Cite A.2.6 for locality and scope membership, the current plane or units declaration and conversion rule, each versioned value and its edition plus G.11 when refresh is current, A.21 forDesignRunTagand the gate decision, and A.15.5 for a prospective work-entry boundary. If no current predicate, applicability condition, occurrence rule, conversion rule, or decision test can state a claim on which the crossing depends, returnmissing-governor. If the governor exists and the available case basis is sufficient to apply its positive test but that test fails, returnfactually unsupported; if a fact or declaration needed to decide the test is unavailable, returnmissing-information. State a negative crossing only under an applicable non-obtaining criterion or complete closure basis and satisfying facts. - A semantic F.9 Bridge is additional, not constitutive. Use it only when the case identifies two F.17
SchemeSenseCellvalues from different semantic contexts and the Bridge predicate obtains. Keep the proposed structural use in a separate C.2.1 claim, recover current A.10 or B.3 reliance when relied on, and keep any Bridge Card orCLoptional and non-constitutive. - An EntityOfConcern change remains with A.6.4. E.18 may place the independently identified r, q, and current-case judgement; it creates none of them, records no operation application by implication, and supplies no
KindBridgeor mandatoryCL.T^D↔T^Ris handled at the exact A.21 gate withDesignRunTagFromandDesignRunTagToand the current A.15.5 or publication locus, without implying Work occurred.
S10 - Non‑mechanism boundary
- Within E.17's optional morphism profile, publication uses typed projections that carry no Γ-semantics. Ordinary publication uses E.24.PUB's form, carrier and publication relations. Building, rendering or uploading is separate work on carriers; a claimed dated Work occurrence still needs its A.15.1 basis.
S11 - Coordination wording labels (when current)
Coordination wording may be published as LexicalView labels over a P2W carry-through flow valuation; it is orientation-only unless an exact structural crossing, work relation, semantic Bridge, or gate decision is independently current. The wording adds no current structure locus kind, check, or mechanism. A published crossing cites CrossingRef and ChangedBindingAccountRefs[], with gate-decision and permission-claim refs separate when current; an F.9 block is added only for a separately established semantic Bridge and bounded use.
S12 - Exact viewpoint references to E.18 constructs
Use S12 only when a current claim maps one identified viewpoint episteme to the E.18 constructs used. Ordinary work with a selected transformation-flow structure, valuation, path slice, or crossing does not open S12, and one mapping may stop after one row.
Imported interface. E.17.0 defines viewpoint membership and episteme–viewpoint conformance. E.17.1 defines local catalogue declarations and U.ViewpointRef members. E.17.2 provides the project-local TEVB authoring template; it ships no catalogue, reference, or viewpoint episteme value. S12 uses those results and does not copy their catalogue or conformance procedure. E.24.PUB defines publication, and C.29 defines any separately current representation or correspondence.
First useful move. Resolve one viewpointRef : U.ViewpointRef under the effective reference scheme to the named viewpoint episteme. Then name only the E.18 participant positions, selected internal relation occurrences r, resolved connection positions q, gates, crossings, paths or valuations consumed by the mapping. A relation reference names r under its direct predicate; a connection reference names q with rho(q)=r.
The viewpoint reference is not a relation, the viewpoint episteme is not a template position, and the mapping establishes neither the selected subject relations nor episteme–viewpoint conformance.
Stop there unless the current claim separately needs candidate-view conformance, whole-family coverage, retargeting, cross-context meaning, publication, representation, or actual Work. Follow the defining pattern for that claim rather than reproducing it here. A token such as VP.Functional may remain P's ordinary reader-facing designator after resolution; it is not a viewpoint id, reference, family member, or conformance result.
The four rows use one grammar: a specified reference resolves to the identified P, and a separately current mapping claim names the E.18 constructs it uses. A project may use one row without materializing the other three. Four rows are required only by a separately identified whole-family coverage claim under E.17.1/E.17.2.
Conditional map row. Persist UTS.ViewpointMap only when the mapping claim is made or consumed:
The optional branches carry references to independently established results. They do not repeat the classification, assignment, responsibility, Work, publication, representation, Bridge, or retargeting tests. If a semantic-context comparison is current, use the exact F.17/F.9 path and bounded-use claim; catalogue provenance or equal labels supply none of them.
S12-scoped checks, only when UTS.ViewpointMap is current.
ViewpointRefresolves P under the stated effective scheme. The row never calls the reference a relation or P a position.- Every
PrimaryE18ConstructRef, crossing, and gate resolves the E.18 value or occurrence consumed by the mapping; the row creates none of them. - Candidate-view, whole-family, publication, representation, cross-context, retargeting, and Work branches appear only when that separate claim is current and cite its defining pattern's result.
- One-viewpoint use needs one row. Familiar labels or four unbound template names establish no whole-family coverage.
Refresh every actual dependent use
Start with the exact changed source, binding, calibration edition, rule or result. Recover the argument-use, premise-use, source-use or other relations that establish which continuations rely on it. The incidence representation helps locate them; graph reachability alone does not establish a dependency.
Apply G.11 to obtain the least closed set containing those direct uses and every further use whose result or admissibility depends on them. Include both dependent branches and their join, and further gate or publication uses.
At an n-ary use, recover every operand so that reconsideration is complete; an independently supported operand may be reused. Stop expanding only when no such unresolved dependency remains. A known dependence on an unavailable new value remains in the affected set with its governed unknown or stop.
Use PathSlice identifiers to address the work, and extend the set of addressed slices until it covers that closure. A changed source shared by two slices reopens both and their join. A cross-member dependency follows the exact E.18.NET relation and retains each member's local state. “Slice-local” means that each action is addressed to the affected portion; it does not confine refresh to the slice that first detected the change.
For each affected use, decide whether its direct rule permits current reliance on the old result, requires a new assessment, requires a new application, or stops continuation.
A stale judgement does not erase an actual historical occurrence. A new application has its own bindings; selecting those instead of the old occurrences reopens A.22 identity. Reconsidering a use does not always require rerunning it.
Join the refreshed branches only when all the inputs required by the join rule are available and compatible, including their relevant editions. When a new gate decision is required, publish it separately under A.21. Re-emit a face when its current claim or required publication status changes. Preserve unaffected results and independently supported uses.
Archetypal Grounding
Worked case: one measured result, two checks and their joint decision. A practitioner is considering a welding job. Its separately identified WorkScope admits humidity below 60%, current from 35 to 45 A, ER70S-6 wire, and an FW-2.1 controller. The job has the stated wire/controller and humidity 55%; current must be obtained from the measurement. Qualification is checked separately. This is a constructed case.
The sensor-read declaration identifies an application by its actual acquisition cycle on sensor S7. Each calculation declaration identifies an application by the controller invocation from accepted arguments to return. The A.6.1 binding rules apply within those extents. Equal argument values in another invocation do not make it the same application.
The argument sourceResult has a declaration-local range of result-binding occurrences: a member must be an actual binding under the sensor-read result declaration, resolve to its independently identified application on S7 in the applicable sample window, and bind a value in that declaration's raw-count range. The range is the ValueKind. Below, rM17 is one argument value.
A/B result arguments used by the join have their own local range: actual result bindings under the respective check declarations, with the declared current-assessment and recognition-result meanings. No universal result-binding kind or common input/output range is introduced.
C3's independently established calibration rule supplies 1 A/count for S7 and this window. Its applicability and calibration facts are part of the case, not consequences of citing C3.
The numeric current channel contains numeric quantities. A declaration may return a recognition judgement unknown when an identified, admissible input lacks a deciding value; that does not insert unknown into the numeric range. An unresolved argument, a notRun application and an evaluation error remain different cases.
bA-source(A17,rM17) and bB-source(B17,rM17) obtain because the result-binding occurrence participates under the declared argument meaning. A stored “40” or an unused reference fails that test. bA-cal(A17,C3) and bB-cal(B17,C3) separately establish both calibration uses. The claim that A17 used M17's result follows from those actual bindings and the operation semantics; it does not require another transfer occurrence.
J17's declaration-local ApplicationPredicate has the complete participants <controller, leftResult, rightResult, joinRule, returnedValue>. It obtains when one completed invocation consumes those two result bindings, applies JP1 and returns that value. Its occurrence identity follows the invocation and its call/return extent. j17 denotes that same application occurrence; rJ17 is its distinct result binding. Two incoming lines in a drawing do not replace the five participants.
OperationalGate G17 separately applies GP1 to J17's check result and the independently identified qualification result Q17. GP1 maps true and qualification pass to pass; it maps false or qualification failure to block. Its explicit rule also maps a required unknown, error, missing or notRun result to block while retaining that source outcome and the condition for reconsideration. Here the check and qualification pass. The resulting GateDecisionResult does not establish later Work or an actual change to a weld; those use A.15.1 and A.3.4.
Select and use the organization. S17's constituents are the named applications, sensor/controller, calibration and rule values, their actual arguments and returns, the result bindings used as arguments, and the gate/check-result participants. R contains the four operations' declared actual argument/result bindings, j17 and the independently established GP1 check-consumption relations. Selecting every mandatory binding of those declarations fixes the participant closure; Q17's internal performance need not be selected.
C requires exact source-result identity, complete arity, both calibration-dependent branches, their joint basis and separation of check results from gate decision. F asks which results of episode 17 support current launch reliance and which uses must reopen when that basis changes; it stops a stronger continuation at unresolved participants, relations or incomplete join input.
beta(pA)=A17 and beta(pJ)=J17. rho(qA-source)=bA-source and rho(qA-result)=rA17. rA17 is also an argument of J17, so its participant node and result-binding relation node are identified as specified in 5.1. The complete source-use walk is pM17 → qM-result=p-rM17 → qA-source → pA17 → qA-result=p-rA17 → qJ=pJ. Each step follows the corresponding application, sourceResult or leftResult incidence. At qJ retain rB17, JP1, the controller and returnedValue as well. The path shows actual source use; A17 and B17 may run in parallel.
For a concrete cut at J, choose the entry boundary {p-rA17, p-rB17, p-JP1, p-controller} and exit boundary {qJ=pJ}. The path family contains exactly the four one-incidence paths from those entries to qJ under j17's argument/application meanings. The cut kJ consists of its leftResult, rightResult, joinRule and controller incidences. It intersects each named path. Preserve the full qJ, including its returnedValue incidence, even though that incidence is not in this cut. A cut valuation may return the role-indexed tuple <left:rA17, right:rB17, rule:JP1, controller:the identified controller>; it is not a count of independent measurements.
nu0 reports the historical returns on their connection positions and has a separate channel for present reliance suitability. It retains that A and B used the same sample and calibration. sigma3 uses L with current 40 A, one ReferencePlane P with declared raw-count and corrected-current channels, historicalCalibration:C3 and currentCalibration:C3 in Evec, and the design tag consumed by this work-entry use. A/B perform their declared calibration arithmetic; a raw connection carries rM17 under the unchanged channel declaration. PublicationScope constrains use of L and supplies none of these bindings.
ConstrainTo at qA-result applies the named current-region rule. CalibrateTo cites C3 for the identified measurement claim. CiteEvidence retains the measurement evidence and the claim it supports. AttributeTo cites the actual result/argument chain supporting the provenance claim. The four annotations preserve sigma3 and the historical occurrences; an ordinary sentence can carry them when no durable replay record is needed.
Change the shared basis and complete the join. Suppose independently grounded C4 applies to the same raw sample and window with 1.1 A/count, and the currentness rule now requires C4 for continued use. First nu1 can reassess reliance on the unchanged S17: its historical return is still (40 A,true), while both dependent uses and their joint result require reconsideration. Preserve historicalCalibration:C3; change currentCalibration to C4 with its from/to account and required GateCrossing. Establish a changed L when the corrected current is obtained. These use changes do not rewrite bA-cal or bB-cal.
The actual dependence closure contains A, B, J, the consuming gate use and dependent publication faces. It excludes new acquisition of the unchanged raw sample and rerunning independently supported Q17.
If recalculation is chosen, new A18/B18 use rM17 and C4. Their returns are (44 A,true) and (44 A,false). J18 consumes those result bindings and returns false; the new GP1 application returns block. Selecting the new occurrences yields another S18 under A.22, not a rewritten episode 17.
Mixed rA18/rB17 with known different calibration editions yields false under JP1 even if both individual recognition values are true. With admissible identified arguments but an unavailable deciding value, JP1 returns unknown; with an unresolved required argument, the application is not yet established. GP1 retains that distinction while blocking continuation. Refreshing only the first detected slice cannot produce the required joint result.
A useful short note is enough for immediate coordination:
C4 now governs current reliance on sample M17. Reconsider both current checks and their JP1/GP1 decision; retain the raw sample and independent qualification. The corrected current is 44 A: the work region admits it, but the 43 A limit does not. Retain the old episode's results and cite the new application/result bindings for the recalculation.
For replay, retain the exact source/result bindings, C3/C4, rules, application extents, both refreshed results, joint decision and affected faces. The note and record report the grounds; they do not create the occurrences.
A second acquisition M18 may return the same 40 counts with another rM18; equality does not change which result A17 used. In a continuing stream application, two equal-value episodes are distinct only when its declaration and facts establish separate maximal binding extents; two log rows during one continuous binding do not.
A new uncertainty or suitability valuation may preserve S; substituting a selected occurrence reopens it. Corrected 42 A remains inside the original WorkScope, while 46 A falls outside it without changing that scope's extension. A publication scope containing two admissible slices still selects no unique L or P/Evec/D. These variations apply the same identity, valuation and dependency rules.
One first-principles P2W specialization. A first-principles-to-work path is one path through a selected transformation-flow structure, not P2W as a whole: U.Signature(profile=FormalSubstrate) declaration, principle frame, mechanism, normalization, selection, planning, pre-run work-entry claim, later Work occurrence when one exists, and current evaluation or currentness relations occupy identified positions. E.18.1 separately carries the accepted problem-side claim through whichever of those relations becomes current.
Cooling-loop first use. For an emergency-load review, begin with the independently grounded cooling-loop stabilization Transformation, its control-law mechanism, the measurement and setting-change WorkPlan, and any dated test-run Work that occurred. Select only the relations whose predicates and case facts connect the participants for that review, then apply the constraints and use frame in 5.1. A work-to-change or production claim uses its direct predicate or the local claim route stated in the Intent; proximity of the plan, run and stabilization supplies no such claim. Locate the needed path or slice. Stop at that first useful selection when no crossing, launch, publication, comparison, cycle, refresh or assurance claim is current. An emergency-load model valuation can reuse S only while its selected constituents and obtaining relations remain unchanged.
Supply chain. The loci run from procurement through inbound quality control and normalization to supplier selection; selection and planning inform each other; planning leads to execution and then refresh. Execution includes receipt Work admitted under U.Work and keeps its receipt records separate; refresh uses quality telemetry and re-emits the affected faces.
When an incoming lot moves from the supplier-receipt position to the internal-QC position and its locality binding changes, record the crossing with CrossingRef, the A.2.6 locality fact, the rule that permits the change, and the A.21 gate. If the two labels also use different local senses, identify their F.17 cells and test the F.9 Bridge, the C.2.1 bounded-use claim, and current reliance separately.
A penalty appears only when a current policy is cited through PolicyIdRef, the policy applies, its rule supports the penalty, and any separately needed authority relation and participants are established. Comparators remain pinned to the named CG-Spec value and edition.
Neural-net functional structure. Loci: U.Signature(profile=FormalSubstrate) declaration (typed tensor-operation declaration) -> mechanism (combinator algebra) -> UNM (dataset normalization; TransportRegistry^Phi) -> selection (architecture and hyperparameter set; Pareto set over accuracy@ratio and FLOPs@ratio) <-> planning (compute budget horizon) -> Work (training-run occurrences admitted under U.Work; any Delta is stated in a separate record) -> refresh (parity inserts; covering the actual dependency closure in 5.13). Faces pin DescriptorMapRef.edition and DistanceDefRef.edition when QD telemetry values are shown; illumination remains report-only telemetry by default.
Developed product, then application - network case. Development, later application, and further use keep separately identified TFS values when they have their own identified objects, Work occurrences, local position bindings, DesignRunTag boundaries, and change boundaries. A tool may be made, then used to make a chair, then the chair may be used while a person writes a text. Apply E.18.NET to select those TFS members and cite each exact production, use, participation, or other cross-member relation with its defining predicate and occurrence rule. The cross-member claim uses that direct occurrence, not an internal connection position from one member. If its required predicate or occurrence rule is absent, return missing-governor.
FPF pattern development and use - network case. Pattern development, application to an EntityOfConcern, and use-found evaluation keep separately identified TFS values when each has its own identified object, Work, positions, and local state. Apply E.18.NET and cite the exact use, evaluation, evidence-return, or repair-trigger relation that connects their positions, together with the pattern or declaration that defines its predicate and occurrence identity; if no such rule is available, return missing-governor.
E.18 defines each member's internal structure; its 5.13 refresh rule covers every dependent use through the needed PathSlices. A reader-facing role label remains ordinary wording until E.10.ROLE recovers its meaning; neither that label nor a feedback arrow makes the members one TFS or supplies the cross-member relation.
Cross-pattern boundary slice (QD archive). A QD selector returns an archive. Under E.18, the selector occurrence and returned archive may occupy positions on one PathSlice; the archive is returned by the selector relation, not by the slice, and remains a set or archive rather than a hidden scalar. Under A.20, an archive insertion or update step may have exact constraint-validity results and a complete required-set summary; no acceptance is inferred. Under A.21, a comparability gate or LaunchGate publishes a decision only when its gate relation consumes the declared independent check results. Under E.20, a newly introduced selector-mechanism definition remains the meaning locus. These are four identified loci, not one prescribed work order.
Bias-Annotation
- Acyclic-bias risk. Tooling accustomed to DAGs may discourage admissible feedback loops. E.18 permits loops with budget and sentinel controls (CC-E18-13, -18).
- Scalarization-bias risk. Cultural defaults to single-score rankings can suppress Pareto fronts and QD archives; E.18 keeps declared order relations and return sets visible (CC-E18-10, CC-E18-12).
- Interop-dominance risk. File and format ecosystems (CWL, RO-Crate, and lineage) can be mistaken for semantic sources; E.18 places them in InteropCard and keeps the applicable semantic definitions and tests explicit at the loci and gates.
- Over-formalization risk. Category-theoretic formalisms can obscure operational guard-rails; ordinary E.18 states each crossing in readable prose, while replay-facing use keeps exact positions, per-binding accounts, one separate A.21 gate decision, and
CrossingRef(CC-E18-11, -23). - Retrospective rewrite risk. Keep historical occurrences and edition pins intact. Reassess their current uses through the dependency closure in 5.13 and CC-E18-18; a new actual execution has its own identity.
Mitigations. Select checks from the current use before applying a profile. Require edition pins when a current comparison or publication depends on them, inspect the GateDecisionResult for a current decision and an optional DecisionLog only when audit, history, replay, or reuse depends on it, and cover every dependent branch, join and consuming face through the required PathSlices (5.13).
Conformance Checklist — Unified checklist (normative)
Conformance use. The branches in use determine which checklist rows and clauses apply. Start with the ordinary selected-structure core: one selected structure, independently grounded locus values, selected obtaining relations with complete position resolution, and the current position, path, path slice, or valuation only when the use needs it.
Apply a row only to the Solution use named by its requirement. If that use is absent, the row—or the branch-specific clause within a mixed row—is not applicable.
Activate branches from the current use. Add crossing and gate tests only for a current GateCrossing, OperationalGate, StructuralReinterpretation, or work-entry boundary. Add launch tests only for a current LaunchGate or launch claim. Add publication tests only for a current MVPK face or publication claim.
Add comparison and selection tests only for a current comparator, selector, comparison, or returned set or archive. Add cycle and refresh tests only for a current loop, freshness question, edition change, or refresh use. Add assurance, guard, decision-log, evidence-lane, or replay tests only when that claim or downstream reliance is current.
Choose a profile after identifying the active branches. It can strengthen checks inside an active branch but cannot activate another branch or require its absent objects.
The whole table remains available when a use combines many branches. Where one CC row contains clauses from more than one branch—for example path composition and publication functoriality, or compare and launch pins—apply only the clauses for the branches that are current.
Coupling note.
CC-E18‑07preserves the independent source results thatCC-E18‑21amaps and joins. Evaluation order may save work, but it cannot change applicability or erase a deferred required check. Scope of E.18. Use the definitions, constraints, and tests named in this pattern's Relations for mechanism-specific checks and publication obligations. E.18 fixes only selected-structure obligations: selected obtaining relations and their resolved positions, gate crossings, typed valuations, publication pins, separation between internal constraint results and profile-fit results, and complete dependent refresh through addressed slices.
Glossary (additions)
-
Open-world locus uses - positions for independently identified values with their own kinds, defining sources and identity rules, including domain-specific specializations. The common markers in 5.1 are examples, not required parent kinds. An A.6.1 application or result binding retains its declaration-local kind and identity.
-
Signature locus - structure-positioned use of A.6.0
U.Signature(universal block). It is an independently defined value bound into the selected structure, not a local kind and not aC.3.2 KindSignature. -
KindSignature (C.3.2) - definition of a
U.Kindby intent, extent, and formality; unrelated to E.18 locus kinds; never agenus. -
Species (domain-scoped) — typed specialisations
speciesOf(kind=...)that declareKindDefinition=<pattern id for the current definition>(e.g.,kind=Mechanism; KindDefinition=A.6.1). -
Semantic Bridge boundary — F.9 defines and tests an obtaining semantic relation between two identified F.17 cells. A structural crossing or A.6.4 retargeting does not imply that relation; a Bridge Card and
CLare optional episteme/evidence apparatus. -
Eulerian interpretation - a use assigns typed values on the connection/cut domain of 5.2 and applies the assurance operations of 5.4. Actual transport or receipt episodes retain their own direct predicates; valuation does not supply token-passing semantics.
-
GateCheckKind boundary.
GateCheckKindis a recognition label within one identified A.21 check application, not a structure locus kind and not enough to identify or merge results. No such label becomes an E.18Checklocus unless anOperationalGate(profile)locus is present. -
GateCheckRef boundary. Where a publication face over a selected structure carries a
GateCheckRef, that value refers to one A.21GateCheckApplicationResult. It must resolve the checked subject, criterion and edition, applicable rule application, case, scope, and window. -
GateDecision, GateDecisionRationale, and GateDecisionExplanation (terminology).
- GateDecision - the lattice value inside one A.21
GateDecisionResult, derived from one identified profile application and its complete required set of identified check-application results. - GateDecisionRationale - the structured rationale inside that result: retained source outcomes, explicit mappings, aggregate, and action consequence. A current publication or optional
DecisionLogmay cite it; neither supplies the rationale or decision. - GateDecisionExplanation — an optional human-readable narrative derived from the rationale; it carries no decision value. It may explain any retained result and mapping, including why an A.20 input prevented passage; absence of a narrative does not make a check inapplicable.
- GateDecision - the lattice value inside one A.21
-
GateFit (aspect, not an entity). GateFit names the aspect of checks that evaluate profile‑fit; there is no separate GateFit entity. “Gate decision under GateFit” means “the gate’s decision computed from GateChecks with
aspect=GateFit”. -
VALATA (VA, LA, and TA) — value-annotation scheme used on AssuranceLane; carriers are referenced via SCR and RSCR; detailed evidence obligations use the definitions and tests in
A.10and the named evidence, publication, or crossing pattern for the current case. -
Connection, transfer and transport - a connection q resolves one selected subject relation r. An actual transfer requires its domain predicate and facts. TransportRegistry^Phi cites conversion rules and policies used by UNM; it neither creates a relation occurrence nor supplies a penalty.
-
GateCrossing - an E.18 structure-local transition between identified source and receiving position/state bindings at one gate whose profile and decision test come from A.21; it is not a semantic Bridge or gate decision.
-
Admissible path - a typed path obeying the GateCrossing discipline: no hidden crossings, every witness required by a current crossing rule is present, compare and launch publication faces are Gamma-pinned when present, and
T^D<->T^Roccurs only atLaunchGate; see S2.
Common Anti-Patterns and How to Avoid Them
Gating Profiles (applied to E.18)
Profiles set strictness only through a current application to one gate, subject, action, scope, and window.
A profile description or name does not by itself introduce a crossing, LaunchGate, publication face, comparator, selector, cycle, refresh, audit record, evidence lane, or Work occurrence. A.21 defines the profile-application, check-application, decision-result, and optional reuse-record boundaries.
Profile selection and change. Cite the current policy-application fact, including its rule and edition, applicability, gate and subject, scope, window, required set, mappings, and any separately required authority. A PathSlice only bounds changed data or currentness and may trigger reevaluation; it neither selects, inherits, overrides, nor weakens a profile. G.11 supplies refresh wiring only when refresh itself is current.
E.18 LEX Discipline (registration)
Register Tech tokens (ASCII) used by this pattern with twin labels: TransformationFlowStructure, TransformationFlowValuation, StructuralReinterpretation, OperationalGate, GateCrossing, CrossingRef, CrossingBundle, GateProfile, GateCheckRef, GateCheckKind, DecisionLog, USM.CompareGuard, USM.LaunchGuard, FlowPositionRef, SubflowRef, FlowEmbed, SentinelId, PathSliceId, SliceRefresh, FinalizeLaunchValues, VALATA. Bridge, BridgeCard, and CL retain their F.9/C.2.1 meanings and are not E.18 crossing tokens. Reference MVPK E.17 naming for faces.
CtxState Extension Registry. Register any extra CtxState slot beyond ⟨L,P,E⃗,D⟩ with: slot id, informal intent, partial‑order rule (with neutral or absorbing), SquareLaw compatibility note, and the applicable change rule, including any required Gate profile. Absence of registration ⇒ non‑conformant.
Consequences
Benefits.
- Universality with discipline: resolved subject relations and explicit gates keep actual participation distinct from work and method order and make cross-domain flows (ML, supply-chain, TAMP and MPC, scientific work structures) analyzable and auditable under the same rules.
- Comparability and replayability: an adequate common basis, normalization when required, and edition-pinned comparison rules support declared return semantics and replay without covert scalarization.
- Locality of change: actual dependency closure identifies all affected PathSlices while leaving independent uses intact; large selected structures remain stable under frequent edition bumps.
- Clean work-entry boundary: When a design-run-tag claim and applicable profile rule are current, a
LaunchGatemay consume the corresponding identified check result. The gate never establishes actual launch values. Those values obtain only through direct relations or A.6.1 application bindings involving one later Work occurrence; acceptance claims and telemetry records remain separate epistemes or relations that may designate that occurrence. - Assurance visibility: When publication or reuse is current, MVPK can make the exact profile application, gate-decision result, optional audit record, and any sufficient reuse witness locally checkable without making them mandatory for an ordinary local structure use.
Trade‑offs.
a) Higher upfront modeling cost: exact crossing positions, per-binding replay accounts, gate refs, and optional durable crossing bundles demand care. When no downstream reliance needs replay, readable prose can describe the ordinary local crossing without a bundle.
b) Longer publication face sets: Required path and publication refs can lengthen faces; lean face sets can be used for low-risk segments.
c) Tooling alignment: some incumbent DAG-only orchestrators conflict with budgeted cycles and set-return semantics; adapters project E.18 semantics to their interop boundary, while E.18.2 carries the mathematical graph-description relation when that projection matters.
Rationale
E.18 governs the selected structure. The patterns below supply the definitions and tests for the relations that the current use requires:
- What the selected structure is: one A.22 selection O/R/C/F, with independently identified participants and obtaining relations resolved through Pos_S/beta and Conn_S/rho. Graph, morphism, tuple, category and algebra descriptions express this subject only for the mathematical use at hand.
- How use bindings change: identify the source and receiving positions, changed
CtxStatevalues and their per-binding bases. An applicable gate profile adds itsOperationalGate, GateCrossing andCrossingRef; gate decisions and permission claims remain separate. A separately established cross-semantic use additionally retains its F.9 Bridge, bounded-use claim and applicable reliance basis. - How comparability works: reuse an adequate common basis or establish the needed re-expression under A.19.UNM. Retain the applicable units, plane, transport rule and comparison editions. Current selector or archive cases preserve their declared set or archive returns and the conditions of
A.19.SelectorMechanism,C.18,C.19,G.5,G.9andG.11. - How change propagates: complete actual dependency closure under 5.13, addressed through the needed PathSlices; editions are monotone. When the selected structure contains a current
LaunchGaterelation for one prospectiveworkEntryClaimRef, that gate is the pre-run decision locus for that claim. Actual launch values are established only through independently obtaining direct relations or A.6.1 bindings involving a later Work occurrence and may be cited by a separate finalization witness.
This arrangement gives checkable conditions for functorial publication on crossings and keeps inner constraint validity distinct from profile fit. A.21 can therefore aggregate mapped check results without evaluation order changing which independently established facts remain available.
SoTA-Echoing (post-2015, multi-Tradition)
Each row connects a source idea to an E.18 invariant, its practitioner implication, and the shortcut it rejects. Vendor, tool, and literature terms are informative. The invariant and its implication explain the pattern.
Relations (explicit pattern-to-pattern relations)
-
E.18 -> coordinates with -> A.15.5 WorkEntryReadiness. A selected structure may position a launch or work-boundary readiness locus only in relation to A.15.5. E.18 supplies the current path and slice, any crossing present in the case, and any applicable
LaunchGateposition or structure-local pins; A.15.5 defines and testsFullKitCondition, planned preparation references, commitment disposition, resource-readiness references, and whether intended work is ready to enter performed-work execution. -
E.18 -> coordinates with -> C.32.P2S ProblemToStructureArchitecturingFlow. P2S may cite a selected transformation-flow structure, path, crossing, or valuation as architecture content or uncertainty. When Plain wording calls that value a method handoff, work handoff, or feedback input, C.32.P2S must identify the receiving entity or relation occurrence independently, including its participants, obtaining condition, and the pattern content that defines or tests it; those labels supply none of them. E.18 still defines the transformation-flow structure and does not become the whole architecturing flow.
-
E.18 -> coordinates with -> C.33, C.34, and C.35 structural-information patterns. When a transformation-flow carrier, path, generated map, or independently identified changed entity or relation occurrence that carries or describes structure needs architecture-specific capture, preservation, or discovery adequacy, use
C.33,C.34, orC.35for that architecture use.Before a selected structure is returned to a named architecture use, cite the selector or selection relation, or another relation occurrence, that returns it; name that relation's predicate, participants, obtaining condition, occurrence identity, and the content that defines or tests it.
Also cite the exact source-to-use relation and the pattern that defines or constrains the receiving architecture claim.
C.33,C.34, andC.35supply definitions or tests; they are not participants in those relations. E.18 keeps the selected transformation-flow structure, path, crossing, valuation, and any exact slice-local subject relation cited by that architecture use visible; it supplies no generic result, return, or receiving relation. -
E.18 -> coordinates with -> A.22.CGUS through E.18.3 when transformation-flow unfolding is current. Under E.18, independently identify the one-TFS or parent-relative internal-subflow substrate; use E.18.NET for an independently identified network substrate.
E.18.3qualifies one separate A.22-selected CGUS only when that CGUS uses exact substrate positions, bindings, and already-obtaining occurrences under current applied-condition claims, any E.18GuardFailevents with their gate-assignment facts, and any independently defined guard-relation occurrences. The substrate ref does not resolve toselectedCGUSRef. Neighboring values and stronger claims remain independently identified and connect through exact supporting relations, predicate-definition content, and current facts. Ordinary E.18 use is not automatically substrate for a CGUS, and narrative, abductive, typing-grounding, improvement, evidence, refresh, and first-entry seed structures do not become E.18 structures by route-shaped wording alone.
Relation rows use the named relation kinds builds_on, constrains, coordinates, specializes, publishes_on, requires, and provides_checks_for.
Foundations
- E.18 -> builds_on -> E.17 MVPK (for publications of selected-structure content). Faces, pins, lanes, functorial publication, Lean, Core, and Regulated profiles.
- E.18 -> builds_on -> A.6.0 U.Signature and A.6.1 U.Mechanism. Locus kinds and governing-definition content boundaries.
- E.18 -> builds_on -> A.7 Strict Distinction (EntityOfConcern, Description episteme, Description episteme admitted for specification use, and publication and carrier separation). No new claims on faces; publication faces project selected structure, crossing, or flow-valuation information without becoming the selected structure, Description episteme, specification use, evidence, gate decision, work occurrence, or carrier.
Flow semantics and checks
- E.18 -> coordinates -> A.20 Flow Constraint Validity. A.20 reports exact internal-constraint results for transformations, operation applications, or A.6.4 retargeting uses when those constraints are current. E.18 supplies no constraint truth, plane or unit declaration, gate consequence, or acceptance.
Terminology discipline (A.20 boundary). Preserve A.20 applicability, evaluation state, outcome, summary, and witness or reason.
GateDecisionRationaleandGateDecisionExplanationremain A.21 terms. - E.18 -> coordinates -> A.21 Gate decisions. When an
OperationalGate(profile)decision is current, it consumes independently identified check-application results under one profile application. An unsatisfied or incomplete A.20 input affects the aggregate under that current rule without making another applicable check inapplicable; each deferred required check remainsnotRun. A.21 defines check-application identity, mappings, aggregation, result and rationale, and optional publication or reuse records. - E.18 -> uses -> USM.CompareGuard and USM.LaunchGuard. Guards publish scope and responsible gate; guard failures are handled by the declared gate.
- E.18 -> coordinates with -> F.9 and F.17 only for a current cross-semantic use. Use E.18 for the structural
GateCrossing; F.17 identifies the two exact local sense cells and F.9 alone decides whether a semantic Bridge obtains. The proposed structural use, reliance, optional Bridge Card, optionalCL, actual gate decision, and any policy-based penalty remain separately identified. - Operational interpretation (default): Eulerian. A flow valuation uses the typed connection/cut domain of 5.2. Declarative assurance operations annotate the resolved subject under 5.4; an actual transfer, receipt or displacement requires its direct predicate and facts.
UNM and comparability
- E.18 -> uses -> A.19.UNM when normalization is required. Reuse an adequate common basis directly. Otherwise establish the normalization and its applicable declarations under UNM; publish the
CG-Spec,ComparatorSetandUNM.TransportRegistryPhivalues consumed. - E.18 -> constrains -> G.5 SelectionAndTuning. Source-backed outcomes, edition-pinned comparators when used, and explicit scalarization; cite the selector-declared set, handoff, abstain, or escalation outcome. Any next-step tuning remains in its separately identified
U.WorkPlanwith any declaration-local A.15.3 planned-filling rows, or in a separately identified configuration or policy that passes its own applicable rule, with no launch-value slot filling. - E.18 -> constrains -> G.11 EvaluatingAndRefreshing. EditionBumpProposal, any applicable two-phase update through the normalization declaration, and dependency-closed refresh through addressed slices. When current, identify
RefreshPlan@Context, dated Work, later measurement and calibration, andRefreshReport@Contextseparately; no request, plan, record, audit artefact, or publication substitutes for another or becomes the returned world-side result by label.
Work boundary
- E.18 -> coordinates with -> A.15.1 Work occurrences and A.15.5 work-entry readiness. When a current
LaunchGaterelation consumes one prospectiveworkEntryClaimRef, its current profile application selects any required freshness, tag, ingress, or other checks and maps their results to the attempted-entry consequence. If Work occurs, A.15.1 defines and tests the exact Work individual and requires the relevant world-side relations involving it to obtain independently; a separateFinalizeLaunchValueswitness, telemetry record, or acceptance claim may designate the occurrence but is not that occurrence. - E.18 -> coordinates with -> A.3.4, A.15.1, and A.15.PROD at actual-change and production boundaries. A
Transformationlocus points to one independently identified actual change underA.3.4; an adjacentWorklocus points to a dated occurrence underA.15.1. A work-causes-change assertion uses A.6.RCD disposition 1 when its predicate and case facts are current; disposition 2 supplies only one local C.2.1 compound claim when no direct predicate expresses it and admitted base-predicate semantics support this receiving use. Work, Transformation, and that claim remain separate. Production-work participation, entity-identity inception, and historically indexed production completion cite separate localA.15.PRODclaims; E.18 neither derives them from proximity nor introduces replacement relation kinds.
Structure and reuse
- E.18 -> provides selected-structure base for transformation-flow families. Flow patterns such as P2W and EvaluatingAndRefreshing use E.18 for selected structure, valuation, crossings, guards, MVPK faces, and slice-local refresh.
A.3.4defines and tests each independently identified actual boundedU.Transformation; E.18 defines the selected compound structure over transformations and adjacent identified loci without asserting transformation composition; and the named neighboring patterns define or test method, work, mechanism, work-to-change, production, evidence, publication, gate, decision, and refresh claims when those claims are current. - E.18 -> coordinates with -> E.18.NET Network of Transformation-Flow Structures. Use E.18 for one TFS, its
FlowPositionRef, parent-relativeSubflowRef, valuations, paths, slices, local state, and selected internal relation occurrences. E.18.NET starts only when independently identified TFS or nested-network members are selected with identified cross-member relation occurrences; it does not replace a detailed internal portion or several valuations of one TFS. - E.18 -> coordinates with -> architecture transformation-flow relation patterns. When a selected transformation-flow structure is used in an architecture-flow relation, the architecture transformation-flow relation pattern records the relation between
TransformationFlowStructureandArchitectureOf@Context; E.18 keeps selected structure, crossing, and flow-valuation discipline. - E.18 -> publishes_on -> E.17 MVPK views (
PlainView,TechCard,InteropCard,AssuranceLane) when selected relation occurrences, resolved connection positions or participant uses are published; Lean mode applies only as per profile.
Conformance Use Checks
Choose tests from the current use, then apply the selected profile to those tests. The full CC-E18 table is available for combined or high-assurance uses; it is not an instruction to run every row for every selected structure.
- Ordinary selected-structure check: verify the selected structure, independently grounded locus values, selected obtaining relations with complete position resolution, and the position, path, path slice or valuation the question uses. Follow the selected relations back to the receiving use. Which conclusion do they support about the organization, and which part of the requested explanation or design still needs other grounds? An ordinary first use can close here when it asserts no crossing, launch, publication, comparison or selection, cycle or refresh, or assurance branch; missing faces,
LaunchGate, selector,DecisionLog, and SquareLaw are then not defects. - Crossing or launch check, when current: for a
GateCrossing, apply its crossing and gate rows, including the exact positions and changed-binding account; add launch rows only for a currentLaunchGateor work-entry claim. Do not infer a Work occurrence from either gate. - Publication or assurance check, when current: for a published face, apply the MVPK, pin, and no-new-claim rows. Inspect
DecisionLog, evidence-lane, replay, or SquareLaw material only when the current decision, crossing, publication, or named downstream reliance needs it. - Comparison, selection, cycle, or refresh check, when current: check the comparison criterion, the actual outcome, and set semantics for an actual set return; apply budget, sentinel, edition, and slice-local replay tests to a current cycle or refresh. Hold editions fixed only for a replay claim, and test an edition bump only for a current refresh use.
- Structural reinterpretation check, when current: confirm one A.6.4 arrow r, an affirmative q, a separate current-case judgement of
satisfies, unchangedCtxState, andPathSliceIdlocality. Apply A.20 only when q also raises a current internal constraint. Test an independent F.9 Bridge and its bounded-use claim only when cross-semantic correspondence is also claimed; keep optionalCL, evidence, reliance, any application, and Work separate.
For an independently identified value at a locus, use its defining pattern for the subject claim. A locus binding does not fill slots in another pattern.
E.18.1 P2W Child-Pattern Relation
E.18.1 is a child pattern for problem-to-work carry-through. It describes the practitioner carry-through practice and, when durable replay is needed, defines its optional C.2.1 note or stop-description claim content. It introduces no local P2W relation kind or occurrence.
Each next method, plan, dated Work, transformation, evaluation, decision, entity, or relation occurrence keeps its independent identity and uses the pattern that defines or constrains the current claim about it. A P2W application consumes this pattern's selected-structure discipline only when a named receiving decision or use relies on an explicit TransformationFlowStructure, path, flow valuation, selected relation occurrence, resolved connection position, crossing or gate position; when branches, joins, guards, or positions for the patterns that define or test their claims must be recoverable, E.18.3 defines that fuller structure.
Under this division of work, E.18.1 carries the accepted problem-side claim and local continuation, while E.18 carries selected transformation-flow structure without making it mandatory for ordinary P2W use.
E.23 Improvement-Loop Boundary Relation
When a transformation-flow structure contains a cycle, budgeted retry path, monitor/escalate path, or slice-local refresh relation, E.18 defines the selected structure: identified constituents and relations, resolved positions, path or slice, gate positions, pins, and the dependent refresh scope. The cycle becomes an E.23 quality-improvement loop only when a named object version is changed and then re-evaluated by a declared object-under-improvement evaluation.
Otherwise it remains a transformation-flow structure, work-control cue, gate relation, or refresh relation; apply the pattern whose Solution answers that exact claim.
Agent-loop diagrams often contain both kinds. A monitor/retry/escalate loop over physical execution state may be a valid TransformationFlowStructure and may include an A.21 gate, but it does not prove that the controlled object improved. If the harness itself is improved, use the E.23 object-version improvement definition and test; if the harness only runs work, use the A.15 family to identify and test the work occurrence.
E.18.3 Constraint-Governed Unfolding Relation
Open E.18.3 when one A.22-selected CGUS is being qualified by its use of an independently identified E.18 substrate. Name selectedCGUSRef separately from the mutually exclusive one-TFS, parent-relative internal-SubflowRef, or E.18.NET substrate branch; then recover the transformed concern, exact substrate positions and bindings, selected obtaining relation occurrences with their direct predicates and resolved connection positions, paths or slices, crossings, applied-condition claims, any E.18 GuardFail events with their gate-assignment facts, any independently defined guard-relation occurrences, optional valuation, preserved and lost transformation structure, non-admissible overreads, and the ordinary stop or reconsideration question.
E.18.3 can organize a transformation-flow slice inside P2W, P2S, work-control, architecture-feedback, evidence, narrative-publication, or refresh situations, but every stronger neighboring claim needs its independently identified value, exact supporting relation, applicable predicate-definition content, and current facts.
A path card, graph expression, route prose, workflow diagram, or demonstrative slice remains a description or teaching slice until both the A.22-selected CGUS and its independent E.18 substrate use are recoverable; the pattern reference adds no connection relation.
E.18:End
P2W Problem-to-Work Carry-Through
Tech-name:
ProblemToWorkCarryThroughPlain-name: problem-to-work carry-through Type: Architectural pattern (E) Status: Stable Normativity: Normative unless marked informative Placement: Part E -> E.18 child pattern Builds on:E.18Transformation Flow Structure,C.22.2ProblemCard@Context,A.6.0U.Signature,A.6.1U.Mechanism,A.3.1U.Method,A.3.2U.MethodDescriptionmembership,A.3.4actual bounded change, the A.15 work family,A.15.PRODlocal production-claim recovery,A.6.RCDblocker boundary and local-claim dispositions,A.6.RELrelation-occurrence and receiving-use discipline,A.6.Prelational precision restoration,A.6.P.WMRwording-to-relation recovery,C.29,C.16,A.19.CPM,A.19.SelectorMechanism,C.18,C.19,F.8,F.18,F.17,F.9,G.5,G.9,G.11,A.20, andA.21. Purpose: preserve selected distinctions from an accepted problem-side record as method selection, planning, performed work, result interpretation, and return become current.
Problem frame
Use this pattern when an accepted ProblemCard@Context can guide work, but the next FPF use is unsettled. Ask which accepted distinction should shape the next question, which relation and participants that question asserts, and what result or stop is needed before the next action.
Method selection, planning, dated work, actual change, result interpretation, and return remain separate continuations. For each current question, apply the pattern that answers it and carry the result it returns or the reason work on that question must stop.
Distinguish the accepted ProblemCard, the subjects of the questions answered by the cited patterns, and the materials that describe those claims. The accepted ProblemCard is the EntityOfConcern of a materialized P2W note. The note is identified under C.2.1 by its ClaimGraph, the accepted card, and its effective reference scheme; its ClaimGraph names the receiving use and designates a separately identified viewpoint or model-use structure only when the claim uses that object and its organization changes how the receiving claim is interpreted.
Each cited PatternID locates the Solution passage needed for the question about its subject EntityOfConcern. A practitioner or another capable system applies that guidance to the project entity or relation—for example, a System, episteme, Method, Work occurrence, or direct relation. When source wording says role, apply E.10.ROLE before treating it as a local system-role kind, separate System-classification judgment, assignment occurrence, participation or functioning relation, ordinary non-use, or missing-governor case.
The compact note, diagram, plan, trace, and publication are epistemes or publication-side values that describe, constrain, or make those claims inspectable. Later method enactment or dated work can change or preserve a subject EntityOfConcern; improving a P2W note or completing its fields does not establish that subject change, work occurrence, evidence, acceptance, or result.
When P2W helps. Use P2W only when keeping the accepted distinction changes which relation you assert, what result you write, or whether you continue, split, stop, or return. If the relation and result are already settled and another P2W note would add nothing to the decision, apply the pattern whose Solution answers the current question.
E.11.PUA covers a smaller use and may begin without ProblemCard@Context: use one selected pattern for one current practical question. Obtain the smallest useful result that answers the question or state why work on it must stop. That ordinary use may stop there; name a receiving use only when the enclosing P2W continuation or another later use is current.
E.18.1 begins only when the wider work-facing continuation depends on preserving accepted problem-side material. PUA may support one pattern inspection inside a P2W flow, but it does not replace the accepted-problem carry-through.
Use this when
- an accepted
ProblemCard@Contextnames a working problem and the team needs a way to choose its next use of FPF toward method, planning, performed work, or result interpretation; - an invariant,
U.Signature(profile=FormalSubstrate),PrincipleFrame, mechanism-position, method-position,A.15.2 U.WorkPlanor plan-item wording cue, performed-work, result-record, or source-currentness cue is present, but the FPF kind or relation to use next is still unsettled; - a transformation-flow structure, mathematical path relation in a graph-shaped description, flow diagram, principle scheme, scenario, functional description, or source publication helps the team think, while the next FPF use still lacks an FPF kind or relation named by value;
- a result artifact, telemetry line, acceptance record, quality-evaluation record, done-state update, feedback pin, or integration claim needs to be unpacked before it can guide the next FPF use.
What goes wrong if missed
The team jumps from a convincing problem-side formulation into downstream language without naming the FPF relation being used. The work then looks responsive to the accepted problem, but the next record is unclear, the result phrase becomes too broad, and the team has no basis for deciding which earlier uses to reopen when a measurement or source currentness changes.
What this buys
The practitioner gets one concrete next move: keep the accepted claim in view, state the question and participants, apply the pattern that answers it, and use the result it returns. Split several relation claims before applying their patterns. If the relation or needed facts are missing, keep the cue and stop. If a relied-on result changes, reopen only the continuation that used it.
Add the compact note only when another person or later action must replay that path. The accepted problem-side distinction remains useful without becoming hidden permission to start work.
Not this pattern when
- there is no accepted problem-side record; use
C.22.2or the problem-side pattern named by value first; - the FPF kind under repair, relation, and record to write are already settled; use the applicable pattern directly;
- the requested output is a local project procedure, schedule, or work-management method; use the relevant work, planning, method, gate, or operational-management pattern;
- the requested record or claim is an evidence case, assurance case, gate record, decision record, architecture description, publication-use claim, or wording-use repair; recover the relation and apply the pattern whose Solution answers that evidence, assurance, gate, decision, description, publication-use, or wording question.
Problem
An accepted problem-side distinction becomes useful when it is ready to guide downstream work or work-planning use. The accepted problem card may expose an invariant, mathematical lens, unresolved functional role cue, mechanism-position candidate, method candidate family, planning constraint, result cue, or changed measurement assumption.
Route unresolved role wording through E.10.ROLE before it affects a continuation; the recovered local system-role kind, classification, assignment occurrence, direct participation or functioning relation, ordinary non-use, or missing governor remains with its own pattern. Without P2W, the team either reduces the accepted problem-side distinction to "we have a solution" or scatters it across several related FPF patterns before preserving it.
First say which accepted claim must affect which decision or use. Then write one ordinary relation-specific question, name its participants, apply the pattern that answers it, and keep that pattern's result or stop. Add a compact note only when another person or later action must replay the path.
P2W succeeds when the accepted claim, receiving use, concrete question, applicable pattern contribution, and result remain inspectable without turning their use-specific connection into a relation kind or treating a note, diagram, plan, trace, or publication as the subject entity or as proof that work occurred.
Forces
Solution
Local P2W mantra. Use this wording to recall the question behind one working decision. Which P2W continuation, if any, is justified now?
What work does the accepted problem require? Choose the next question by what its answer could establish for that work. Apply the relevant method, bring its result or stop back to the original question, and reopen the uses that depend on a changed answer.
Filled cooling use. ProblemCard@Context PC-FAB-042 says that method comparison must preserve the conserved heat-flow structure. The current decision is whether a mathematical-lens continuation is justified. Ask which structure the proposed lens preserves, which it loses, and where its use stops; apply C.29; keep the returned lens-use result.
If the lens subject, declared use, preservation/loss account or stop is unresolved, keep that C.29 question open and do not advance by wording to method selection, planning or Work.
The wording helps a practitioner recall which question to ask. A claim about a Method, MethodDescription, plan, dated Work, Transformation, CGUS or P2W relation requires identification of that object under its own rule. The five rows below display conditional continuations. The applicable relation determines which one to take.
If a selected CGUS already exists, an A.22 demonstrative slice may include this display in its ClaimContent for a declared use. The table itself admits no structure, continuation-row kind, relation occurrence, MethodDescription, plan or Work.
Here and elsewhere in this pattern, move is Plain wording for the current use action or continuation: stating a question, applying the pattern that answers it, keeping its returned result, stopping, splitting, or reopening. In another case it may instead refer to an independently defined recommendation, PlanItem, enabled continuation of a qualified CGUS, dated Work, or actual Transformation. No universal Move object or shared identity connects proposed, chosen, and performed work.
The decision aid below helps the practitioner choose which relation question to answer now. Fill a compact carry-through or replay episteme only when another person or later action must recover the path. The aid shows the accepted claim, concrete question, relation and participants, pattern contribution that answers it, returned result or stop, and the smallest continuation to reopen after a relied-on result changes.
Choose the first of these three levels that lets the current reader act and, when later use requires replay, lets that reader recover the path to the result or stop:
- Ordinary conversational use. State the question and its participants, apply the method that answers it and bring that answer back to the receiving work. An answer about one relation can finish a commission confined to it. Write no P2W note when feedback is fast, the use is local and nobody later needs to replay the path.
- Reliance-bearing use. Add the compact episteme in
4.1when transfer, audit, delayed feedback, expensive reversal, automation, or durable reuse depends on recovering the accepted claim and direct continuation. - Structure-bearing use. Add the selected structure defined and tested by
E.18.3, A.22.CGUS and, for independent members, E.18.NET in4.0bwhen a named use needs that structure's identification or qualification, or relies on it for comparison, publication or a stronger downstream claim. Section4.0bexplains when an ordinary account of branches and conditions already suffices.
Choose a higher level only when its transfer, audit, delayed-feedback, costly-reversal, automation, durable-reuse or explicit-structure need is present.
Choose only the extension the receiving use needs.
No conditional extension may add a mandatory input to ordinary P2W use, change the kind or identity of a result returned by the cited pattern, or mutate the stable core. Omit unused extensions, including their fields.
Assurance scope by use. For a materialized positive episteme, check the accepted ProblemCard edition, carried ClaimGraph slice, decision or use that relies on the result, effective ReferenceScheme, returned result kind and ref, the cited pattern contribution used, and carry-through rationale. Check a separately identified U.Viewpoint episteme or BoundedModelUseStructure only when the claim designates that object and its organization changes how the receiving claim is interpreted.
For a stop use, check that no result was fabricated and that the cue and stop are stated. In both positive and stopped uses, the episteme is about the accepted ProblemCard under C.2.1, not a P2W relation occurrence or RelationSignature.
For practitioner guidance or conformance, verify that a practitioner applying the guidance can obtain the required result or the reason to stop. None of these checks adds Work, transformation, evidence, gate, MethodDescription membership or downstream subject facts.
P2W result without a new relation species
An ordinary P2W application is a practitioner move: preserve one accepted problem-side claim, name the receiving decision or use, ask one concrete question per independent relation, and return each answer or justified stop to the work that required it. When several answers bear on one decision, establish what they warrant together under compatible conditions.
Start with the part of the accepted ProblemCard that can change the answer needed for the next decision or action. State what the reader must preserve and why it matters there. Retain the claim's status and the scope and conditions on which its use depends. For example, an accepted problem formulation can carry a causal hypothesis that needs testing.
Choose the next question from the work that needs the answer. Start with the alternatives the recipient still faces. Which fact about a participant, its arrangement with other participants or the use of a result could change the decision or its grounds? If the practitioner must design that arrangement, leave the choices entrusted to that practitioner open in the task they receive. If a returned result changes the subject or the question, use C.39.QR:4.1 to establish its contribution to the original work. An explained choice of guidance can finish a request for that choice.
If the next action needs a subject result, such as a plan or a measurement, apply the supplying method to the case and obtain that result or identify the exact gap. Use an adequate existing way directly. If the way itself is missing, use C.39 to find or develop it; E.4.CM guides that construction when it must become a maintained framework contribution. When the next decision depends jointly on how a suitable method becomes available, how a performer gains access or capability, and how the job is supplied, C.32.MWA:4 connects those contributions. For a readiness decision alone, use A.15.5 for the intended job.
Bring the answer back to the receiving decision or action. Explain how the carried claim constrained the question, choice or interpretation of the answer. If the local answer leaves a connection required by the larger task unresolved, keep that connection as unfinished work. A stop can settle the present inquiry while leaving the receiving task dependent on further facts or a still-missing way.
For example, C.22.2:20.1a supplies an accepted ProblemCard about two support-policy clauses that use "complete" for different events. Carry that difference and the policy edition into the question of which event the policy means by completion. Resolve the question through the applicable declaration or interface rule; if that rule is unavailable, retain that gap. The observed difference supports clarification, while the proposed causal connection with reopened cases remains a hypothesis. If both clauses are later shown to mean the same event, return to the problem formulation because that comparison no longer supports the ambiguity claim.
This move introduces no ProblemToWorkCarryThroughRelation@Context, reusable predicate definition, RelationSignature, or P2W relation occurrence.
When transfer, audit, delayed feedback, costly reversal, automation, or durable reuse requires the claim to be retraced, use the C.2.1 episteme in 4.1. Its identity is its ClaimGraph, the accepted ProblemCard@Context as EntityOfConcern, and the effective reference scheme.
The ClaimGraph records the accepted card edition and carried claim slice, decision or use relying on the result, concrete question, applicable pattern reference, contribution used, returned result kind and ref or stop, and why the carried content remains relevant. It designates a separately identified U.Viewpoint episteme or BoundedModelUseStructure only when the claim uses that object and its organization changes how the receiving claim is interpreted; neither becomes another identity discriminator.
These are use-specific claim contents, not SlotSpecs of another relation species; the returned result retains its own kind, relation semantics when applicable, identity, and subject-specific basis.
In each case, participants need a grounded claim and a way to retrace how they reached it. Conversation can supply both on the spot. Transfer, audit, delayed feedback, automation or durable reuse may require someone to retrace it later. None of these uses asks whether repeated P2W relation occurrences are the same individual.
The conversational move, optional positive note, and stop description therefore close those uses without a relation-kind candidate. A.6.RCD, E.24, E.24.UK, A.6.0, and relation-species naming do not open.
If a later use asks whether a P2W relation occurrence persists, recurs, ceases, or participates in another relation, reopen A.6.RCD and obtain the direct subject settlement and admission before declaring or instantiating such a kind.
A positive use closes only when the accepted card and carried slice remain current for the receiving use, the cited pattern has returned its result for that use, and the rationale remains a truthful claim in the note or is directly recoverable in conversation. A preceding P2W note may be cited for replay, but it supplies neither occurrence continuity nor a supporting relation; use each relation's defining predicate and identity rule. If these conditions fail, correct the value-kind pair, apply the pattern that answers the question, split the claims, or retain a reduced-use cue and stop.
P2W Declarative Carry-Through Structure
Use P2W as a declarative interface from one accepted ProblemCard@Context claim to results or stops obtained by applying the patterns that answer its continuation questions. P2W preserves the carried claim and receiving use, states the concrete relation-specific question and its participants when relation-like, cites the applicable pattern, and keeps each result or stated stop on a separate continuation. It defines none of the selected relations or results. Refer to the neighbouring guidance without restating it.
Use the following questions to carry the result or blocker returned by the applicable pattern.
When the conditional structure extension opens, recover one admitted A.22-selected CGUS and, under E.18.3, the independently identified E.18 substrate positions, bindings, already-obtaining occurrences, and any relation-reference epistemes needed for replay.
The condition basis—an applied claim with its test and current facts, an E.18 GuardFail event with its gate-assignment facts, or an independently defined obtaining guard-relation occurrence—supports an ordinary stop or reconsideration question; no return relation follows from the pattern reference.
P2W keeps only its accepted claim, current use, one returned result or stated stop, split, and the dependent continuations; it adds no structure field. Plain actions such as carry, recover, write, split, stop, and return guide this P2W use. They are not P2W relation kinds, commitments, permissions, gates, or substitutes for the cited pattern's rules.
Conditional structure extension through E.18.3
First explain which continuations are available, blocked or unknown, using their conditions and the facts of the case under A.22.CGUS:4.1. E.18.3:0 permits a sufficient ordinary explanation of the transformation-flow connections. Stop at that explanation when it answers the receiving question.
Open this structure extension when a named use needs the selected structure's identification or qualification, or relies on that structure for comparison, publication or a stronger downstream claim.
Identify one U.Structure under A.22.CGUS by its independently identified constituents, selected obtaining relation occurrences, applied constraints, and named selection-use frame. E.18.3 qualifies that selected CGUS only when it uses positions, bindings, and already-obtaining occurrences from one independently identified E.18 one-TFS, parent-relative internal-SubflowRef, or E.18.NET substrate. E.18.1 declares no P2W subset schema, wrapper relation, or hybrid record.
Before choosing the structure branch, distinguish three cases. Several FlowValuation values remain valuations of one TFS only while its A.22-selected O/R/C/F remain the same under E.18. A detailed internal portion remains one parent-relative SubflowRef when its positions and selected relation occurrences resolve in that parent, with every required participant retained.
Two or more independently identified TFS or nested-network values connected across their boundaries by already-obtaining relations require one E.18.NET TransformationFlowStructureNetwork; do not flatten them into one giant TFS. Each member retains its boundary. Work, actual transformations, valuations, position bindings and any DesignRunTag stay with the leaf TFS where they occur or apply.
Each nested boundary reference resolves through one finite acyclic memberPath[] to an ExposedFlowPositionRef; each selected cross-boundary claim cites its obtaining occurrence and complete ordered endpoint bindings through one resolvable NetworkCrossFlowRelationRowRef. Membership is acyclic; a feedback relation may cycle only when its predicate and occurrence facts permit that cycle.
If explicit structure is not required, use the stable conversational core. If replay but not structure is required, use 4.1.
If the next question is work-facing, apply the A.15 family before claiming a plan, readiness, launch or dated U.Work. Use A.3.4 for each actual transformation and A.15.PROD only for the exact production-work, identity-inception or completion claim currently made.
G.11 handles source currentness; E.18:5.13 covers all dependent uses through the needed PathSlices; E.18.NET retains independent members and cross-member occurrences. P2W reopens each affected application and dependent continuation, preserving independent uses.
Compact carry-through episteme (conditional reliance extension)
Open this extension only when transfer, audit, delayed feedback, expensive reversal, automation, or durable reuse requires someone to replay the stable P2W core. Materialize one ordinary C.2.1 episteme whose EntityOfConcern is the accepted ProblemCard@Context, whose ClaimContent is the current positive or stopped carry-through account, and whose effective ReferenceScheme governs its designations. Carry-through note and stop description are Plain use labels for those two ClaimContent shapes, not local U-kinds or relation species.
A positive use fills the returned result and leaves honestStop absent. A stopped use fills the returned blocker or reduced-use cue and fabricates no positive result.
When the claim uses a separately identified U.Viewpoint episteme or BoundedModelUseStructure and that object's organization changes how the receiving claim is interpreted, the ClaimContent designates it; otherwise no surrogate field is filled. Citing a PatternID does not by itself admit a U.MethodDescription. The episteme, its claim and its predecessor pointer are neither a reusable predicate definition nor a P2W relation kind or occurrence.
A positive use is well formed only when the named result kind is one that the cited pattern returns for the stated question, the carried ClaimGraph content remains relevant to the receiving use, and every independent continuation stays separate. When the result is a relation occurrence, assertion, or description, cite that exact object; keep its obtaining or claim basis, occurrence-identity rule, and any receiver-conditioned reusable declaration or typed SlotSpecs with that returned object under the pattern content that defined or tested it. A predecessor pointer supports replay only and establishes neither episteme continuity nor another occurrence.
For first-minute use, state the question, apply the pattern that answers it, and continue with its result or stop without materializing this episteme. Materialize it only when replay is required.
Each continuation names one applicable pattern reference and one question or use that its result answers. Do not combine value kind and relation signature, method and mechanism, evidence and assurance, plan and dated Work, actual Transformation and production, or refresh and residual triage in one field.
The use closes positively when the cited pattern has returned its positive result and the carried problem-card claim remains visible in that result or its stated basis. It closes by bounded stop when the cited pattern returns a blocker or reduced-use cue and no positive continuation can be stated.
Conditional development-loop relation-selection extension
Open this didactic extension only when cheap generation, open-ended search, or evolutionary-engineering work has produced many variants before the project has a stable problem, comparison basis, selected set, work entry, or currentness relation. Apply the unchanged P2W core to the one question that changes the next action. Use the four rows to distinguish these cases. Once you have stated the question, use the single map in Relations.
If the current question is still problem formulation or opportunity, return first to C.22.2 to accept or revise the ProblemCard@Context and its carried claim. That is an upstream return, not another downstream P2W result.
Cheap variant generation shifts effort toward problem production, characterization, archive stewardship, fair comparison, explicit choice, autonomy boundaries, evidence, assurance, performed work, effect measurement, currentness, and repair. P2W preserves the accepted problem-side claim while one of those relations becomes current; an archive, front, selected set, confidence phrase, or choice rule supplies neither an A.2.8.PER permission result nor performed work.
Source wording such as trust budget, problem factory, solution factory, or factory of factories remains a project label until the evidence, assurance, autonomy, work-organization, or other direct relation is named.
Conditional development-for-developed first-minute extension
Open this didactic extension only for a fast DPF seed, and keep the source-use and hardening continuations distinct. An accepted problem-side record may cite a G.2 source-use relation, selected source U.Episteme, exact EpistemePublicationRelation occurrence reference when availability is material, source-pack cue or return, and provisional framework purpose.
If choosing a DPF, an access-only route, or stop must settle a downstream-used framework boundary, use E.4.PFAD to profile that framework-specific content in one E.9 DRR; a cheap seed or route that settles no such boundary stops without that DRR.
State each material initial pattern relation with the predicate that defines it, and use E.4.PFR only when a named maintenance use needs relation records. Using the E.4.PFAD profile adds no second decision or decision record.
If the receiving use needs a way that the sources do not yet supply, develop it through C.39; for a framework contribution, use E.4.CM to develop the connected Methods and their explanation before allocating them to patterns. An adequate supplied way can be used directly.
Use E.8 for pattern authoring, E.21 for evaluation, E.23 for improvement, and G.11 for currentness or refresh. Keep the source result, selected answer and DRR, direct relation assertions or optional records, authored patterns, quality results, and currentness results separate.
When the receiving question is which guidance to use next, an explained selection can close that question. When the receiving work needs the subject result, keep the carried claim through application of the selected guidance to the result or stop.
Cooling-module example. ProblemCard@Context PC-DEV-041 states that cheap generation produces many cooling-module layouts while fair problem framing and comparison remain weak. The carried claim is that the current candidate set retains maintainable low-energy variants until energy use, service access, manufacturability, thermal margin, and test cost are represented in the current characteristic and comparison relations.
A C.18 archive and front are current. A.19 defines the characteristic space and its comparability boundary; A.19.CPM comparison becomes current only when that characteristic space and comparator are current. The G.5 selected-set result declaration remains stopped until that comparison and front are current; actual audience availability is a separate later question that uses E.17 for a source-backed publication face and return to source and E.24.PUB for the publication occurrence, form, carrier, audience, bounded use, and availability.
An E.16 generator boundary may separately bound search and test spending. Recover the prototype observations and their support for the receiving claim. Use A.10 when that source-to-use account is implicit or must remain replayable; use B.3 for assurance-sensitive confidence use.
A C.30 architecture-candidate relation appears only for retained layouts that change selected structure. No U.WorkPlan has yet been produced under A.15.2, and no dated U.Work occurrence has yet been admitted under A.15.1.
Thermal and serviceability measurements can feed but cannot create three separate results: applying A.15.5 may return WorkEntryReadiness@Context for one named intended-work concern; applying A.21 may return a GateDecision only for one current OperationalGate(profile) and its declared checks; applying A.2.8.PER may return one named non-prohibition, granted-permission, permission-exercise, non-violation, or permission-conflict result with its required participants and basis.
An actual release action is an A.15.1 U.Work occurrence; a further claim that a subject was released needs its named subject predicate and participants. No predicate definition or occurrence rule for that release relation is current in this example, so an approved, authorized, or released cue stops as missing-governor for that attempted use rather than inheriting the measurement, readiness, or gate result.
When descriptors, tests, competitor information, or cited publication editions change, reopen the currentness-dependent continuations under G.11.
Which retained layouts belong in the current front? Apply C.18 to obtain its current front record. Architecture comparison, selected-set result declaration, actual publication, planning, and work are possible later continuations, not alternative fillers of one field.
Conditional naming and publication extension
Ordinary P2W use skips this extension. Open it only when a pattern author, publisher, trainer, or tool builder must cite the already defined practice outside its local use. The header's Tech/Plain pair identifies this pattern for readers: ProblemToWorkCarryThrough / problem-to-work carry-through. It does not classify a U.Method, U.MethodDescription, relation, Work or result. If the use requires MethodDescription membership, first identify one C.2.1 episteme, require one independently admitted U.Method as its EntityOfConcern, and apply the A.3.2 substantive way-of-doing claim threshold.
The compact positive, stop and replay shapes in 4.1 and 4.8 are local ClaimContent uses of ordinary C.2.1 epistemes. Their field labels are local phrases, not reusable U-kinds, NameCards or term rows.
Cite the settled Tech/Plain designation directly when its existing meaning serves the receiving use. Use F.8 for an unsettled or stronger naming need; apply F.18 for the durable designation and F.17 for a public scheme-local row when those uses require them. F.9 opens only if two independently identified scheme-sense cells require an exact Bridge. No Bridge is current merely because two readers use similar P2W wording.
Keep the practice, pattern episteme, any admitted Method, any qualifying MethodDescription episteme, local carry-through episteme, publication occurrence, publication form and presentation carrier separate. Naming or publication admits none of them and adds no stable-core field.
Reopen only what the change affects: a changed practice or practitioner use reopens E.18.1; changed wording reopens F.18; changed public reader use reopens F.17; changed source basis reopens its exact source-use relation. For a changed publication, use E.17 for the source-backed face and return to source and E.24.PUB for the occurrence, form, carrier, audience, bounded use, and availability. A source phrase or remembered title supplies no source-to-use relation, authority, evidence, result or performed Work.
Positive carry-through: one executable first use
Use the cooling case to see which conclusion a calculation of the whole can support and where the original comparison still needs work. The fourth row applies if the source also asserts a signature. For another subject, use the branch rule in 4.6 and the relation-selection map for the questions that subject raises.
In this example the ordinary route uses one carried distinction, one concrete question, one map lookup, one result from the pattern that answers the question, and one visible stop.
A case with several claims splits before any pattern is applied; a case with only a cue stops under 4.6.
Direct-relation distinctions that change the branch
P2W carries the value or stop returned by the neighboring pattern. Apply that pattern’s test there, without restating it in P2W. Keep a local distinction here only when it changes which question the reader asks:
- Lens or declaration? Ask whether the current use judges a mathematical representation or declares a signature. Split the claims when both are present; the first-use case in
4.2shows the difference. - Mechanism or method? Ask whether the claim concerns a law-governed operation application or a reusable way of doing. A shared noun supplies neither; split the questions and use the Relations map once for each current claim.
- Change or timing? Ask whether the claim concerns an actual bounded change, a temporal aspect such as an interval or cadence, or the adequacy of a temporal claim for one use.
- Work, change, or their connection? Identify the dated
U.Workoccurrence and actualU.Transformationseparately, then ask whether a work-to-change claim is current. Apply the pattern that defines or tests that claim, or the applicable A.6.RCD route, and carry only its positive or negative result or blocker. Shared timing does not answer the question. The BuildOps and Pump 14 slices in5.1show a positive result; Pump 14 also preserves an earliermissing-governorstop without copying the result's proof. - Approved, ready, released, or permitted? State which result is being sought: a gate decision, permission result, work-entry-readiness result, release
U.Workoccurrence, or subject-release relation. Apply the pattern that answers that question and carry its result or blocker;authorizationis not a result type. - Result or production? Let
A.6.P.WMRseparate the concrete result questions. OpenA.15.PRODonly for a production-work, entity-inception, or production-completion question; its returned claim or blocker stays separate from work, change, delivery, acceptance, and release.
For every other exceptional object, state the relation-specific question and consult the canonical map in Relations.
Boundary and relation discipline
A neighboring pattern's detail appears outside Relations only when one local discriminator in 4.3 or one worked case needs it to choose, split, or stop. Use the branch rule in 4.6 and the single question-to-pattern map in Relations. Neither place restates a neighbour's occurrence basis, recovery algorithm, production criterion, derivation method, or admission law.
A local P2W application closes positively when a practitioner or another capable system has obtained or amended the result by applying the cited guidance and the carried distinction remains visible in that result or its stated basis.
It closes by bounded stop when no continuing relation can be recovered and the reduced-use cue plus stop condition are stated. A following method selection, planning act, work occurrence, evaluation, or other use of neighboring pattern content is not unfinished P2W work.
A wider P2W carry-through slice remains current only while a named downstream receiving use relies on the accepted problem-side distinction. It closes when no remaining receiving use relies on that distinction and no return condition is current. A later changed assumption opens a local return to each application that depends on it, continuing through dependent branches and joins. Earlier applications with independent grounds remain closed.
Return and refresh rule
Reopen the relation that supplied the changed value, then only the continuation that relied on it. Do not replay the whole carry-through.
A dated occurrence already admitted as U.Work remains the same world-side occurrence. Return may change a later interpretation or plan; it does not rewrite that occurrence retrospectively.
Plain relation-selection branch
Put the unsettled question into one ordinary sentence: "Did this work change that pressure here?", "Does this grant let this technician do this work now?", or the equally concrete sentence for the current case. Name the participants and relation, then use the matching row. Do not scan every pattern first.
A question about one relation can close on the first row. The Relations map helps locate the relevant pattern or check an exceptional branch.
Lowering and reopen block
Lower only the claim that cannot be made. Keep any independently grounded value and preserve the practical question that would reopen the branch.
Conditional reliance replay after a relied-on value changes
Open this extension only when transfer, audit, delayed feedback, costly reversal, automation, or durable reuse requires a durable account of what still follows after source-currentness repair, appearance-based reliance repair, changed measurement, changed problem-side record, FPF pattern change, or a use-found defect. For an ordinary local return, use 4.5 without creating a replay episteme.
Materialize one ordinary C.2.1 episteme whose EntityOfConcern is the accepted ProblemCard carried by the original carry-through episteme, whose ClaimContent is the replay account below, and whose effective ReferenceScheme governs its designations. Replay note is Plain wording for this use, not a local U-kind, refresh process, change log or authority record.
Reapply the guidance recorded for the earlier use before filling the replay episteme, then use the basis it supplied or located with current facts to reassess the earlier claim or result about the independently identified changed value.
If the changed object is a relation, that reassessment first judges whether the relation obtains; apply A.6.REL afterward only when relation-occurrence identity is current. The earlier result keeps its participants, obtaining or claim basis, occurrence-identity rule and any reusable RelationSignature or typed SlotSpecs.
The replay account records only what still follows, what no longer follows and which P2W continuation reopens. Citing a PatternID does not admit a MethodDescription.
P2W may cite a readable relation assertion, an explicitly individuated occurrence, or a typed assertion or description, but it cites the independently identified object that served as the earlier result together with its obtaining or claim basis. Citation does not make relation use signature-dependent; a receiving episteme carries a signature reference only when the defining content for that claim requires one.
The changed object may instead be a source edition, measurement, unit, reference plane, Method set, comparator, module-interface relation, publication-use relation, problem record, or FPF pattern publication.
Whatever changed keeps its own kind. If the reopened use depends on maintenance, responsibility, or authority, name the direct relation and its participants; an owner-shaped label is not enough. Reapply the guidance used for the earlier result. Add a G.11 line only when one exists.
The practitioner or another capable system applies the guidance to the reopened question and decides whether to continue, stop, split, retain a reduced-use cue, or return upstream.
Archetypal Grounding
Seal-failure carry-through
A maintenance team has an accepted ProblemCard@Context for recurrent seal failure. It records the operating conditions, the distinction between thermal deformation and material degradation, and the observations that would challenge that distinction. The team uses E.18.1 because diagnostic-method selection, repair planning, dated repair work, interpretation of the post-repair measurements, and return after a changed diagnosis all depend on preserving these accepted problem-side distinctions.
E.11.PUA may help the team inspect and apply one diagnostic-pattern candidate inside this flow. Its result might be one fit finding or one diagnostic method-selection input. That smaller result does not replace the accepted problem material, the repair plan, the repair work, or the later interpretation and return relations.
In System-facing work, an accepted problem-side record may lead toward method choice, planning, performed work, result records, and result measurement. In Episteme-facing work, the same record may lead toward a U.Signature(profile=FormalSubstrate) declaration, mathematical-lens use, description, publication, evidence, or gate-related claims. Both uses ask the same P2W question: which FPF kind or relation can carry the next claim being made?
Worked slices
Each slice shows only the P2W contribution: the carried distinction, current question, pattern that answers it, independently obtained result or blocker, and next continuation or stop.
-
Thin first-principles start. The accepted card says that a conserved structure, not one more tuning defect, matters to the next decision. The current question is a mathematical-lens question, so the practitioner applies
C.29and carries its lens-use result or stop. A separate declaration question goes toA.6.0; method selection waits for its own question and participants. -
Planning from a selected-enough method. The carried distinction constrains planning and the current question asks for a plan. The practitioner applies
A.15.2and carries the plan result returned there. Any compact P2W note cites that result and the problem-side claim it preserves; the plan keeps its own content and authority. -
Performed work and a positive store-change connection. The carried claim makes actual population of the artifact-store partition material to the next release question. The current question asks whether
ReleaseBinary12_BuildWork_2026-07-21T0900_0912is connected toArtifactStorePopulationTransformation_12.A.15.1andA.3.4identify those two participants, and applying the BuildOps work-to-change predicate yields positive assertionBuildWorkPopulatedStore-12. P2W carries that assertion and continues to the separate release question; if the defining pattern instead yields a blocker, P2W stops there. It does not recreate the predicate test, performed-application proof, or negative-claim rule. -
Result interpretation without a generic result. The sentence the work result proves the approach worked leaves the result question unresolved. Apply
A.6.P.WMR; carry each concrete returned claim or blocker on its own continuation. No returned item becomes a generic result or production value merely because the source used the word result. -
Functional explanatory order. A source diagram places formal declaration, principle framing, mechanism, normalization, method selection, planning, performed work, and measurement in one readable order. Treat each as a possible question, apply its defining pattern only when that question is current, and carry the separate returned values or stops. Display order supplies no project sequence or authority.
-
Interface split before P2W use. A source says a port-throughput limit makes a solution feasible after integration. Ask the module-interface question through
A.6.Mand the selected transformation-flow question throughE.18. Planning, work, evidence, gate, function, and architecture cues remain stopped until their own questions are stated. P2W carries only the result or stop that matters to the current decision. -
Measurement returns to planning. A source says one work occurrence produced telemetry and an artifact.
A.6.P.WMRfirst returns the distinct artifact, telemetry, production, or unsupported results needed by the case; P2W keeps separate continuations. If a laterC.16result andG.11currentness result change the reference plane used by planning, reopen only the planning, method-comparison, or problem-side continuation that relied on it. The earlier dated Work occurrence is not rewritten. -
Pump 14 pressure adjustment; positive continuation after an earlier stop. The carried distinction makes the relation between
W-P14-ADJUST-1010-1020andT-P14-PRESSURE-RISEmaterial. In the earlier case record, no current predicate could state that connection, so applying the defining work-to-change pattern yieldedmissing-governorand P2W stopped.In the current record, each precise performer has an independently established A.13 core and A.15.1 has independently admitted the Work. Because this record also carries exact assignment-bound attribution, F.6 afterward establishes that relation through the same obtaining assignment. That basis and the
P14-REL-2026application support the positiveAdjustmentWorkCausesPressureRiseresult; P2W carries the result without reconstructing the agency, Work-admission, assignment-attribution, transformation, or causation proof.The separate claim that
PC-P14-PRESSUREguidedWP-P14-2026-07-15still has its ownmissing-governorresult and stays stopped. Later measurement and decision questions remain separate.
Additional worked situations
Pilot examples for transformation-flow structures and networks
Distinguish three cases in the examples: several valuations or path slices of one TFS; one parent-relative internal SubflowRef; or an E.18.NET network of independently identified TFS or nested-network members connected by already-obtaining cross-boundary relations. A diagram, common product, display order, shared Work or source wording decides none of them.
For one TFS, every valuation uses E.18:5.2’s resolved connection/cut domain while retaining the same A.22 selection O/R/C/F. A new actual application or selected relation occurrence first reopens that identity.
In a network, each member retains its boundary. Keep Work, actual transformations, valuations, position bindings and any DesignRunTag at their respective leaf TFSs. Cross-flow occurrences retain their defining predicates, signatures, participant order and endpoint bindings. Membership is acyclic; feedback relations may cycle when their own rules permit it.
Use a pilot to check the carried object's exact member-local position, the direct relation that crosses a boundary when one exists, and every dependent member or continuation. When two branches use the same changed calibration, the pilot covers both branches and their join; the first detected slice is only an address into that work.
Filled P2W carry-through notes
These examples show how to fill and replay the compact note in 4.1.
Cooling-loop mathematical-lens continuation.
Port-throughput continuation split.
Bias-Annotation
These five lenses were tested: Gov, Arch, Ontological and epistemic, Prag, Did. Scope: accepted problem-side record plus carried distinction moving toward FPF applications.
- Governance bias (Gov): permission, gate, release, assurance, and decision cues remain local cues until the relation and participants are stated: an
A.2.8.PERpermission result,A.21 GateDecision,A.15.1releaseU.Workoccurrence plus any required named subject release predicate,B.3assurance result, or direct decision result. The wordauthorizationsupplies none of them. - Architectural bias (Arch): diagrams, selected structures, and module-interface language help formulate the next relation question; they do not replace the accepted claim, receiving use, separately identified viewpoint or model-use participant, applicable pattern contribution, or returned result.
- Ontological and epistemic bias: a source publication, diagram, compact note, or formal declaration remains separate from the subject EntityOfConcern and from the relation or result claimed through the pattern contribution used for the current question.
- Pragmatic bias (Prag): the carry-through structure is useful for action without becoming a prescribed project procedure.
- Didactic bias (Did): the local P2W mantra and positive carry-through structure show how to use P2W before the reader needs the more detailed relation aids.
Conformance Checklist
CC-E18.1-1The P2W use starts from an acceptedProblemCard@Contextor stops before P2W begins.CC-E18.1-1aThe accepted ProblemCard as the note's EntityOfConcern, the note's ClaimGraph and effective ReferenceScheme, any separately identifiedU.ViewpointorBoundedModelUseStructuredesignated by that ClaimGraph, each cited pattern's subject EntityOfConcern, and every supporting compact note, diagram, plan, trace, or publication remain distinct. Note completeness does not prove a P2W relation occurrence, subject change, performed work, evidence, acceptance, or result.CC-E18.1-1bEvery materialized carry-through episteme identifies one accepted ProblemCard as EntityOfConcern, carries one ClaimContent for the receiving use, names its effective ReferenceScheme, and designates a separately identifiedU.Viewpointepisteme orBoundedModelUseStructureonly when the claim uses that object and its organization changes how the receiving claim is interpreted. It cites the carried ProblemCard slice,applicablePatternRef, returned value kind and ref or a justified stop, and rationale. It introduces no reusable P2W predicate,RelationSignature, relation kind, local note kind or occurrence.CC-E18.1-1cWhen external naming or publication is current, F.8/F.18/F.17 apply only to the already identified value and receiving use. The pattern label and local positive/stop/replay phrases create no NameCard, U-kind, relation, Method or MethodDescription. Any MethodDescription claim separately passes the A.3.2 EntityOfConcern and substantive-claim threshold.CC-E18.1-2A positive carry-through ClaimContent cites one returned result and one or more separate continuation descriptions. A stopped ClaimContent instead states the reduced-use cue or blocker and stop without fabricating a relation. Local non-overread and return conditions appear when relied on; absent fields are not filled by generic unions.CC-E18.1-3The stable core works without a separate carry-through episteme or explicit structure: accepted claim, receiving use, concrete question, applicable pattern contribution, returned result or a justified stop, split, and smallest local return. When explicit structure is needed, A.22.CGUS and E.18.3 select the exact structure; E.18 keeps several valuations or one internalSubflowRefon one TFS; E.18.NET keeps independently selected flows or nested networks and exact cross-member occurrences. E.18.1 adds no hybrid schema.CC-E18.1-4One wording span from an admitted source may split into several FPF applications; the record does not compress them into one generic token.CC-E18.1-5Result wording is unpacked into concrete result-related relations; a genericWorkResultkind is not admitted.CC-E18.1-6PrincipleFramereferences keep postulates and CHR observability distinct from units, planes, comparators, thresholds, ontology editions, CHR editions, plans, work, evidence, and gates.CC-E18.1-7Measurement,G.11source-currentness relation, reference-plane, method-set, comparator, or problem-side changes return to every application, branch and join that depends on the change.CC-E18.1-8The stable P2W core contains only accepted claim, receiving use and concrete question, applicable pattern contribution, returned result or a justified stop, split, use of the supported answers for the receiving decision, and local return. Reliance notes, explicit E.18.3 structure, development examples, and naming or publication are optional extensions. No extension may add a core input or change a returned result. Relation obtaining and identity, occurrence declarations, admission, production, evidence, gates, decisions, and other neighbouring algorithms remain in the patterns whose Solutions answer those questions.CC-E18.1-9Local boundary wording remains only where it names a near-miss that changes the next P2W application.CC-E18.1-10The pattern leaves one usable next move: apply the pattern that answers the question and use its result, write a compact note when another person or later action needs replay, split independent claims or establish how several answers support the same use, keep a cue and stop, or reopen only the continuation affected by a changed relation.CC-E18.1-11For a structure-bearing use, identify what the receiving question needs from the structure. Resolve its values as valuations of one TFS, a parent-relative internalSubflowRef, or an E.18.NET network of independently selected members and cross-boundary occurrences, as applicable. Preserve each member’s boundary. Keep Work, actual transformations, valuations, position bindings andDesignRunTagvalues at their respective leaf TFSs. Section5.3illustrates the distinction. A request to replay one of those pilots can close on that pilot. The self-evolving-spec case keeps use-found evidence outside practitioner-facing prose.CC-E18.1-12Every carried claim family can be lowered, stopped, split, or reopened throughE.18.1:4.7; a cue from a wording span in an admitted source or from a source-pack cue that cannot name the recovered FPF kind or relation remains a reduced-use cue.CC-E18.1-13Every materialized replay identifies the changed value, occurrence, assertion, or description; its kind andchangedValuePatternRef; what still carries and what no longer carries; the smallest reopened continuation; any currentG.11currentness line; andnextApplicablePatternRef. If the changed object is relation-bearing, the cited result—not a P2W copy—retains its kind, participants, obtaining or claim basis, occurrence-identity rule, and any receiver-conditionedRelationSignatureor typed SlotSpecs.CC-E18.1-14When a generated DPF seed or cheap framework seed enters P2W, the record names theG.2source-use record, selected sourceU.Epistemereference, exactEpistemePublicationRelationoccurrence reference when availability is material, source-pack cue, or source-pack return when that source use is current; the problem-side cue when that is current; the next concrete claim or relation-specific question, including its participants when a relation is asserted; the next applicable pattern selected from the canonical Relations map; and the stop condition that prevents the seed from becoming public authority by generation alone.CC-E18.1-15An actual-transformation continuation carries only a current value or blocker returned byA.3.4; E.18.1 does not reconstruct the occurrence basis or infer actuality or composition from a method, plan, model, description, flow position, adjacency, shared work, or common referent.CC-E18.1-16A work-to-change continuation cites the positive or negative claim or blocker already returned by the named subject predicate's defining pattern or the applicableA.6.RCDroute. It keeps the actualU.WorkandU.Transformationreferences when they are part of that result, but does not reconstruct occurrence proof, predicate tests, failure classification, or negative-claim closure. The BuildOps and Pump 14 slices show positive carry-through; Pump 14 also preserves the earliermissing-governorstop. A production continuation likewise carries only the result or blocker returned byA.15.PROD.CC-E18.1-17A PatternID reference, selected or recommended continuation, imperative wording, intended realization, plan seed, graph or filled table admits noU.MethodDescription. Membership exists only for an independently identified C.2.1 episteme whose EntityOfConcern is one admittedU.Methodand whose ClaimContent contains at least one substantive way-of-doing claim.CC-E18.1-18Move remains Plain wording for the exact current object or use action. Proposed or chosen work remains distinct from dated performed Work; no universal Move kind, record or relation is introduced.CC-E18.1-19The wording of the local mantra in §4 recalls one decision. Each phrase refers to an independently identified value, as illustrated in the cooling case. The practitioner returns that answer or missing basis to the decision that motivated the question. A bounded stop on one branch leaves independently answerable branches available. The five-row table displays the possible continuations.
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
E.18.1 is a child of E.18 because a P2W use may need transformation-flow structure when the accepted claim spans several slices, typed positions, or returns. It does not define graph semantics or prescribe performed-work order.
It helps a practitioner keep the accepted claim visible while selecting the pattern whose Solution answers the next question. P2W preserves the carried claim; the practitioner or another capable system obtains or amends the downstream result by applying the neighbouring guidance.
SoTA-Echoing
The table identifies the comparators for specific P2W moves in the assessment dated 7 August 2026. Each row states the adopted contribution and its use limit.
The synthesis that combines these moves into one P2W carry-through discipline is an FPF-scoped architectural hypothesis, not established SoTA.
The sources support the problem-first, relation-separated, replayable moves named in their rows; they do not establish that P2W is a universal workflow or that one carry-through claim is sufficient for every downstream claim.
The hypothesis concerns the passage of one accepted problem-card claim to a result or justified stop for an identified use. Independent questions can be answered separately. When one decision needs several answers together, the practice that makes that decision must supply the grounds for combining them.
The Jiao article, QD survey, manufacturing digital-thread papers, historical Modelica 3.7 specification, and cited Dyad component and analysis documentation are publication or practice anchors. The Dyad comparison concerns separation of component models, analysis definitions and execution; Modelica remains historical lineage only. At the time of that assessment, the DGM paper provided a recent system result. The 2026 EvoTrace and harness papers were then preprints, with the uncertainty that status carried. Reopen these adoptions when stronger studies change problem-first method selection, distinguish generated structural novelty differently, revise evaluator-hack controls, alter QD archive semantics, or show that digital-thread continuity warrants a stronger use than the exact direct relation currently supports.
Relations
-
Apply
A.22.CGUSwhen P2W identifies one A.22 structure whose local loci, selected relations, applied constraints, and at least two potential continuations are recoverable. Judge enabled, disabled, unknown, and error outcomes for the present case separately from structure identity and membership. -
E.18.3qualifies that A.22-selected CGUS through positions, bindings, and already-obtaining occurrences from one independently identified E.18 substrate. E.18 defines the one-TFS and parent-relative internal-SubflowRefinterfaces; E.18.NET defines independently identified network members and obtaining cross-member relations. P2W cites those values, adds no subset, reciprocal record, or hybrid structure schema, and neither reidentifies nor routes them. -
Before hardening a source-derived DPF seed, recover the needed source contribution and its limits through the sufficient source route in
E.4.DPF:4. UseG.2when that requires a broad, refreshable CG-Frame pack across competing traditions. AddA.10when the source-to-use account is implicit or must remain replayable,G.6for addressable path citation or shared provenance representation,B.3for assurance of a named reliance use, andG.11for currentness and refresh only when that stronger use is current. -
E.4.DPFguides DPF authoring;E.4.CMdevelops the connected Methods and their explanation from the field's working questions before they are assigned to patterns. When the framework-architecture question is live,E.9records the selected answer andE.4.PFADprofiles its framework-specific content;E.4.PFRhandles an optional framework-relation record only when a named maintenance use needs one. -
E.23defines and tests repeated quality improvement only after the object version and evaluation are recoverable; P2W may carry a seed to that point but does not become the improvement method. -
G.11defines and tests currentness, admitted-source decay, source-use relation change, edition change, and refresh when a changed source publication, source-use relation, or telemetry reopens the P2W applications that depend on that change and their consuming continuations. -
E.18defines and tests the selectedTransformationFlowStructure, its participant positions and resolved connection positions q, flow valuations, crossings and local state bindings. Each q resolves through rho to one selected internal relation occurrence r; establish r under its direct predicate and occurrence-identity rule. ApplyA.20for a current internal constraint-validity claim andA.21for a current gate profile application, check mapping or gate decision; cite those separate results only when the P2W continuation uses them. -
C.22.2defines and tests the accepted problem-side record and problem-side claims related to the carried distinction. -
A.6.Psupplies recovery and readable statement of each direct relation.A.6.RELdefines and tests direct obtaining, occurrence individuation, and receiver-conditioned use of any reusableRelationSignature; P2W cites the occurrence, assertion or description returned there without repeating its defining rules in ClaimContent. UseA.6.RCD,E.24, andE.24.UKfor any later P2W relation-kind candidate and admission, whileA.6.0declares aRelationSignatureonly after that settlement. F.8/F.18/F.17 open only when an external naming or publication use is current; the header's Tech/Plain pattern label and local note-field phrases create no NameCard, term row, U-kind, relation or MethodDescription.
Canonical question-to-pattern map. Read each row independently. The row order is not a declaration or work sequence, and each pattern keeps the definition, test, and basis of the result it returns.
E.18.1:End
Transformation Flow Mathematical Description
Tech-name:
TransformationFlowMathematicalDescriptionPlain-name: mathematical description of a transformation-flow structure Type: Architectural pattern (E) Status: Stable Normativity: Normative unless marked informative Placement: Part E -> E.18 child pattern Builds on:E.18Transformation Flow Structure,E.18.NETNetwork of Transformation-Flow Structures,C.29Mathematical Lens Use,C.2.1U.Episteme,E.17publication machinery,A.3.4U.Transformation,A.6.0U.Signature,A.6.5slot discipline,A.15work family,A.20,A.21, andC.30architecture family. Purpose: record how a mathematical expression describes exactly one selectedTransformationFlowStructureorTransformationFlowStructureNetwork@Context: what it represents, what it preserves, what it loses, which declared use it serves, and which relation a stronger project claim asserts and how to test that claim.
Problem frame
Use this pattern to describe exactly one selected subject mathematically. That subject is a transformation-flow structure or a network of such structures. The description may focus on an independently identified part of that subject. Section 4.2 distinguishes expression families by the structure and use they expose.
The primary EntityOfConcern of this pattern is the mathematical description TransformationFlowMathematicalDescription@Context: a C.2.1 U.Episteme specialization whose described ontic subject is exactly one selected TransformationFlowStructure under E.18 or one selected TransformationFlowStructureNetwork@Context under E.18.NET.
The one-TFS and network reference branches are mutually exclusive; CandidateMathObject, ExpressionKind, MappingMode, PreservedStructure, LostStructure, and DeclaredUse are fields of the description record, while PublicationFaceRef? references a separate E.17 publication face.
A single diagram can present several of the five values below. Identify which value the claim concerns.
When the described selected structure is one A.22-selected CGUS qualified under E.18.3 through an independently identified E.18 substrate, E.18.2 defines only the mathematical description.
A graph, path expression, category object, algebra, tuple, or matrix may describe substrate positions, crossings, and condition labels, but the expression does not decide whether a condition is an applied claim, an E.18 GuardFail event, or an independently defined relation occurrence.
The expression may also describe preserved or lost structure, supporting relations to independently identified neighboring values, and stop or reconsideration questions; those claims remain in TransformationFlowMathematicalDescription@Context or a C.29 lens-use claim. The expression does not become the selected CGUS or its substrate and does not carry method, work, evidence, architecture, publication, or refresh authority.
Use this when
- you need a mathematical description of one selected
TransformationFlowStructureor one selectedTransformationFlowStructureNetwork@Context, possibly focused on an independently identified part of that subject; - a diagram or equation set helps compare composition, decomposition, coarser/finer partitioning, internal transfer, crossing, or refresh inside one TFS, or cross-member relations in one selected network, but the mathematical expression must not by itself authorize work;
- a source says "graph", "network", "path", "morphism", "algebra", "category", "workflow", "pipeline", "dataflow", or "functional diagram" and the claim being made is the mathematical description of one already selected TFS or TFS network;
- a reader needs to decide whether the visible object is one E.18 TFS, one E.18.NET network, an E.18.2 mathematical description, a C.29 lens-use claim, or only an E.17 publication face.
What goes wrong if missed
A project source expression, source publication, or diagram can make a graph-shaped expression look like the flow structure itself. The team may then take mathematical neatness as evidence or as grounds for claims about work completion, gate readiness, architecture adequacy, or permission to act. The opposite error is also common: every graph-shaped structure is demoted to "just a diagram", so the selected structure, its slices, and its refresh boundaries disappear.
What this buys
The practitioner can tell which inferences the mathematical expression supports. The record names exactly one represented E.18 TFS or E.18.NET network, the expression used, what the expression preserves, what it loses, the declared use, and the result returned after applying the pattern whose Solution answers any stronger claim.
Not this pattern when
- one selected transformation-flow structure is the EntityOfConcern; use
E.18; - one selected network of independently identified TFS or nested-network members is the EntityOfConcern; use
E.18.NET; - one A.22-selected CGUS whose E.18.3 qualification uses an independently identified E.18 substrate is the EntityOfConcern; use
E.18.3; - one bounded transformation is the EntityOfConcern; use
A.3.4; - the claim is general mathematical-lens adequacy outside transformation-flow structures; use
C.29; - the claim is a publication face or view publication; use
E.17and the relevant view or architecture-description pattern; - the claim is work planning, performed work, evidence, assurance, gate fit, gate decision, release, decision, or architecture adequacy; use the applicable row in §4.4 and keep the plan, Work, evidence relation, assurance result, gate result, release claim, choice, or architecture result returned there.
Problem
Transformation-flow structures are often easiest to inspect through mathematics. A graph can show dependency and reachability, a category can show composition, and a quotient can show coarser structure. A fold can reveal aggregation and a refinement can reveal lost detail. A wiring expression can show interface placement, and a tuple can make slot positions explicit.
Those expressions are useful because they preserve selected structure while ignoring other structure. If the expression is treated as the structure itself, the project may believe that a path in a graph proves a possible performed-work order, that a commutative square proves a real bridge, that a fold proves safe aggregation, or that a wiring diagram proves integration readiness.
Use E.18.2 to record a mathematical expression over one already selected E.18 TFS or E.18.NET network and state what that expression may be used for. This description does not select or reidentify that world-side subject, decide an atomic transformation, establish a work occurrence, pass a gate, settle an evidence case, or establish an architecture claim.
Forces
Solution
Write a TransformationFlowMathematicalDescription@Context only when the mathematical expression changes how you describe the transformation flow. Name exactly one subject that the expression describes: one E.18 TFS or one E.18.NET network. Keep that subject reference, the mathematical description, any C.29 lens-use judgment, and any E.17 publication face separate.
If a separate representation, transfer or adequacy question remains beyond the sufficient local description, use C.29 and retain its result; add a lens-use record only when the receiving use needs one.
First-use account
Start with what the recipient wants to infer about the selected structure or network. If you are to choose the expression, ask which changes in its participants or their connections could alter that inference. Could two arrangements have the same graph or summary and require different answers? Retain the distinguishing relation, supply an independent premise, or limit the inference to what both arrangements support. If the expression is given, establish which of those conclusions it already supports.
State the subject, mathematical object, correspondence, what the expression preserves and loses, and its declared use in ordinary prose or equations when sufficient. Use this compact record when the receiving use needs a separately inspectable description:
Exactly one of DescribedTransformationFlowStructureRef? and DescribedTransformationFlowStructureNetworkRef? is present. The first points to one E.18 TFS; the second points to one already selected E.18.NET network.
DescribedSliceOrLocusRef? may cite an existing path, slice, FlowPositionRef, ExposedFlowPositionRef, member path, E.18.NET NetworkCrossFlowRelationRowRef, or other independently identified part without copying the fields that define that object.
CandidateMathObject identifies the mathematical object; ExpressionKind names its expression family. MappingMode states how its relevant elements and operations correspond to the described subject, the direction of inference, and the assumptions on which that correspondence relies. Distinguish analogy, fitted representation, simulation and an exact structure-preserving map as C.29 requires.
PreservedStructure, LostStructure, DeclaredUse, and BoundaryStop follow the C.29 discipline when the expression is claim-bearing. PublicationFaceRef? points to a separate E.17 publication face.
The compact record has no generic neighboring-object reference. When the receiving use needs a neighboring claim, cite its C.2.1 claim-bearing episteme in the subject-specific account; identify an ontic subject or relation occurrence only through a separately named, correctly typed reference supplied by the pattern for that claim.
For a one-TFS description, recover the E.18 selection S and the mappings the expression uses. The mapping beta resolves participant positions to constituents. The mapping rho resolves each connection q to one selected relation occurrence r. Retain the full incidences, participant meanings, order and multiplicity needed by the declared inference. If r also participates as a constituent, the expression must preserve its identity across the participant and relation representations when an inference follows that connection.
A valuation claim uses E.18:5.2’s tagged connection/cut domain, declared channels and regime. A value on a drawn binary edge is not automatically a value on r or on a cut. Name the cut’s path family, boundary and crossing incidences, and declare any aggregation rule. A coarse drawing may omit detail, but an inference that needs an omitted n-ary argument, multiplicity or boundary must return to the complete representation. Record that loss and stop the unsupported inference.
When the question is whether a proposed calculation on the summaries preserves an answer, compare the two orders for that answer. Connect the detailed models under the subject's constraints and interpret the result, then compare it with combining the summaries and their retained relations. A proof about the admitted arrangements can establish the comparison. For a threshold or an existence question, stop when a sufficient bound, compatible realization or independent ground already gives the required answer. A summary can give an exact answer if all admitted detailed cases it merges agree on that answer and the proposed calculation returns it. For a complete list of attainable outcomes, cover every detailed case and establish a detailed realization for each listed outcome. For one possibility, a single compatible realization or an existence proof suffices (C.29.1:4.3–4.5, C.37.MM:4.2).
Empty-path identity and associative concatenation belong to the mathematical description. Check E.18’s complete-path admissibility separately, including required side inputs, editions, modes and budgets. Those laws establish neither actual execution nor a prescribed Work order.
For a guaranteed limit, an upper bound on all outcomes of a nonempty admissible domain can suffice. When the agreed criterion permits a specified risk, compare a supported probability or risk bound with that criterion. A list of possible outcomes by itself gives no probability. Reopen the omitted relation only when the available answer leaves the intended judgement open.
Expression families
Use these families to recognize expressions. They introduce no new FPF kinds. A local expression may combine several families; the record still names exactly one selected TFS or network subject, one current described part when relevant, and the declared use.
Five-way subject, description, lens, and publication discriminator
Use this discriminator before writing or accepting a mathematical description:
The same visible source may require several records, but each E.18.2 description chooses one described ontic subject branch.
A refrigerator principle scheme may include an E.17 publication face, a functional-architecture view, one selected E.18 TFS, a thermodynamic mechanism claim, and an E.18.2 graph or equation description. A network diagram may similarly publish an E.18.2 description of one already selected E.18.NET network.
If the expression is evaluated as a lens, apply the C.29 adequacy test; if it is rendered or published, identify the E.17 publication face and any current view or architecture-description membership. Neither record reidentifies the TFS or network.
Related claims
E.18.2 defines the mathematical description of the selected TFS or network. For any neighboring claim, use the row below that names the contribution needed for that claim:
Archetypal Grounding (Worked Slices)
Refrigerator principle scheme. A vapor-compression diagram can be a publication face. The cooling cycle can be a selected TransformationFlowStructure. The thermodynamic laws are mechanism or formal-substrate claims. The graph or equation set that describes the cycle is an E.18.2 mathematical description. It may preserve transformation order, heat-transfer constraints, and cycle closure while losing maintenance work, sensor uncertainty, and installation context. It does not prove the refrigerator works or authorize a repair.
Two descriptions of one build-the-builder network. A nested wiring description can preserve finite member paths and exposed positions while hiding an n-ary relation's qualification. A hypergraph description of the same E.18.NET value can preserve relation arity and endpoints while flattening recursive member boundaries. Both E.18.2 records cite the same network ref and state different preserved and lost structure; neither graph creates or reidentifies the network. A rendered diagram is a further E.17 publication value.
Two branches on one supply. At one instant each branch may draw 1 A or 3 A. Their currents add at the common boundary and the model has no other load there. An interlock allows exactly the pairs (1,3) and (3,1). Compose under that rule first: either pair gives 4 A. Now keep just the two individual sets of currents and combine them independently. The full list of pairs is (1,1), (1,3), (3,1), (3,3), giving totals of 2, 4 and 6 A. Both calculations establish compliance with a 6 A limit. The 6 A pair in the second calculation has no realization with this interlock. To decide a 4 A limit for this installation or report its exact total, retain the interlock or obtain an independent sufficient bound or measurement for the same use. A bound of 4 A alone settles the limit, while an exact total requires bounds or evidence that determine the value.
With an interlock allowing at most one high branch, the possible totals are 2 and 4 A if both-low is permitted. For the inquiry about possible totals, adding that constraint to the independent model is a strict refinement: all three surviving pairs were allowed before, at least one survives, and 6 A is excluded. For the question solely about compliance with a 6 A limit, all four original pairs already pass. A claim over a switching interval also needs the reachable intermediate states.
P2W carry-through. A P2W source expression or publication may draw a graph-shaped path from formal substrate to principle frame, mechanism position, method selection, work planning, work, and evaluation. The graph-shaped expression can be an E.18.2 description of the selected carry-through structure. The P2W move itself remains E.18.1; work planning remains A.15; dated work remains U.Work.
Neural-network dataflow. A transformer architecture diagram may describe layers, attention blocks, residual connections, and graph-like connection structure. If the current claim selects one TFS, use E.18; if it selects independently identified TFS or nested-network members plus cross-member relation occurrences, use E.18.NET; if it is an architecture claim, use C.30.
If the current claim is the mathematical graph, tensor-shape relation, or wiring expression that describes one such already selected subject, use E.18.2.
For benchmark superiority, apply the relevant comparison test. For training Work, apply A.15.1's occurrence and identity rules. For an evidence claim, recover its subject-support grounds and use A.10 when the source-to-use account is implicit or must remain replayable. For release, test the release action as Work and any separate subject-release predicate. For causality, apply the causal predicate and test for that claim. The diagram supplies none of those project results.
Circuit and algorithm. A logic-circuit schematic can describe a transformation-flow structure realizing a Boolean relation. The netlist, wiring graph, algebraic normal form, and truth table are different mathematical or formal descriptions. They do not by themselves decide whether the selected method exists, whether the CMOS mechanism is valid under voltage and timing conditions, or whether a dated powered run occurred. Take a drawing that partitions a fixed set of gates into disjoint, nonempty blocks covering the set. Splitting one block into two such blocks while retaining the others strictly refines the partition. Each new block lies inside an earlier one, and the same gates are covered. An extra label can leave all the blocks unchanged. If the truth table and the expanded circuit allow exactly the same Boolean output for every admitted input, the input–output answers stay the same. Timing at the output would require a timing model and its own correspondence.
Bias-Annotation
Conformance checklist
CC-E18.2-1The current EntityOfConcern isTransformationFlowMathematicalDescription@Context, not the selected E.18 TFS or E.18.NET network itself.CC-E18.2-2Exactly one described ontic subject branch is present:DescribedTransformationFlowStructureRef?orDescribedTransformationFlowStructureNetworkRef?. The optionalDescribedSliceOrLocusRef?resolves through the selected E.18 or E.18.NET subject and does not duplicate its fields.CC-E18.2-3The mathematical expression family is named without minting a new U-kind.CC-E18.2-4For a claim-bearing expression, state its declared use, the structure preserved and lost, and itsBoundaryStop. Identify which inference would require returning to the omitted information or obtaining other grounds. If the claim is that combining the summaries preserves the answer from the detailed models, establish that comparison for the stated domain. For the decision itself, a sufficient bound can settle a threshold, a jointly admissible case or an existence proof can establish possibility, and independent grounds can settle the requested answer.CC-E18.2-5C.29 is used when mathematical-lens adequacy, payoff, obstruction, preserved/lost structure, or stop condition is being evaluated beyond the local description relation.CC-E18.2-6Graph, path, slice, morphism, algebra, category, tuple, quotient, fold, refinement, factorization, wiring, and network-expression language can describe the selected subject mathematically. To assert which ontic structure or network is selected, identify it independently using E.18 or E.18.NET.CC-E18.2-7No mathematical expression proves work occurrence, authorizes action, passes a gate, settles evidence, or establishes architecture adequacy by itself.CC-E18.2-8A rendered graph, table, equation, diagram, or other publication face remains separate from the mathematical description and is handled throughE.17; changing it alone reidentifies neither the description nor its selected TFS or network subject.CC-E18.2-9When selected TFS, selected network, work, method, mechanism, signature, evidence, gate, decision, architecture, function, module-interface, or reusable-structure claims are current, apply the contribution named for that claim in §4.4 and keep the result it returns. E.18.2 records only the mathematical-description relation for one already selected ontic subject.CC-E18.2-10Distinguish the claims carried by a source expression or publication face by their current EntityOfConcern and relation position. Keep separate records when the receiving use needs their independent identity or reconstruction.
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
A label such as “graph” or “morphism” does not by itself identify what is being described. To use the label, name the current EntityOfConcern: E.18 keeps one selected TFS, E.18.NET keeps one selected network, A.3.4 keeps bounded transformation, E.18.1 keeps P2W carry-through, and E.18.2 keeps one mathematical description of exactly one selected TFS or network.
E.18.2 has a narrower scope than C.29. C.29 answers the general question "is this mathematical lens use adequate for this declared purpose?" E.18.2 answers the local question "what mathematical expression describes this one selected TFS or network, and which declared use does that expression serve here?" Using C.29 for adequacy avoids a second set of lens-use rules. Graphs, paths, categories, tuples and algebras remain useful for describing transformation flows.
SoTA-Echoing
Relations
- Apply E.18's one-TFS identity, position-resolution, selection-constraint, and local-value rules to select one
TransformationFlowStructureand identify the flow valuation, path, slice, crossing, resolved connection annotations, and actual dependent refresh used by the claim. - Apply E.18.NET's membership, boundary, and cross-member relation requirements to select one network of independently identified TFS or nested-network members and identify its obtaining cross-member relation occurrences.
- Apply A.3.4 to identify an actual bounded
U.Transformation, its changed referent, boundary, facts, and continuity or reidentification rule. - Apply C.29 to evaluate mathematical-lens use and retain its returned adequacy, preserved/lost structure, payoff, obstruction, or stop result when that claim is current.
- Use C.2.1 for description-episteme identity and E.17 for publication faces and their publication boundary.
- Use A.6.0 for formal-substrate signatures, A.6.1 for mechanisms and applications, A.6.5 for slot discipline, and E.20 for mechanism-method placement.
- Apply A.15.1 to identify performed Work, A.15.2 to identify work plans, A.20 to obtain internal-step validity, A.21 to obtain gate results, A.10 for an implicit or required-replayable source-to-use account, B.3 to obtain assurance, and C.11 to obtain local choices.
- Apply C.30 to state architecture claims, C.30.AD to identify architecture descriptions, C.30.ASV to evaluate structural views, A.6.F to recover unsettled function-like claims and their subject patterns, A.6.M to state module-interface relations, and C.31 to state reusable-structure characteristics.
E.18.2:End
Constraint-Governed Transformation-Flow Unfolding Structure
Type: E.18 transformation-flow specialization of
A.22.CGUSStatus: Stable Normativity: Normative unless marked informative
Use This When
Use this pattern when a team is planning, reviewing, or explaining a transformation and a route-like flow card is useful, but branches, joins, guards, or connections to independently identified neighboring values or neighboring claims already shown to obtain determine what can follow. Recover those transformation-flow relations without treating displayed order as performed-work order, evidence, decision, or authorization.
The admitted object is the same selected U.Structure already identified under A.22 and qualified as a CGUS by its constituents, selected obtaining relation occurrences, applied constraints, and one named selection-use frame. E.18.3 recognizes that object under an additional transformation-flow unfolding condition; it creates no second structure alongside that CGUS.
That condition uses one independently identified E.18 substrate branch: one TFS, one parent-relative internal SubflowRef within a TFS, or one selected E.18.NET network. The A.22-selected CGUS uses substrate positions, bindings, and already-obtaining occurrences; the substrate ref does not resolve to selectedCGUSRef and is not a second CGUS.
Do not use this pattern merely because a visible record or description is a route, path, graph, process map, chain, loop, or swimlane. First ask whether a branch, join, condition, dependency, crossing, or connection changes the continuation question for the thing being transformed. E.18.3 membership requires the A.22 identity, CGUS locus bindings and potential topology, and one of the three E.18 substrate cases with its flow positions, bindings, and obtaining occurrences. The current continuation result is evaluated separately. Description loss, C.33 notes for a declared architecture use, publication, and stronger neighboring claims are added only for uses that need them.
Start with the question about what can follow. Identify the subject or subjects whose conditions determine the answer. For one transition, two places on a flow card may show where the connection matters. For a join, find which contributions must be available together and what relates them. An ordinary explanation can supply the answer or name the missing fact. A separate C.2.1 episteme, E.18 position mapping or selected A.22 structure is needed only for a use that relies on it.
Recover the following values only when a named use depends on the identity and qualification of the unfolding organization as an E.18.3 structure: the A.22 identity, CGUS locus bindings and potential continuation rows, the E.18 or E.18.NET positions and bindings, obtaining relations, and applied constraints. For a decision about what can follow now, also evaluate the current continuation candidates. Add replay fields, C.33 loss notes for a declared architecture use, publication material, or downstream assurance only when that named use needs them.
Here move is Plain wording for the current action, not a universal kind or relation; a proposal, selection result, or formalized description does not establish that the proposed action was performed.
Problem Frame
E.18 describes transformation-flow structure: selected relations, dependencies, paths, crossings, guards, valuations, publication faces, comparability, complete dependent refresh, and structure-position bindings. A.22.CGUS supplies the broader constraint-governed structure. E.18.3 asks when that CGUS uses one E.18 substrate's positions, bindings, and obtaining occurrences in its potential continuation topology. Current continuation results and neighboring stronger claims remain separately testable.
Problem
Readers can mistake a path diagram for a workflow, a flow card for performed work, a P2W chain for work authorization, or a graph expression for the whole structure. Showing a gate, evidence path, architecture decision, or publication face next to a flow can likewise suggest that it is part of the transformation-flow ontology.
Changing a label cannot establish qualification. E.18.3 qualification depends on one A.22-selected structure, its CGUS-local loci and potential topology, the correct E.18 or E.18.NET case, typed transformation subjects, flow-position mappings, and obtaining occurrences.
The case- and time-indexed continuation result separately cites each candidate, its applicable test or obtaining-relation basis, case inputs, facts or evidence, dependent occurrences, window, and outcome. Description adequacy and stronger uses are separate decisions.
Forces
Solution
E.18.3 is a membership-and-use profile for one selected A.22.CGUS U.Structure. The selected structure keeps the four A.22 identity discriminators.
Its applicable E.18 substrate is independently identified as one TFS, one parent-relative internal SubflowRef, or one E.18.NET network. E.18.3 asks whether the selected CGUS uses positions, bindings, and already-obtaining occurrences from that substrate, together with its current constraints and use frame, to satisfy the transformation-flow unfolding conditions below.
Use the display to recover the selected values. It introduces no record kind and supplies no structure identity.
The four A.22 discriminators identify the selected CGUS. Filling in the display does not establish that identity. The mutually exclusive substrate fields identify independently current E.18 objects used by that CGUS; none is another identity field and none resolves to selectedCGUSRef. flowCase and the remaining rows show why that one CGUS qualifies and let the current use replay its substrate, subject kinds, relation predicate definitions, position bindings, and reconsideration boundaries.
No ambient context, transformed-subject label, path, valuation, tag, record edition, demonstration, or profile field becomes another identity discriminator.
Distinguish three kinds of basis when evaluating a continuation. An applied condition claim keeps its predicate or test, applicability, case inputs, and facts. An E.18 guard-failure event keeps its gate-assignment facts. A relation reference resolves only an independently obtaining relation occurrence. Each candidate's ContinuationJudgementResult cites the branch used for that candidate and records enabled, disabled, unknown, or error; the current-set result aggregates those candidate results without changing structure membership.
Distinguish a path in one TFS from a demonstration. PathId, PathSliceId, FlowValuation, and FlowPositionRef stay with one E.18 TFS. A post-qualification demonstrative slice is a separate C.2.1 episteme about the CGUS. Before qualification, a card or explanation remains about the actual subject, question, or proposed continuation set and need not be materialized as an episteme unless persistence or replay requires it.
A pattern-selection flow, selected-pattern-application flow and downstream-subject-work flow keep different EntitiesOfConcern, changes, Work occurrences, results, applicable definitions and tests, constraints and reconsideration conditions.
If all relevant participant positions and selected internal relation occurrences resolve to one TFS with their complete incidences, use the positions needed by the question and, when current, one complete top-level demonstration locator <transformationFlowStructureRef, pathSliceId, DesignRunTag>. A detailed internal portion remains one parent-relative SubflowRef.
If independently identified TFS or nested-network members cross, use E.18.NET to recover the network membership and exact cross-member occurrence requirements, including the applicable predicates and current facts showing that the membership and occurrences obtain; the mutually exclusive A.22 network locator applies and the top-level one-TFS triple is absent.
A result, tool, context, constraint, shared label or displayed arrow neither merges network members nor supplies their relation. Each member keeps its own boundary. Work, actual transformations, valuations and position state stay with their respective leaf TFSs. Nested pattern-selection content is present only while its exact source or selection-provenance relation is current for the declared demonstration use. When present, it contributes its own candidate, fit finding or recommendation rather than borrowing a later application result.
Identify preserved transformation structure by its U.Structure references. State captured, expected-but-uncaptured, lost and hidden structure for the declared use. Use exact C.33 epistemes for this account when carrier loss affects a declared architecture use.
A stop or reconsideration condition is an ordinary use boundary unless a relation occurrence is independently defined and shown to obtain by its applicability conditions and current facts. G.11 supplies the source-currentness and decay tests; E.18 supplies one-TFS slice-local refresh.
There is no generic method-to-work linkage here. When one named use relies on a Method-to-Work claim, cite the already-obtaining relation or result and the concrete definition, test or rule that supports it; keep Method, qualifying MethodDescription, WorkPlan, readiness and dated Work separate.
A pattern ref, intended realization, selected continuation, imperative sentence or displayed sequence does not admit any episteme as U.MethodDescription. A.3.2 supplies the membership test: one already identified C.2.1 episteme whose EntityOfConcern is one admitted U.Method and whose ClaimContent makes at least one substantive way-of-doing claim.
Each Method, qualifying MethodDescription, WorkPlan, work-entry result, dated Work, actual Transformation, production/inception/completion, evidence, evaluation, or source-use object must first be independently identified; any membership, occurrence, evidence, evaluation, or source-use claim obtains only when current facts or evidence satisfy the applicable definition, test, predicate, or rule. Only current objects and already-obtaining relations may enter the structure.
Ordinary start and conditional formal recovery
Begin with the alternatives the recipient is considering. Which part of the transformation, other participant or enclosing arrangement could enable one of them or exclude it? Trace the necessary conditions to where they hold, and ask which differences among possible cases change the answer. For two stages that must join, both may be individually enabled yet their outputs may belong to incompatible configurations. Check the configuration when the recipient needs the joint continuation. When one established condition already excludes that continuation, that fact can settle the question. An ordinary explanation can also identify what remains unknown. Use it directly when sufficient.
Use the numbered recovery branch below only when a named use depends on the identity and qualification of the unfolding organization as an E.18.3 structure. Comparison, publication and stronger reliance retain their own requirements for the claims made. That recovery preserves the membership and current-result distinctions; it is not a prerequisite for understanding or correcting an ordinary route-like card.
-
Recover one selected A.22.CGUS and its four identity discriminators; do not create a reciprocal E.18.3 structure.
-
Name the current transformation subject or subjects, their kinds and the E.18 positions and bindings used by the question.
-
Classify the independently identified E.18 substrate used by
selectedCGUSRefas one TFS with its valuations, one parent-relative internalSubflowRef, or one E.18.NET network of independent members and exact crossings; do not resolve the substrate ref to the selected CGUS. -
Discriminate every continuation basis before judging a candidate. Keep an applied condition claim with its test, applicability, case inputs, and facts; keep a
GuardFailas an E.18 event with its E.18/A.21 assignment facts; and use a relation-reference episteme only for an independently defined obtaining relation occurrence. For each candidate, record the dependent selected occurrences, window, outcome, and reason in aContinuationJudgementResult, then derive the current continuation set. Carry a relation signature only when declaration-level replay needs it. -
For each
neighboringValueUseRows[]entry, recover the independently identified neighboring value through its exact kind and ref and one already-obtaining supporting relation. If the row makes a stronger claim, state in ordinary language what the neighboring content contributes; a bare label such astestormethodis not enough. A definition, constraint, predicate, test, evidence rule, or assurance rule may supply the applicable criterion, with current facts or evidence showing that the claim obtains. A Method contributes a reusable way of doing and its applicability or bounds, and a MethodDescription may state that content; any claim that its use produces, supports, evaluates, evidences, or assures a result still needs a separate applicable rule and current basis. Require an exact claim-bearing episteme, ClaimGraph, edition, or other content identity only when that identity changes the selected stronger use. Reuse an existing exact ref when available. Cite a relevant pattern only when it locates that content. A result label, return arrow, or comparison layout is not the relation. -
Name the ordinary stop and reconsideration conditions. For a post-qualification description or demonstration, state preserved or omitted structure and add C.33 only when carrier loss affects a declared architecture use. Choose exactly one complete locator family for a one-TFS or network demonstration, or neither for a generic slice.
-
If an A.22 discriminator, CGUS locus binding or potential-continuation row, E.18 flow binding, or direct relation is missing, do not claim E.18.3 membership.
If only a fact or test result for the current case is missing, retain the structure and apply the candidate's condition logic. Return unknown when the gap prevents determination; retain an independently established disabled result and distinguish evaluation error. For a required conjunction P ∧ Q, known P=false establishes disabled even while Q remains unknown; known P=true with Q unknown leaves the outcome unknown. Preserve the unresolved Q in the basis in either case.
If only a description-loss or stronger-use value is missing, narrow that use rather than demoting the structure.
Exact relation references
When another person or later use must replay how one selected relation occurrence participates in the selected transformation-flow structure or supports a separately current subject use, materialize one ordinary C.2.1 episteme. Its EntityOfConcern is the already-obtaining relation occurrence, its ClaimContent contains only the current reference use below, and its effective ReferenceScheme governs every designation.
Transformation-flow relation reference is Plain wording for this use, not a local U-kind. The episteme’s edition and currentness remain ordinary C.2.1 and G.11 concerns; they do not add an identity field or ambient context.
The relation kind, predicate definition, ordered participants, current basis, and any network endpoint bindings carry the transformation-flow meaning; E.18.3 adds no separate structural-function classifier.
An internal relation occurrence is cited through its E.18 connection q and rho(q), retaining the direct predicate and every participant incidence inside its selected TFS. A dependency is recoverable only when the predicate truth conditions make one admitted continuation, state, or value depend on another and the participant order preserves that direction. A cross-member connection is recoverable only from an obtaining relation whose ordered endpoints bind admitted positions in different selected E.18.NET members. Use these tests to distinguish the three cases. Each occurrence retains its own relation kind and predicate.
An E.18 GateCrossing is a structure-local transition, not a U.Relation occurrence, and never enters this relation-reference field. A domain condition informally called a guard enters a relation reference only when an independently defined relation kind and an obtaining occurrence exist.
An applied constraint or condition claim is not the EntityOfConcern of this relation-reference episteme; keep it in appliedConstraintClaimRefs[] with its test and current facts. A GuardFail emitted by USM.CompareGuard or USM.LaunchGuard is an E.18 event, not a relation occurrence; recover the event and GuardOwnerGateId aggregation-assignment facts under E.18/A.21 instead. The word guard alone admits neither branch.
The optional supporting-use fields appear only when an independently current claim or relation says how the cited occurrence is used. That claim or relation keeps its own kind or predicate, current basis, and receiving use when the receiving use distinguishes it. No broad evidence, assurance, architecture, narrative, publication, gate, decision, comparison, currentness, or other use label makes the stronger use obtain.
One selected relation occurrence may support several separately established uses without becoming several occurrences; cite each claim or relation that matters rather than extending a classifier.
For a selected network mapping, resolve NetworkCrossFlowRelationRowRef to exactly one row in its named current record edition. The row, the relation-reference episteme and the direct occurrence must agree on occurrence identity, kind, predicate-definition source, optional signature, participant order, endpoint members, positions and bindings. The endpoint set adds no relation and makes none obtain; it preserves how the already-obtaining occurrence reaches admitted transformation positions.
A pattern identifier or reference is not a U.MethodDescription. A relation signature is carried only when that declaration exists and the replay needs it; citation does not make every use signature-dependent.
Connections to independently identified neighboring values
E.18.3 defines no universal neighboring-value relation. A neighboring Method, plan, Work, evidence, assurance, gate, decision, architecture, narrative, publication, evaluation, or currentness value must be independently identified. A claim about its kind, current status, or use obtains only when current facts or evidence satisfy the criterion supplied by the applicable definition, constraint, predicate, test, evidence rule, or assurance rule.
A Method contributes its reusable way of doing and applicability or bounds; a MethodDescription may state that content, but any truth, result, evidence, assurance, or Work claim about using it still needs its separate applicable rule and current basis.
A positive connection exists only through an already-obtaining relation. A stronger neighboring claim states its concrete contribution in ordinary content-bearing language; an exact content identity is added only when that identity changes the selected use.
Use this display row when a reader must recover the connection:
connectionQuestion is one free-text question, not a code, kind, relation, or closed question-type set. For example, a question may concern basis dependency, a result, a governing constraint, or a comparison.
A basis-dependency question creates no obligation. A result question is positive only after the exact result entity or relation and what it is a result of or for are recovered. A governing-constraint question needs the exact current constraint claim or occurrence. A comparison question needs its comparator, participants, scope and exact comparison definition or test; juxtaposition supplies none.
A positive answer connecting the neighboring values requires an obtaining supporting relation. Direction, participant order, applicability, occurrence identity, dependence and currentness come from its predicate definition, exact declaration when replay needs it, and current facts, not from the question wording.
When a stronger neighboring use is current, exactStrongerUseClaimOrRelationRef points to the independently governed claim or relation that establishes it; no broad use category substitutes for that ref. concreteContribution then says what the neighboring content does—for example, defines a term, constrains a claim, supplies a predicate or test, describes a Method's way of doing, or supplies an evidence or assurance rule.
These are examples of actions, not field values; definition, test, or method alone cannot fill the field. relevantPatternRef is only a locator. An exact claim-bearing episteme, ClaimGraph, edition, or other content ref is required only when that identity changes the selected stronger use. None of these fields creates a relation.
An ordinary stop uses stopCondition; reconsideration uses reconsiderationConditions[] to name the condition claim, affected structure and next question, with relevantPatternRef only when cited content supplies a needed contribution. Neither creates a receiver or connection relation.
If the supporting relation is missing, keep the neighboring values separate and record the attempted question. Use the A.6.RCD missing-governor result only when no applicable rule can state or test the claim for the stated participants and question; otherwise distinguish missing facts or bindings (missing-information) from a positive test that fails on a sufficient case basis (factually unsupported).
A failed positive test alone does not establish the negative. A recommendation, intended realization, rationale, common EntityOfConcern or adjacent box cannot supply the missing supporting relation.
Ordinary provisional explanation and admitted slice
Before the selected A.22 structure passes admission and the E.18.3 membership condition, a path fragment, flow card, worked example, or first-use account may remain an ordinary provisional explanation. It can name the concrete subject, recognizable places or states, proposed relations, possible continuations, and the missing fact or constraint without asserting a structure, position, or relation occurrence.
When replay, comparison, publication, or another current use needs that narrower account to persist as a claim, constitute one ordinary C.2.1 provisional episteme. Its EntityOfConcern is the independently identified subject designated by its claims; this may be the transformation subject, current question, proposed continuation set, or an A.22 structure whose CGUS or E.18.3 membership remains unresolved.
The provisional episteme’s ClaimContent may name the visible candidate places, proposed relations, presentation form, unresolved coordinates, and the exact condition that would resolve each one. Use the explanation or episteme to guide discovery; any claim that the proposed structure, position, or relation occurrence exists still needs its own identity basis and applicable test.
After qualification, a separate ordinary C.2.1 demonstrative-slice episteme may teach one enabled traversal. Its EntityOfConcern is the same selected CGUS recognized by E.18.3. Its ClaimContent cites established CGUSLocusBinding values, the relevant current continuation judgements, relation-reference epistemes or obtaining occurrence refs, any omissions that matter to this carrier (under C.33 for a declared architecture use), alternatives, presentation claims, admissible and forbidden uses, and the return condition. The slice creates none of those values.
For two actual Work occurrences identified under A.15.1, establish the order of the events in question (B.1.4, C.27.TA). On a reliable common clock, completion of A at 09:10 and start of B at 09:15 establish that A finished before B began. Relative order can also follow from an enforced start condition. If this B actually started through an interlock that could release it only after this A completed, the dependency establishes their order even without timestamps. The elapsed time remains open. A planned interlock, an unused permission or the arrival order of reports alone leaves that finish-to-start claim open. A.15.2 supplies the test for planned order. Use A.3.1 and A.3.2 when the claim concerns the reusable Method or its description. A demonstrated sequence can finish its explanatory use at the enabled path.
Do not infer that a demonstrated path is the whole topology. When the selected structure branches, joins, cycles, keeps alternatives live or is partially ordered, record what the slice omits or compresses before relying on it for comparison, architecture, evidence or planning.
A pre-qualification card can still help discover candidate CGUS loci and proposed E.18 positions. Name the subject-domain object or question, the proposed flow position and binding, and the missing A.22, CGUS, or E.18.3 membership value. Once those values are established, qualify the structure first and constitute a separate slice only if that presentation must persist. If a current fact is missing, report the candidate as unknown only when its condition cannot otherwise be decided. Missing description-loss material narrows only the description.
Admit network-aware demonstration mappings
A demonstrative slice can carry a network mapping only after its network and E.18.3 structure qualify. First select and verify one E.18.NET-conforming network. Then recover the one selected A.22.CGUS, its E.18.3 transformation-position mapping rows, and every required relation-reference episteme. Only then may the E.18.3 slice add its network demonstration mapping rows; those rows supply no missing member, position, relation, constraint, or admission.
For each selectedNetworkPositionMappingRows[] entry, resolve the finite member path to its leaf TFS. A FlowPositionRef names that TFS; an ExposedFlowPositionRef also repeats this network and the complete path. includedLocusBindingRef must be the same CGUSLocusBinding already present in the E.18.3 mapping and the slice's includedLocusBindingRefs[]. The network ref maps that binding to a flow position; it creates no copied position or constituent binding.
For each selectedCrossFlowRelationReferenceRows[] entry, require its NetworkCrossFlowRelationRowRef to name a current record edition whose EntityOfConcern is this slice’s selected network, then resolve exactly one row by occurrence and complete ordered endpoint-binding identity. Pair that row with one relation-reference episteme already cited by this E.18.3-qualified structure and with its matching networkEndpointBindingSets[] entry.
Verify occurrence, kind, predicate-definition source, optional signature, participant order, endpoint members, flow positions and bindings by value.
If the record describes another network, zero or several rows resolve, any field differs, or the relation reference is not already current, omit the mapping and name the missing or ambiguous network, row, position, occurrence, predicate definition or binding.
The complete top-level one-TFS locator is absent from a network slice. FlowValuation, PathSliceId and DesignRunTag remain member- or leaf-local; Work, actual transformations, boundaries and currentness also remain with their exact member and applicable definitions or tests. Member paths are finite and membership is acyclic, while exact cross-flow feedback occurrences may cycle when their predicates and constraints admit them.
Every selected cross-flow relation remains the occurrence whose predicate-definition source fixes its kind and participant meanings and whose applicability conditions and current facts show that it obtains. Do not substitute universal creates, produces, uses, input, output, result, handoff or transfer edges. One C.32.CONWAY result may contribute one exact architecture-influence and transformed-architecture correspondence row after its direct occurrence and endpoint bindings are recovered; it never constitutes the network.
A source phrase or graph enters only through an exact source-to-use claim or relation. A separately identified BoundedModelUseStructure participates only when the current assertion or use selects it and its organization changes interpretation of that claim; shared wording, adjacency, or a crossing display is evidence of neither model-use qualification nor crossing.
Positive case. A four-level build-the-builder demonstration follows one finite member path to an established leaf position, maps it to the same included CGUS locus binding, cites an admitted cross-flow relation-reference episteme, and keeps path slice and tag in one leaf-local row.
Near miss. A graph supplies raw positions or an edge label, mixes locator families, duplicates bindings, assigns one tag to the network, or cites a row without endpoint bindings; keep that demonstration provisional or return the missing member, relation, position, or binding.
Boundary
E.18.3 recognizes one selected A.22.CGUS U.Structure; it is not a second transformation ontology or reciprocal narrower structure. That selected CGUS uses one independently identified E.18 substrate branch and its positions, bindings, and already-obtaining occurrences; the substrate is not the selected CGUS.
The selected structure is not a workflow, Method, MethodDescription, WorkPlan, performed Work, actual Transformation, mathematical graph, publication, evidence relation, gate decision, architecture decision, or architecture description. It organizes independently identified constituents, already-obtaining relations, and constraints for one transformation-flow unfolding use.
A graph, record, table, or demonstration may carry claims relevant to A.22 identity or E.18.3 qualification. Its form alone supplies neither those grounds nor MethodDescription or Work admission. A.3.2, A.15.1, A.3.4 and A.15.PROD supply the applicable membership or occurrence tests; every relation claim still needs its predicate definition, applicability conditions and current facts.
When a continuation is intended to perform work, apply A.22.CGUS:4.1.1 to its constituent and encompassing conditions, including shared capability and capacity in the required window. B.1.5 and B.1.5.CM identify which Methods the whole construction requires or admits as alternatives. B.1.5.EW asks which actual constituent actions enact the encompassing Work. Nested flow positions and temporal arcs can describe either case once its subject relations are established.
Local completion supports the whole result only under the required coordination and result-use conditions. A description that omits those conditions answers a narrower question than whether the whole can be performed.
Replay and change localization
Replay A.22 identity, CGUS membership, and E.18.3 membership separately. The first uses the four A.22 discriminators; the second uses local locus bindings and potential continuation topology; the third maps those bindings to one E.18 substrate case and its positions, bindings, and obtaining occurrences.
Replay the current set from each candidate's basis rows, retaining the values required by each applicable condition-evaluation or obtaining-relation branch in A.22.CGUS:4.2, the relevant case inputs and facts or evidence, dependent occurrences, window, continuation result, and reason. Description loss and every stronger neighboring claim remain separate uses.
Localize changes by the value they affect. A changed A.22 discriminator can reidentify the structure. A changed CGUS locus binding or potential row reopens CGUS membership. A changed E.18 substrate, flow position, binding, or selected occurrence reopens E.18.3 membership.
A changed test, fact, evidence item, guard event, assignment fact, or window reopens only dependent continuation judgements and the current set unless it also changes one of those membership bases.
A changed carrier omission reopens the affected description and, for a declared architecture use, its C.33 episteme; a changed neighboring use reopens its own claim.
A changed demonstration, valuation, path slice, local tag, continuation outcome, or enabled-set cardinality does not by itself create another structure. Reidentify the selected U.Structure only when one of its four A.22 discriminators changes.
Archetypal Grounding — Worked Slices
Ordinary first use — heat-treatment card. A practitioner reviewing the flow card for GearBlank@Lot-14 marks “soak complete” and “quench candidate,” writes “quench remains an admissible continuation only when the measured soak state is within the allowed range,” and asks whether the card may show that continuation or which fact or constraint is missing.
If the measured-state fact or range rule is unavailable, the useful result is a provisional explanation naming that gap. The team may use it to correct or discuss the card and stop; it authorizes no Work and asserts no C.2.1 episteme, A.22 structure, E.18 position, applied constraint claim, E.18 guard event, or relation occurrence. Continue to formal recovery only when the named use depends on the organization's identity and qualification as an E.18.3 structure.
Candidate-set replay entry. When the team must determine whether the proposed candidate-set repair organization qualifies as an E.18.3 structure, name one proposed selected-structure use, CandidateSetComparisonBasis@Review-2026-07 and its kind, then describe candidate ReferenceEditionChangePosition and ComparisonRecalculationPosition plus the proposed dependency ComparisonDependsOnAdmittedEdition.
Because this use needs a replayable claim, constitute an ordinary C.2.1 provisional episteme whose EntityOfConcern is the qualification question about that proposed organization. Its ClaimContent names the G.11 currentness test and A.19.CPM comparison rule as needed contributions and states that the A.22 identity, exact E.18 bindings, and dependency occurrence remain unresolved.
This account keeps a comparison based on a stale edition from appearing current. It does not yet assert a structure, typed position, or relation.
P2W carry-through. Accepted problem-side records may name distinctions, constraints and unresolved relation positions that guide later Method selection, planning, Work, interpretation and reconsideration. E.18.3 may organize independently current objects only after the selected A.22 structure, E.18 position bindings and direct relations are recovered. It does not authorize launch or performed Work, does not admit any MethodDescription from intended use, and does not replace E.18.1 carry-through.
Recursive build-the-builder demonstration. After a network and its E.18.3 mappings are established, a slice follows one finite member path to an established leaf position. The network mapping points to the same included CGUSLocusBinding, and every cross-member row cites a relation-reference episteme with matching participant positions and bindings. The leaf path slice and tag stay in its member-local row. Before those facts are recovered, the graph remains an explanation rather than a network-aware slice.
Complete compact high-reliance case — edition-current comparison basis. This replayable comparison has two potential continuations: recalculate with the admitted edition, or stop and replace the edition. The objects and occurrences named below have already been identified.
The structure has two potential candidates; the present window enables recalculation and disables replacement, with no unknown or error result. The currentness claim remains a condition claim with a test, applicability, inputs, and facts; it is not inserted into relationReferenceEpistemeRefs[].
If those facts disappear, the recalculation candidate becomes unknown and the replacement candidate is judged under its own condition; the topology and structure do not change merely because the enabled set does.
Partial candidate-set recovery display. The larger four-position account below shows a broader teaching slice with several values still unresolved. Use it to recover those values before claiming conformance:
The unresolved position refs and bindings, the full ClaimContents and current bases of both dependency references, the tests and current facts for both applied claims, and every neighboring-value row must be recovered before this larger account can pass the checklist. Neither applied claim belongs in relationReferenceEpistemeRefs[].
After those values and the omission and reconsideration conditions are recoverable, the demonstration ref may name a separate episteme about the same selected structure. Use C.33 for the omission account when carrier loss affects a declared architecture use.
Local edition-relation repair. After ReferencePublicationEdition@v2 is admitted, use G.11 to record the edition change; ComparisonDependsOnAdmittedEdition still references v1. Keep independently unchanged constituents, positions, path and path-slice identifiers, preserved structures, and reconsideration conditions.
Re-evaluate the relation under its predicate definition and current facts, replace the selected occurrence only if the v2 predicate obtains, and then re-evaluate EditionAdmissionGuard as an applied constraint claim under its test and current facts. Reopen the A.19.CPM comparison use only if its basis changed, C.18 only if the comparison result changed, and C.32.PAD only if that retained-set change affects the current decision.
If the selected occurrence changes, the A.22 relation discriminator changes and the selected structure must be reidentified; mere publication wording or a new relation-reference episteme does not do so.
Connected-box proxy failure. A team reports that every flow-card box is connected and adds low-value edges until path coverage reaches its target. The relation count rises, condition labels no longer distinguish applied claims, E.18 guard events, and actual relation occurrences, stale dependencies remain unrepaired, and unsupported neighboring connections increase.
Edge count, labels, and path coverage describe the expression only. Remove edges without exact occurrences and predicate definitions, recover each continuation's actual condition branch, evaluate whether practitioners select the correct continuation and smallest repair, and use E.13 when display coverage substitutes for those outcomes.
Architecture P2S projection. A P2S flow card includes architecture-relevant problem pressure, unknown or selected structures, synthesis positions and actual-structure feedback. If one selected CGUS satisfies E.18.3, cite its exact E.18 positions and relations. C.32.P2S defines and constrains selected and expected epistemic structures and their exact use; realization Work and actual world-side structures remain separate.
C.30.TFS-REL supplies the architecture-use rule and C.32.PAD supplies the architecture-decision test. One C.32.CONWAY correspondence may be one qualified E.18.NET row, never the whole network.
Physical workpiece transformation. A heat-treatment unfolding use concerns GearBlank@Lot-14, independently admitted as a project U.Holon, and selects exact E.18 positions for load, soak, quench, and hardness evaluation. QuenchAdmittedAfterSoakRange is an applied condition claim only when its range test and current measured-state facts are recoverable; it is not thereby a relation occurrence or E.18 guard event. If an exact USM.CompareGuard or USM.LaunchGuard failure is current, recover that event and its gate-assignment facts separately.
Furnace loading and quenching must pass the applicable A.15 plan or dated-Work test; each actual heat-treatment change must pass A.3.4; production, inception, or completion uses the A.15.PROD tests; hardness uses the applicable measurement, evaluation, and evidence rules. A flow card can expose alternatives before execution without claiming that Work occurred.
Clinical transformation planning. A treatment-adjustment unfolding use concerns Patient@Case-17, independently admitted as a U.System, and selects assessment, intervention-candidate, contraindication, observed-state, and reconsideration positions. A contraindication condition remains an applied clinical claim with its test and current facts; a cited E.18 guard failure remains an event with its gate-assignment facts; and the observed-state relation changes admissibility only when its independently defined kind and occurrence obtain.
The selected structure does not authorize treatment, show that evidence is sufficient, replace clinical judgement, admit a MethodDescription, or show that an intervention occurred; those claims require the applicable clinical DPF, permission, Work, evidence, and gate definitions or tests plus the current facts or evidence that satisfy them.
Formal flow-expression boundary. A team expresses the candidate-set repair use as a directed graph or DCR model to ask whether DecisionRepairPosition is reachable after EditionAdmissionGuard. The expression may preserve the dependency topology and a condition label plus the queried path, but it does not decide whether that condition is an applied claim, an E.18 guard event, or an independently defined relation occurrence.
It also loses neighboring claims already shown to obtain, their concrete contributions, description omissions (including C.33 omissions for a declared architecture use), and currentness semantics unless those are separately mapped.
Use E.18.2 for the mathematical description and C.29 for its declared use, preserved/lost structure, and stop. Positive reachability alone shows neither the condition's ontic type, currentness, retained-set validity, decision repair, Work order, nor selected-structure identity.
Reference-currentness repair. A one-TFS path slice may depend on an admitted publication edition, a G.2 source-use relation, a source pack or a telemetry window. E.18:5.13 supplies the complete dependency closure, addressed through all needed PathSlices. G.11 supplies the tests for source currentness, decay, edition shift, deprecation, reship and no-change claims.
Connect these values only through exact obtaining occurrences and their predicate definitions, and reconsider every dependent continuation, including both branches of a shared dependency and their join; do not create a combined currentness-refresh value.
Bias-Annotation
Conformance Checklist
Common Anti-Patterns and How to Avoid Them — Repairs
Consequences
E.18 can describe a transformation-flow use without classifying every route-shaped pattern as a transformation flow. P2W, P2S, agent-loop, gate, evidence, architecture, and currentness cases may use the same selected A.22 structure and exact transformation-flow relations.
Each stronger neighboring claim obtains only when current facts or evidence satisfy its applicable definition, constraint, predicate, test, evidence rule, or assurance rule. A Method may contribute a reusable way of doing and its applicability or bounds, but any claim that using it produces, supports, evaluates, evidences, or assures a result remains separately testable under its own rule and current basis.
Identifying the structure, establishing its qualification and recovering the values for a named use all take work. Undertake that work only when qualification, comparison, publication, or stronger reliance requires it. A selected CGUS qualifies for E.18.3 through its E.18 substrate case, subjects, locus-to-flow mappings, and obtaining occurrences. Current continuation judgements, description loss, publication, and stronger claims remain separate results or uses.
An ordinary card can stop after naming the subject, alternatives, conditions, and missing fact or rule.
The benefit is change locality. A changed demonstration, valuation, path slice or tag usually changes only that use; it does not reidentify the selected structure. A changed selected constituent, occurrence, applied constraint or named selection-use frame changes an A.22 discriminator and therefore requires a different structure selection.
Rationale
E.18 distinguishes transformation-flow structure from the whole work process. E.18.3 uses the same distinction. Constraint-governed unfolding adds a next-use concern—how one selected structure exposes admissible continuations while protecting the differences among structure, description, Method, MethodDescription, plan, Work, transformation, production, evidence, gate, decision, architecture, publication, E.18 slice-local refresh and G.11 currentness.
E.18.3 does not create a reciprocal specialization object or universal connection relation. It recognizes one A.22-selected U.Structure when that CGUS uses positions, bindings, and already-obtaining occurrences from one independently identified E.18 substrate branch and its current transformation-flow constraints support the unfolding use. It uses ordinary C.2.1 epistemes only to make that qualification and its demonstrations replayable.
SoTA-Echoing
In the 7 August 2026 comparison, OCPQ was the research comparator for typed multi-object constraint structure, and the cited Dyad documentation supplied a bounded engineering comparison of relation-first models kept separate from analysis definitions, compilation, execution, and results. Modelica 3.7 supplies historical acausal-modeling lineage only; the older CMMN, Declare, DCR, and artifact-centric rows also supply lineage. Reopen the adoptions when object-centric constraint methods change object-relation treatment, model languages change model-analysis separation, or use evidence shows that these distinctions no longer prevent workflow, query-result, or execution-artifact overread.
Relations
Specializes: the A.22.CGUS use of one selected U.Structure with the same constituents, selected obtaining relation occurrences, applied constraints, and named selection-use frame. The CGUS qualifies when it uses the positions, bindings, and obtaining occurrences from one independently identified E.18 substrate branch and satisfies the transformation-flow unfolding condition. E.18.3 creates no second structure or ambient context identity, and no substrate ref resolves to selectedCGUSRef.
Builds on: E.18 for one-TFS substrates, resolved participant and connection positions, obtaining relations, valuations, paths, slices, and SubflowRef; E.18.NET for network substrates, member paths, exposed positions, and cross-member occurrences; A.22.CGUS for structure identity, CGUS-local locus bindings, potential continuation topology, separate current judgements, and description/slice separation; and A.3.4, A.22, and E.17 for transformation, structure, and publication discipline.
Coordinates with: E.18.1, C.32.P2S, C.30.TFS-REL, C.32.CONWAY, E.23, C.18, C.19, G.5, A.15, A.15.PROD, A.10, B.3, A.20, A.21, A.6.3.NAR, exact source-use patterns and G.11. A network demonstration consumes only already-current E.18.3 position mappings and relation-reference epistemes; one C.32.CONWAY occurrence can fill at most one qualified network row.
Does not replace: the definitions, constraints, predicates, membership or occurrence tests, evidence rules, and assurance rules governing Method, MethodDescription, Work, transformation, production, evidence, assurance, gate, architecture, decision, publication, mathematical-lens, source-use, E.18 slice-local refresh, or G.11 currentness claims.
Nor does it replace a Method's reusable way of doing and applicability or bounds. A Method or MethodDescription supplies no truth, result, evidence, or assurance criterion merely by being cited; any such stronger claim retains its separate applicable rule and current basis.
A pattern ref only locates content, and a contribution-form label does not state that content; exact content identity is required only when it changes the selected use. Pattern refs, selected continuations, imperative wording, graph adjacency and intended realization do not establish the independently governed objects or claims listed above.
E.18.3:End
Network of Transformation-Flow Structures
Tech-name: TransformationFlowStructureNetwork Plain-name: Network of transformation-flow structures Type: Structural pattern for ontic relations (E) Status: Stable Normativity: Normative unless marked informative
Problem frame — intent and first useful result
Use this pattern when one engineering question depends on two or more independently identified transformation-flow structures, or on nested networks of them, and at least one obtaining relation connects positions across their boundaries. Typical situations include a toolchain that builds another tool, a production system related to the product it helps produce, or an operating flow whose observation returns to a separate development flow.
Start from what the recipient needs to establish across the flows. Work back from the use of a result to the flow that supplies it and any other flow on which that use depends. At a nested boundary, ask whether the exposed result suffices or whether two internal arrangements with that same result could change the answer. Follow the difference only as far as the question requires. Use a shared constraint when it affects what several members can provide together. Then select among the three cases in step 1 and continue with steps 2–5 when the question requires a network.
- decide whether the case is several valuations of one flow structure, an internal portion of one flow structure, or a network of independent flow structures;
- identify each candidate member independently;
- name the obtaining relation occurrences that connect positions in different members;
- select only the members, relations, boundary exposures, and constraints needed for the current question; and
- return one network reference, or stop at the proposed description and name either the relation-claim result returned by its governing pattern or the separate missing network discriminator.
A successful selection returns a network reference and its basis. For a request to identify that network, this can be the final answer. For a receiving decision, use the selected connections to establish what can be concluded across the members and obtain any further premise the decision needs.
If selection stops, return the proposed description and the reason. For example:
When the relation claim has a positive obtaining result but a network endpoint is not bound, keep that positive result and state a separate E.18.NET selection blocker:
If a fact needed to decide the relation claim is unavailable, carry the governing pattern's missing-information outcome; a sufficient case basis that fails its positive test yields factually unsupported. Neither outcome alone asserts a negative. Carry an inapplicable or negative result only when that pattern's applicable rule and case basis establish it.
A missing member, applied constraint, or networkUseFrame remains its own network-selection blocker and never becomes a relation result.
Keep proposedNetworkDescriptionRef until all four A.22 discriminators—members, selected obtaining relation occurrences, applied constraints, and use frame—are recoverable; only then assert selectedNetworkRef.
Do not use E.18.NET merely because one flow branches, contains a detailed portion, has several valuations, or is drawn as a network. Use E.18 for one selected TransformationFlowStructure, its valuations and selected internal obtaining relations; use E.18's SubflowRef for one parent-relative internal portion.
Use E.18.2 when the current object is a graph, wiring diagram, tuple, category-theory expression, or another mathematical description. Use A.22.CGUS and E.18.3 when the current object is an admitted demonstrative traversal rather than the network itself.
Problem
Teams routinely connect flows that concern different objects, Work occurrences, architecture boundaries, valuation state, and change cadence. A development TFS has positions for the Work that produces or changes a tool; other TFS values have positions for its use and evaluation, with feedback to development.
A manufacturing system is changed through one flow while products are made through another. A compiler is built by one toolchain and then participates in a later build.
A single picture can hide three different ontic answers:
When the third case is treated as one giant TFS, local state appears global, one internal connection is asked to replace distinct production, use, evaluation, feedback, correspondence, and dependency predicates, and a change in one member appears to reidentify everything.
When the first or second case is over-split into a network, the model invents members and relations that the engineering situation does not need.
Forces
Solution
Select a dependent non-agentive structure
TransformationFlowStructureNetwork@Context is a dependent, non-agentive specialization of U.Structure defined by E.18.NET and selected through the A.22 identity law. It is not a root U-kind, acting system, holon, workflow, graph, record, publication, FlowValuation, WorkPlan, or performed Work.
The @Context suffix qualifies retrieval and use; it adds no identity discriminator.
For N : TransformationFlowStructureNetwork, recover its four identity discriminators:
As in the first-use result, the four fields identify the direct members, selected obtaining cross-flow occurrence refs, applied network constraints and one concrete use frame.
returnCondition is not a fifth identity discriminator; it records when the current use must return and reselect. stopOrReturnCondition states the boundary of the action or use within networkUseFrame; returnCondition names a change that reopens selection.
forbiddenOverread?, also called groundedForbiddenOverread?, records an optional explanatory guard under F.19:4's plausible-reader test. It remains outside networkUseFrame and structure identity. A change to that explanation alone leaves the network unchanged; if its content changes an applied constraint or a use-frame value, compare that existing discriminator.
The direct-member set contains at least two members. Each member is one independently identified TransformationFlowStructure or one independently identified E.18.NET-conforming TransformationFlowStructureNetwork. At least one selected relation occurrence binds positions in different direct members or in different leaf TFS members reached through them.
The use frame states the concrete question or action, how this selection will be used, and its stop or return condition. “Current use”, “appropriate network”, and the title of a diagram are not use frames.
The direct-member discriminator identifies the selected members; record rows cite those independently identified values. If a later use needs a separately re-identifiable world-side membership occurrence, apply the direct relation pattern that defines its participants, predicate, applicability, and identity rule. When no such rule is available, reopen the relation question under A.6.RCD.
Reidentification and change locality
Replacing a direct member, selected relation occurrence, applied endpoint or exposure constraint, acyclicity constraint, or named selection-use frame identifies another selected network. Reidentifying a nested member reopens every parent network that selects that member.
Changing only a name, reference designator, record edition, graph layout, mathematical description, publication, selecting system, selection Work, evidence item, FlowValuation, PathSliceId, or local DesignRunTag leaves the network unchanged when the four A.22 discriminators still resolve to the same values.
Recurse through finite member paths
The selected direct-member nesting is acyclic. No direct or transitive member path from a network resolves back to that network, and every member path used by a reference is finite. The same network kind supports build-the-builder and supply-network recursion at any finite nesting depth.
When functional layers are useful in a flow network, organize them by the contributions that the selected flows make to the stated result. In a build-the-builder case, one flow produces a system that performs work in another flow; recover the production, use and participation relations that connect them.
Nesting one selected network inside another records member grouping for the current question. The depth of that nesting does not determine the functional layers or a modular hierarchy. Use C.30.STRAT:4.2 to keep functional layers distinct from levels in a module and interface structure.
Cycles among selected cross-flow relation occurrences are compatible with acyclic member nesting when their predicates and constraints permit those connections. To claim operating feedback, recover how the returned observation was used to choose or maintain a subsequent action (C.43).
E.18 defines the complete FlowPositionRef identity. Import that tuple unchanged; E.18.NET defines only the ExposedFlowPositionRef extension needed for a boundary position reached through one finite member path:
Every hop in memberPath[] resolves through the preceding network's direct members. Its final member is the TFS named by leafFlowPositionRef. When the path crosses a nested network, the leaf position must be one of the boundary positions that nested network exposes for the current use by the containing network. Two different paths to the same leaf TFS position are two different exposures.
The parent network may compose the finite path and use the exposed boundary. It must preserve the nested member’s internal structure without copying or flattening it. A separate mathematical description may flatten the nesting for a declared use when it states what the flattening preserves and loses. FlowValuation, PathSliceId, actual fillings, and DesignRunTag qualify use of a position; they are not part of FlowPositionRef or ExposedFlowPositionRef identity.
Keep valuation and design/run state leaf-local
Each positionBindingRef cites the relevant E.18 participant-position/use binding or a declaration-local binding whose pattern defines the needed participant meanings, value kind, and reference mode. A network introduces no universal cross-flow value kind.
Recover each leaf TFS under E.18:5.1, including its A.22 identity, beta resolution and the complete participants of its selected relations. Its valuations use E.18:5.2’s declared connection/cut domain and regime.
DesignRunTag belongs to one position binding inside one leaf TFS. A network has no FlowValuation for the network as a whole, global design/run ladder, or automatic crossing that changes the carried entity's kind.
If the same episteme fills local positions in different members—for example one position concerned with design work and another with production, verification, or later operation—record each leaf-local binding and the exact relation that obtains between them. Those ordinary member descriptions create no fixed TFS taxonomy or lifecycle phase.
When a changed basis is used across members, follow E.18:5.13 through the exact cross-member occurrences and every dependent branch and join. Cover the required leaf-local PathSlices; independent uses and each member’s local state remain separate.
Preserve the direct cross-flow relations
For every relation used by the network, recover the following values.
- the obtaining occurrence;
- its admitted relation kind;
- the pattern that defines or tests its predicate, applicability, and occurrence-identity rule;
- the complete participant meanings and order;
- the endpoint member and position binding for every participant; and
- direction only when the direct relation has direction.
An n-ary relation remains n-ary. Do not decompose it into invented binary arrows. A row, edge label, shared entity, temporal adjacency, operation result, plan row, or graph connection never makes the relation obtain.
E.18 resolves each internal connection(q) to its selected relation occurrence r. That position is local to one selected TFS; it supplies no universal cross-member predicate.
For any production, use, participation, evaluation, correspondence, feedback, dependency, supply, or other cross-flow relation, the relation kind must already be admitted. Use its applicable relation pattern to recover the participant meanings, predicate, applicability, and occurrence-identity rule; current case facts or constituting history must satisfy the positive predicate.
The occurrence obtains when those conditions hold. For network selection, identify an occurrence that already obtains. Use A.6.REL only when a named use must distinguish that occurrence from another.
For ordinary network selection, the PatternID and exact relation occurrence are enough; add relationFunctionClaimRef to the defining or constraining ClaimGraph only when comparison, migration, or reliance depends on that exact rule identity. The network selects only the ref to the occurrence that already obtains.
If no admitted relation kind and applicable predicate cover the intended participants and use, carry missing-governor from the pattern governing the relation claim. If required case facts are unavailable, carry its missing-information result; if the available basis is sufficient to apply the positive test but that test fails, carry factually unsupported.
Neither missing information nor a failed positive test suffices to establish a negative. Carry an inapplicable or negative result only when the governing pattern defines that outcome and its current basis establishes it. Only a positive obtaining occurrence may fill selectedCrossFlowRelationOccurrenceRefs[].
After a positive occurrence is established, test the E.18.NET endpoint and position bindings separately. A missing binding blocks network selection but does not change the relation result. Missing members, applied constraints, and use-frame values are likewise separate network-selection blockers.
A row, graph edge, or episteme neither admits a relation kind nor creates an occurrence. In none of these branches substitute creates, produces, uses, input, output, result, handoff, or transfer as a generic edge.
Record the network without replacing it
If the selected answer must remain available after the immediate work, describe it with a separate C.2.1 episteme. Its effective scheme has the C.2.1:4.1.2 range EpistemeInterpretationScheme; recover the rules that interpret the record's network, member, relation and endpoint-binding designations and its claims:
The C.2.1 record describes the selected network. The network and record have different identities. The record’s member and relation rows cite objects that already exist and occurrences that already obtain.
An architecture-correspondence row is a qualified reading only. It contributes no member or selected cross-flow relation unless a separately grounded relation occurrence and endpoint bindings also satisfy the network identity.
E.18.NET defines this composite locator for one nested cross-flow row:
Resolve the record ref first, then match crossFlowRelationRows[] by the occurrence ref and the complete ordered endpoint-binding identity. Exactly one row must match. Zero matches or several matches leave the locator unresolved and stop that consumer; never fall back to the containing record, the occurrence alone, or a prose pointer.
NetworkCrossFlowRelationRowRef is a reference shape, not a U-kind, episteme, or relation occurrence. Its U.EpistemeRef targets the containing record, never the nested row.
Keep descriptions, demonstrations, architecture, and Work outside identity
Use E.18.2 for a graph, hypergraph, network expression, wiring diagram, category-theory object, tuple, fold, or other mathematical description of the selected network. State what that description preserves and loses. Use E.17 when publishing a rendered graph or another face. Use C.29 when the adequacy of the mathematical lens for its declared use is at issue.
Use A.22.CGUS and E.18.3 for an admitted network-aware DemonstrativeUnfoldingSlice@Context. Its finite paths must map to already admitted included positions, its cross-flow relations must cite admitted relation-reference epistemes identifying those occurrences, and its tags remain in leaf-local bindings. The slice demonstrates one traversal; it is neither the network nor an actual trajectory, WorkPlan, or Work occurrence.
Use C.30.TFS-REL when architecture uses the independently selected network. Distinguish one named containing holon with an obtaining architecture relation whose selected structure is that network, a bounded C.30 claim about one named holon, and an explicit inter-holon use with its participating architecture claims.
For the single-holon claim, retain its negative, unresolved or candidate/expected disposition, supporting or missing basis, and return before stronger reliance; the architecture claim does not relax network selection.
Use C.32.CONWAY only for its one-pair architecture-influence reading; the pair does not become the network.
Only admitted Systems perform Work. Selecting a network, writing its record, or drawing its graph may be Work when A.15.1 independently admits the occurrence after each precise performer has an A.13 core; none is performance by the network, and no Work claim is needed merely to select or discuss the network.
When selection Work is material, cite those already established A.13 and A.15.1 results. Cite F.6 only when the current network account also needs precise assignment-bound attribution, and leave its proof with F.6.
Keep the Method, performer, dated Work, result episteme, selection or decision relation, and any C.11 choice result separate. A result episteme is not a decision or accountability relation by form; state accountability, duty, responsibility, or authority only through the direct relation that obtains.
Archetypal Grounding — worked cases
Same surface vocabulary, different ontic answers
Several valuations of one TFS. A cooling-loop review compares nominal-load and emergency-load valuations of the same cooling-loop TransformationFlowStructure. Both valuations retain the same A.22 selection O/R/C/F and use the E.18 connection/cut domain under their declared regimes. A counterfactual load value, path slice or local use tag may differ while those discriminators remain unchanged.
Replacing an actual run, applied selection constraint or selected occurrence first reopens the TFS identity. E.18.NET is not used.
Internal coffee subflow. A coffee-brewing TFS exposes a preparation portion containing grinding, dosing, and wetting positions plus their selected parent-internal relation occurrences with all participants retained. Its entry and exit remain positions of the brewing TFS. The practitioner uses E.18's SubflowRef; no second TFS or network is created.
Independent network. A roastery-production TFS and a café-brewing TFS concern different objects and have separate Work occurrences, valuation boundaries and architecture change cadence. The supply predicate and the dispatch and acceptance facts establish that lot 24 reached the café's accepted-stock position. For today's brewing decision the café also needs enough of this lot available and eligible at the time of use.
The dispatch and acceptance records suffice to trace what the café received. Suppose its current records show 4 kg of lot 24 still on hand and uncommitted, cleared for brewing at 10:00. The café can choose it for a 3 kg batch then. If 3 kg has since been reserved for another order, only 1 kg remains available for this batch. A recall can close the choice even with all 4 kg present. With only the acceptance receipt, report the completed delivery and recover the needed current condition. Create a TransformationFlowStructureNetworkRecord@Context only when the network description must survive the current work.
If the supply claim has no admitted relation kind or applicable predicate, carry the governing pattern's missing-governor result. If required facts are unavailable, carry missing-information; if a sufficient case basis fails the positive test, carry factually unsupported. Neither missing information nor a failed positive test asserts a negative. Only an applicable negative rule and satisfying case basis can supply a negative result.
When the supply occurrence obtains but an endpoint binding is missing, keep the positive occurrence and name the missing binding as a separate E.18.NET selection blocker. A missing member, applied constraint, or coffee-service use frame is also a separate selection blocker.
Project system-of-interest and recursive build-the-builder
For one project question, practitioners ask which independently identified flow structures must be considered together to connect production and later operation of the project system-of-interest, and which builder branches must also be visible. The actual project remains composite U.Work; the selected network is a non-agentive U.Structure.
Project designation and U.System identity remain separate from any local system-role kind, classification, assignment, selection Work, or result episteme. None follows from a project or network label.
For the compiler-and-application use, identify five TFS values by the questions they answer:
CompilerEditionPreparationTFS, whose loci bind compiler-edition preparation and the obtaining source-use occurrences needed by the build;BootstrapCompilerBuildTFS, whose loci bind Work on pre-existing build substrates and the separately grounded production and identity-inception claims for one bootstrap compiler;ApplicationBuildTFS, whose loci bind application-production Work and the use of that admitted compiler in the application build;ReleaseAssuranceTFS, selected for release-assurance questions; andDeploymentOperationTFS, selected for deployment and operation after the application system exists.
These names designate independently identified TFS values, not lifecycle kinds. They assert no transformation of a not-yet-existing compiler or application. Use E.18 for each TFS, A.15.1 for any current Work occurrence, A.3.4 for a change of a continuing referent, A.15.PROD for production or identity inception, and the applicable relation pattern for each cross-member occurrence used here.
The compiler-producing contribution of BootstrapCompilerBuildTFS and the application-producing contribution of ApplicationBuildTFS occupy different functional layers in this build-the-builder account. Their connection runs through the produced compiler and its participation in application-building Work.
The nested networks below instead group members for successively broader questions. Their four depths are not four functional layers or four levels of a modular structure.
Select the nested network values from those already established inputs:
Each named occurrence is independently established under its project predicate before selection. Each network applies its endpoint-binding and boundary-exposure constraints plus the acyclic direct-member constraint, and each keeps the use frame in its row. The local names select or add nothing by themselves.
No claim about who selected these networks is required. If the case also needs CompilerNetworkSelectionWork-5, cite each precise performer's independently established A.13 core and the Work's independent A.15.1 admission. Add F.6 only if the case also needs attribution under the assignment in question; its assignment declaration and proof remain outside E.18.NET. Adding or removing the Work or attribution claim changes none of the four network identities above.
The result episteme may describe the selected structures and cite a separate selection or decision relation, but it is not a decision or accountability relation by form. Any accountability claim needs the predicate and participants of that accountability relation.
A compiler-production case can close on separately grounded identity inception, production completion or readiness, evidence, and decision while naming the application-build position as the downstream use outside that closed case. Project-level reasoning continues into the member where the compiler later participates.
The same joint-selection question recurs for a builder system: select the TFS in which that admitted builder performs the Work at issue together with the independently identified TFS or nested network concerning production and identity inception of the builder, or its later change after it exists.
Shared identity creates no edge; use obtaining production, inception, participation, application, use, or other relation occurrences and their endpoint bindings.
The bootstrap compiler result is exposed from the outer network through one finite member path:
Each path entry is a direct member of the preceding network, the final entry is the TFS named by leafFlowPositionRef, and no network repeats. FlowValuation, path slices, and DesignRunTag remain leaf-local. “Builds”, “uses”, “evaluates”, and “delivers” are ordinary cues until each link resolves to an admitted relation kind, complete participant meanings and order, obtaining occurrence, and endpoint bindings.
Before these identities and relations are grounded, A.1.STM may show the dependency only as a Plain provisional long-mantra map and must name the missing member, the exact relation-claim result returned by its governing pattern, or the separate missing occurrence, endpoint, or position binding. It is not yet an E.18.NET selection.
Once the network is admitted, a separate A.22.CGUS demonstrative slice may traverse admitted positions and relation-reference epistemes; it remains a demonstration, not the project, network, case, or Work order.
N-ary relation and feedback cycle
A manufacturing release relation has three participants defined by one admitted domain relation pattern: one product-definition position in a TFS selected to answer the development question, one equipment-readiness position in a TFS selected to follow the changes that establish equipment readiness, and one release-condition position in a TFS selected for assurance. Keep the three participants in predicate order in one network row. Three unlabeled arrows would lose the ternary relation.
Later, an observation from use reaches a development position. If admitted cross-member relations also lead from that position to the operating position, these connections form a directed cycle. To establish operating feedback, trace the observation into a subsequent development decision that governs operation. For example, the engineer used the reported failure rate to retain the current setting under the applicable rule. A rate merely stored in a log establishes receipt or storage. The network can retain its directed cycle while the decision follows an independent schedule. Direct-member nesting remains acyclic.
Architecture and two demonstrative boundaries
For one containing holon, an independently obtaining C.30 ArchitectureRelation may have the exact network as its selected structure. A bounded C.30 claim Q about one named holon H and the independently selected network N may instead remain negative, unresolved or candidate/expected under C.30.TFS-REL's single-holon claim branch; retain Q's basis and next return.
If the selected members belong to separately named holons and no containing bearer is grounded, use the inter-holon branch and name the participating architecture claims. None of these architecture uses supplies a missing network discriminator.
A Plain A.1.STM long-mantra map may display proposed members and a missing cross-member link before network admission. It names the intended final result and the absent member, relation kind or predicate, predicate result, occurrence, or endpoint binding; it asserts neither an E.18.NET structure nor a CGUS.
After the network is admitted, a separate teaching mantra may show one finite admitted dependency slice. The slice uses the network locator family, cites admitted positions and exact relation-reference epistemes, and keeps omissions and return visible. It does not prescribe project Work order, make the path the whole network, or turn a leaf-local DesignRunTag into a project phase.
Bias-Annotation
These five bias risks apply across uses of this pattern: Gov, Arch, Onto/Epist, Prag, and Did. The table gives a mitigation for each.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Adoption test: use E.18.NET only when the current question needs independently identified members and at least one obtaining relation that connects positions across their boundaries. If one TFS or one parent-relative SubflowRef answers the question, there is no need to add a network, endpoint bindings or member paths.
Rationale and naming
TransformationFlowStructureNetwork names a recursively selected organization over independently identified transformation-flow structures or networks and exact cross-flow relation occurrences, preserving member boundaries and locally exposed positions. Members are structures rather than their valuations, and nesting may have any finite depth. The Plain name is “network of transformation-flow structures.” The shorter cue “transformation-flow network” serves retrieval after that subject is clear.
SoTA-Echoing
Each line below is inherited only while the cited current pattern version retains both the named body decision and the named source-use row for its declared use. E.18.NET relies on the currentness decision recorded with that source-use row; it does not independently turn the cited literature or tool practice into current authority.
When one cited source-use row changes, reopen only the affected line here.
A new need for cyclic member identity, a separately re-identifiable membership occurrence, or cross-flow semantics that cannot preserve the direct relation and its endpoints reopens the E.18.NET architecture decision itself.
Relations
Builds on: A.22 for selected-structure identity and non-agentivity; E.18 for one TFS, its selected internal relations and resolved connection positions, FlowPositionRef, valuations, paths, slices, and local state; A.6.REL, A.6.RCD, and A.6.P.WMR for exact relation recovery and missing-governor; C.2.1 for the optional descriptive record; and F.18 for the stable local name.
Coordinates with: A.15.6 for actual project Work, project system-of-interest designation, and subject- or claim-centred case closure; A.1.STM for a Plain provisional long-mantra display and backward/forward attention use; E.18.2 and C.29 for mathematical descriptions; A.22.CGUS and E.18.3 for admitted demonstrative slices; C.30.TFS-REL for architecture use; C.32.CONWAY for one qualified architecture-influence pair; A.3.4, A.12, and the A.15 family for actual transformation, causal or acting positions, Work, production, and work-to-change claims; E.17 for publication; E.11 for public entry and recognition; and E.11.PUA for using one already selected pattern to reach its first useful result.
Does not replace: the direct pattern that defines or constrains any selected production, use, participation, evaluation, feedback, dependency, correspondence, supply, evidence, assurance, gate, decision, causal, or work relation. E.18.NET selects already obtaining occurrences for one network use; it does not define their kinds or make them obtain.
E.18.NET:End
Pattern Quality Gates: Review and Refresh Profiles
Type: Architectural pattern Status: Stable Normativity: Normative
Use this when
Use E.19 when one new, substantially revised, or aging FPF pattern edition or bounded subset needs a repeatable admission, refresh, or return-for-repair review. Its profiles specify the questions to ask and what each conclusion means. A reviewer applies the selected questions and returns either repaired text with focused verification or actionable findings.
Use it especially when a draft looks structurally compliant but may fail on first-minute usability, primary EntityOfConcern stability, terminology, SoTA grounding, related-pattern boundaries, examples, anti-patterns, or shipping-facing authority claims.
Not this pattern when. Use E.8 to write the pattern body. Use E.9 to record the content decision that explains why FPF should change. Use E.9.DA when the question is whether one exact DRR is adequate for a declared downstream authoring use before drafting or host amendment. Its ordinary result may be precise findings or repaired text, while exact C.2.1 and coordinate-result apparatus is conditional on a requested reusable result or named reliance.
Use E.21 for ordinal pattern-quality evaluation of one pattern version. Use E.23 when the aim is repeated quality improvement against an object-under-improvement evaluation rather than one admission or refresh review profile.
Use local patterns for the domain rule or constraint being reviewed. Use project gate or release patterns when the question is whether a project publication, work-result record, or release candidate passes a delivery gate. E.19 reviews admission or refresh of FPF patterns. Its profiles and results do not certify a state of the world, project, publication, or release.
What goes wrong if missed
Review collapses into heading compliance or personal taste. A draft can pass because it has the right headings while being hard for a practitioner to recognise, too thin against current practice, unclear about its primary EntityOfConcern, relation record, or claim record, or misleading about related patterns and the authority each pattern's content carries.
What this buys
E.19 gives authors, reviewers, and stewards a shared review profile: what must be checked, how deep the check should go, which defects block admission or refresh, and what evidence is needed before a pattern-quality claim is made. It also makes the recognition text visible before the more detailed assurance checks.
First useful move. Recover what the intended practitioner must be able to do from the request for the pattern or the practice it serves. Compare that work with what the named edition or subset actually teaches. Use the difference to frame the admission or refresh question and select PCP-BASE and the risk profiles it needs. Inspect the affected passages, then repair and verify each defect or return the actionable findings.
Local-repair boundary. If baseline triage shows that the review question has no ontology, usability, SoTA, boundary, naming, or authority risk beyond a small mechanical repair, close with that repair direction. Apply further profiles only for risks the question raises. Claim an E.21 quality value only if E.21 has evaluated the pattern version over its required coordinate set.
Three situations for recognizing which questions to select
Primary EntityOfConcern in plain terms. One FPF pattern edition or bounded subset under an admission or refresh review question. The selected checks, reviewer, any repair, findings, optional aggregate result and evidence use, and any authority-bearing decision remain distinct when the review involves them.
Primary working reader. The first reader is an FPF reviewer, with the pattern author close behind. The review must be answerable to the eventual practitioner or manager who will rely on the admitted pattern.
Problem frame
FPF evolves by adding and revising patterns. Over time, the framework accumulates two kinds of risk:
-
Admission risk — a newly authored pattern can be structurally compliant yet fail on ontology, semantics, terminology conflicts and vagueness, scope, SoTA in related disciplines, or cross-context hygiene.
-
Staleness risk — older patterns can remain internally consistent while drifting away from contemporary practice and newer parts of FPF, current internal vocabulary, or updated related patterns and their defining or constraining content. The pattern can still appear clear to the reader while becoming misleading, incomplete, or incompatible.
FPF already contains many checklists and constraints, but they are distributed across patterns and suites. Authors and reviewers therefore lack a common, repeatable way to decide what to check and how deeply before admitting or retaining a pattern.
Problem
Without a stated, common review method, reviewers and authors face four problems.
- Different reviewers focus on formal or template compliance and miss deeper ontological, semantic, and naming issues, producing bureaucratic output that does not improve the enforceable Conformance Checklist.
- Authors “optimize for the visible checklist” and miss hidden requirements (lexical discipline, Bridge hygiene, SoTA‑Echoing quality, scope claims, delta‑class impact).
- Older patterns accumulate conceptual staleness and diverge from current practice, current terminology, or current internal invariants.
- Compliance comes to depend on reviewer taste. That makes the specification’s normative content harder to trust and prevents a repeatable review decision.
Forces
Solution — Profile-based gates for admission and refresh
Establish Pattern Quality Gates (PQG): a family of profiles that declare admission and refresh checks. Different reviews can combine different profiles instead of applying one checklist to every pattern.
A Pattern Check Profile (PCP) is a named bundle of check families. Profiles are additive: every review configuration includes the baseline profile and only the risk-driven profiles needed by the declared question. A PCP specifies questions and closure conditions; the reviewer applies them and returns findings or repaired text. An unselected profile requires no result row or durable disposition.
Choose review depth from the harm if a defect survives, the novelty and complexity of the claim, how widely the pattern will be reused, and how likely its sources or neighbors are to change. Pattern length, official status, and the number of available checks do not justify deeper review by themselves. Use cheap automated or template checks for properties they can test, then spend reviewer attention on semantic, ontological, practitioner-use, and current-source questions they cannot close.
Editorial and runtime gates. PQG and PCP are editorial review constructs in the authoring plane (Part E). They are distinct from enactment and runtime gating constructs such as OperationalGate(profile), GateProfile, and GateDecision (A.21), which govern Work transitions and gate decision policies elsewhere in FPF.
For an ordinary bounded review, keep the reviewed edition or subset, question, selected profiles, checked loci, defects or repairs, and conclusion. When exact replay or a named later use needs a stronger account, also keep independently recoverable:
- the exact reviewed FPF pattern edition or bounded subset and the declared admission/refresh question;
- the review configuration: baseline and risk-selected PCP declarations, exact question scope, use, qualification window, and stop boundary;
- the semantic review
U.Method, when that identity matters; call an episteme itsU.MethodDescriptiononly after it passes A.3.2; - for each actual review, repair, or verification occurrence asserted as dated
U.Work, recover every actual performer through A.13 and use A.15.1 to identify its time, Method, containing System, and Work independently. Add F.6 only when the review account expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment. That attribution must be independently grounded. Holder identity or timing does not establish it. A missing or failed F.6 link alone leaves independently admitted Work intact. Recheck any A.13 or A.15.1 premise invalidated by the same finding and its dependent conclusions; - each PCP check application and A.6.1 binding only when the receiving use must replay those bindings;
- any distinct authoring/repair work, changed pattern edition, and focused verification work/application in inspect-repair-verify form;
- actionable finding or blocker claims, focused-verification claims, and one C.2.1 aggregate E.19 review-result episteme when a durable conclusion is required;
- any separate authority-bearing admission, refresh, return-for-repair, or waiver decision and its decision work;
- witnesses and the support the claim requires, including A.10 when an implicit or required-replayable source-to-use account is needed and any B.3 result the selected assurance claim requires; and
- any F.10 status use, publication occurrence or form, carrier, and currentness relation used by the receiving claim.
Any local system-role kind and its independently evaluated classification are optional separate claims. Neither supplies assignment or performance. Route unresolved source role through E.10.ROLE. Name the intended-reader position or the representation position. When a later claim relies on a dated occurrence, apply item 4 and CC-E19-0.
In Plain, review run refers to the configuration, the reviewer’s actions and their results. The §4 account keeps the declarations, applied Method, actual review work, findings, result, and any authority-bearing decision distinct when those identities are needed.
Define the reviewed pattern or subset
Name the reviewed pattern or bounded subset, its edition or other stable version basis, the admission or refresh question, the selected profile questions, and the review boundary. That is enough for an ordinary bounded review. Add exact scope, window, and review-configuration identities only when a receiving result or named reliance needs them. The profiles determine which questions to ask and how deeply to review them. An ordinary bounded review requires no progress record.
When a reusable result or named reliance depends on how the review was enacted, apply the item 4 actual-Work account and CC-E19-0 to each asserted review, repair, or verification occurrence.
If a durable aggregate result is needed, constitute a C.2.1 result episteme whose EntityOfConcern is the reviewed pattern edition or subset and whose ClaimGraph states the review scope, applicable profile questions, actionable findings or aggregate cleared boundary, conclusion, and reopen condition. Add a non-use boundary only when it changes a named receiving use under the F.19 plausible-intended-reader test. Witnesses, evidence use, the optional result publication, and any authority-bearing admission or refresh decision remain separate.
Choose inspect-repair-verify when the reviewer may edit and same-turn repair fits the declared use. Choose independent findings when the review needs separation from the author or an unchanged candidate. Independence changes who edits. The selected questions remain the same, and no additional dossier is required.
Choose one review form.
- Inspect, repair, and verify. One bounded review may include inspection, repair, and focused verification. A reviewer performs those actions. Distinguish their performer, Method, affected object, or occurrence only when the positions differ or a named later use needs them. Apply item 4 and
CC-E19-0if the account asserts dated Work. Apply every selected question, repair every in-scope defect, and reapply the affected checks. The changed edition and focused verification carry the substantive evidence. Constitute an aggregate E.19 result episteme only when a receiving admission or refresh decision requires it. Make a separate findings record only for an unresolved blocker, a decision outside current authority, or transfer to another author. - Independent findings. A reviewer applies the selected questions without changing the reviewed pattern or subset. One C.2.1 findings-result episteme or handoff file records every actionable defect and blocker, with repair direction precise enough for the author to act without repeating the diagnosis. It is neither the reviewing action nor an admission decision.
A selected question that reveals no defect requires no durable pass entry. Independent review records actionable findings without compiling positive recitals. In repair mode the repaired text and its verification show the completed repairs, with no parallel findings record.
If another pattern defines a reusable value or decision required by the declared use—such as an E.21 coordinate, a DRR decision, or a landing result—that value belongs to the result required by that pattern rather than to an E.19 progress account. E.19 specifies the substantive questions and outcomes independently of how a working environment keeps place during the review.
Complete the selected scope. Inspect every independently answerable question in the declared baseline and risk-selected scope. The first defect, blocker, or already-negative admission conclusion may prevent a positive verdict, but it does not complete the review and does not suppress findings that remain independently obtainable. Stop before the selected scope is complete only when a missing source, missing authority, unsafe boundary, or equivalent condition makes the remaining questions impossible to judge truthfully or safely. In that case, identify the unexamined scope and the reason it could not be judged, and report the findings as partial.
A nontrivial pattern-quality review SHOULD state its quality-evaluation purpose before depth is selected. Use E.22 or an equivalent compact question frame to say whether this review is a floorEvaluation, exceptionalImprovementEvaluation, paretoTradeoffEvaluation, openQuestionDiscoveryEvaluation, absorptionEvaluation, or a declared combination. If the purpose is absent, E.19 treats the review as an admission-refresh blocker read, not as a request to raise every evaluated coordinate toward exceptional expression.
When coordinate values, PatternQualityStatus, or all-4/all-5 claims are needed for one pattern version, the review opens or consumes an E.21 result instead of assigning those values inside E.19.
When the review opens or consumes E.21, E.19 treats E.21 as a hard pattern-quality evaluation, not as a selectable profile. The review must not accept an E.21 claim that omits required coordinates, omits ShortRationale, omits PrecisionRestorationProfile, uses inactive/triggered-coordinate language, narrows the requested use to make the result pass, or replaces coordinate values with blocker triage. In repair mode, repair or re-evaluate the affected result where that work is in scope. In independent findings, identify the defect.
Baseline triage can answer only the E.19 review boundary when no E.21 quality value, all-4/all-5 claim, landing-quality claim, or pattern-improvement movement claim is being made.
If the aim is repeated improvement against an object-under-improvement evaluation, use E.23 for the repeated method. An E.19 review configuration may supply PCP questions and its result episteme may supply findings inside that loop, while E.23 supplies the repeated Method. An E.19 result gives no ordinal quality value. Only a separate E.21 assessment application and result episteme can state the E.21 coordinate values for the changed pattern version.
In E.19, the reviewer judges the quality of an FPF pattern or bounded subset. E.19 governs FPF admission, refresh, return-for-repair, blocker, and review-profile claims, not E.21 coordinate assignment and not project-side publication interpretation, explanation interpretation, comparative review-unit use, or participation in a named project-side review relation. When those project-side relations are used, use the publication or project-side pattern that names the object being interpreted or reviewed.
Project-side reuse boundary. Use this boundary when an E.19 review-result episteme is cited for project certification, evidence, safety or compliance assurance, gate input, release justification, work authority, or publication truth. First identify the FPF pattern-quality claim it states: admission, refresh, repair return, or selected pattern-quality boundary.
Any project-side reuse then opens the concrete relation that governs that use: A.10 for an implicit or required-replayable source-to-use account, G.11 for source currentness, B.3 for assurance, F.10 for status use/interpretation, A.20 for a current local CV status when applicable, A.21 for gate decision, A.15 for work, or the relevant project-side pattern.
The E.19 result may be evidence about FPF pattern quality; it is not certification of the project world. Plain wording in the reviewed text remains ordinary unless it changes admissible use, evidence, gate, assurance, work, decision, status use, or FPF pattern application. A project refusal or approval requires a project-side governing relation that states the project claim and its admissible use.
Give semantic or ontological defects and Solutions that fall short of current SoTA priority over formal or template defects, such as failure to follow E.8 structure or RFC deontic terminology. In repair mode, fix the formal or template defects within the declared boundary. In independent findings, identify them and say how to repair them.
For example, if the header block is missing or incomplete, continue with ontology and semantic review first. Treat missing header fields as one mechanical defect, not as a reason to stop (PCP-BASE #7).
When a proposed or accepted pattern change needs a best-known Delta-Class (Δ-0…Δ-3) and initial impact radius, place them in the governing change, decision, or landing result using E.15's actual-effect and actual-dependency tests. E.19 repairs or reports an omission that matters to the selected review; it does not copy a successful change account into a second review record.
Apply the baseline profile to every run
Every run MUST include PCP‑BASE as a triage baseline. Full-depth checking is selected only where the relevant risk is present; reviewer depth SHOULD prioritize the FPF-governed sections and enforceable requirements in E.19:4.2.1.
- Internal coherence (problem <-> conformance claim <-> solution) Follow the same practical question from its source through the Problem, Solution and Conformance Checklist. Could the pattern satisfy every item in its checklist while its user still needs someone else to make a choice or produce a result it promises to teach? For a method of designing a whole, check where its user chooses the parts, determines how they work together and establishes what the arrangement will provide. If the commission specifies that arrangement, correct application of it may be sufficient. Compare the Solution with the checklist in both directions for omitted and orphan requirements.
- Lexical discipline & reserved vocabulary Terms and registers follow lexical rules; ambiguous "everyday" synonyms do not silently replace kernel vocabulary.
- SoTA-Echoing minimum compliance (E.8) SoTA-Echoing satisfies the E.8 authoring requirements applicable to the pattern kind (Architectural vs Definitional), including stated adopt/adapt/reject stances and the E.8 two-part SoTA test: current best-known problem-solving practice for the named practice question, and by-value incorporation into FPF-governed pattern loci. If a SoTA Synthesis Pack exists for the topic, SoTA-Echoing binds to it rather than forking an untracked narrative. Any divergence of pattern norms from contemporary practice is explicitly stated as such. SoTA-Echoing MUST be non-decorative, MUST reflect best-known current practice rather than official status, source recency, institutional adoption, or merely popular defaults for the declared problem, and MUST govern the Solution and other FPF-governed sections, or those sections MUST justify divergence explicitly.
- Cross-pattern compatibility & impact radius Relations are consistent with declared dependencies and dependents. Declared scope and impact are compatible or explicitly limited.
- Didactic grounding Apply E.8:4.2.1.3 to the connected explanation and its cases. Recover the common relation or Method being taught, its grounds, and the correspondence to the example. For a Method example, distinguish the reusable way, the depicted performance and its subject material. Inspect whether the cases make the material variation of the declared use intelligible; a correct result in the example's domain answers only that part of the review.
- Reader-fit
The pattern body addresses the intended FPF user in the working role governed by that pattern. FPF developers, package architects, reviewers, and evaluators are appropriate readers when they occupy that role. FPF-governed sections explain in user terms admissible use, costs, boundaries, the concrete definitions, constraints, tests, or other contributions used from FPF patterns named by value, and the project-side FPF kinds, references and related relations also named by value. Architecture placement, freeze or merge state, package-boundary rationale, reference boilerplate, quality or projection evidence, corpus-entry evidence,
PatternQualityStatus, monolith-parity evidence, landing evidence, and broader package-development rationale stay inDRR, architecture documents, review handoff,E.21result,E.19findings, README, ToC,E.11, cards, retrieval or projection carriers, release or landing evidence carriers, companions, or ordinary references unless they change the working reader's first admissible move. - Template & section integrity This is lowest priority for review depth and SHOULD NOT consume effort that would displace ontology, semantics, modularity, slot discipline, or SoTA checks.
- Modularity & contradiction hygiene
The pattern SHOULD NOT be overloaded or significantly expand requirements or dependencies without an explicit reason and impact record.
Checks include: scope containment, split/refactor recommendations when warranted, and contradiction scans against neighbor patterns in Relations.
The pattern SHOULD balance cohesion and coupling across FPF.
Identify which kind or declaration owns each slot or parameter. For a subkind use the membership criteria of
U.SubkindOfin C.3.1. For a specialized mechanism use its own declaration rules (A.6.1:4.8). A slot added by the specialization does not become mandatory in its parent. UseUsesfor a dependency on another pattern. When two models of the same work or system are compared, B.5.MR establishes what their assertions preserve and whether a previously admissible answer is excluded. - Substantive solution and locus adequacy For the reviewed pattern, ask a short set of questions about its problem and the change under review. Does the pattern still solve the stated problem? Are the decision points and applications of the relevant patterns correct? Are kind boundaries and selected companion or projection functions preserved? Has anything worsened? Are the SoTA rows current enough for the claim they discipline? Does that claim have enough support without excess material?
- Triggered method, performer, work, and result separation
When a Solution says how to do work, first distinguish content defining, constraining, testing, or guiding a Method from a claim that dated Work occurred or the world changed. Method guidance alone does not trigger a fictive performer or Work. If an account asserts dated
U.Work, verify the §4 actual-Work account. If it asserts a world-side change, identify the change relation, the pattern that defines it, and the things it relates. Keep the intended-reader position, any qualifying A.3.2 method-description episteme, actual performer, Work, and problem-facing result separate. For a literal datedU.Workclaim, return a finding when an episteme, checklist, plan, prose, or intended-reader or representation position is made to perform Work, or when Work and result are collapsed. Judge ordinary or metonymic wording through the complete-claim test inF.19; a familiar instrumental expression alone does not require a formal Work account.
Where to focus baseline review
Focus review effort where it can improve semantic and ontological trust. Keep routine checks feasible by giving the following parts different depths.
- Treat FPF-governed sections and deontic requirements as the primary depth loci:
- the pattern’s Problem frame, Rationale, and worked slices when a new family, profile, or specialization would otherwise be intelligible only from project context,
- reader fit in Problem, Solution, Consequences, Rationale, and worked slices whenever the draft risks mixing user guidance with package-development rationale,
- the pattern’s Conformance Checklist (the enforceable conformance check set): keep items universal and easy to use, limit prohibitions to what is needed, and leave checks that belong to another pattern there,
- deontic clauses (
MUST/SHALL/SHOULD/MAY) that define requirements on the authoring/validation plane (not laws of nature or mathematical facts; ensure an explicit conformance subject), - admissibility constraints (
Invariant:/Well-formedness constraint:) that define valid models (cardinality, typing/kinds, totality) and are written as non-deontic predicates (no RFC keywords inside the predicate), - definitions and mint/reuse decisions (new terms, renamed terms, scope claims baked into names, avoiding overloaded names and choosing names that fit the intended use),
- cross-context and cross-plane claims (Bridge hygiene and “sameness” assertions),
- SoTA (when the pattern claims state-of-the-art rather than a popular-but-outdated solution or vocabulary),
- substantive solution and locus adequacy: one reviewed-pattern-specific content pass checks whether the repaired text still solves the stated problem, assigns claim-bearing material to the correct governing loci named by value, preserves kind boundaries and selected companion or projection functions, keeps quality/projection evidence and executor/reviewer correspondence out of the pattern unless the pattern's own
EntityOfConcernand user-facing action are that evaluation/projection work, and has not become either under-grounded or over-bureaucratic, - modularity and Slot discipline of A.6.5 that provide evolvability of FPF,
- absence of contradictions in a pattern,
- Relations that define compatibility and impact radius.
- Check mechanical corrections briefly when their scope is limited to the named mechanical property and their semantic and practical effects are unchanged, for example a micro-typo or heading-format correction. For live recoverability or contribution questions in stylistic or narrative rewriting, including RFC-form deontic cleanup, use the whole-span
F.19reading below. Automate a check only for a named property the tool tests. A clean result establishes that property alone. Semantics, ontology, practical usefulness and source currentness need their own grounds. - Do not block semantic review on template and RFC compliance defects. Missing header block fields (E.8 H-5), missing canonical sections, or a missing footer marker are fixable integrity defects. Record them as repair items and continue with the FPF-governed section checks in the same run.
- Whole-span precise language. Reviewers SHOULD apply the complete connected Solution in
F.19to the selected FPF-governed span: recover meaning, test the contribution of each optional expression, and compare useful content before and after repair. Retain its precision-before-coarsening order, cold intended-reader recovery, evidence-selection and coverage conditions. For a claim spanning kinds of holon, composition scales or levels of model abstraction, use F.19:4.3.1.1. If the text supports only a narrower conclusion, apply C.39.GA when it promises the broader argument. If an FPF meaning remains unresolved, use E.10 or its subject pattern. - Precision-restoration distribution must be preserved. Apply
CC-E19-21; keep only review-specific questions here and use the declared language or subject owner for the repair. - Review-specific continuity questions. Apply these to the changed claim and affected uses:
- Is the pattern's own
EntityOfConcern, first useful move, practical delta, and any action-changing applicability boundary recoverable, with its action guidance before auxiliary wording, publication, architecture-placement, package, or quality apparatus? - After wording or reference migration, does the claim still reach the same referent through the intended slot or reference position and alignment path? Record any deliberate retargeting in the governing change decision.
- When phrase apparatus, semio bias, architecture placement, package rationale, or quality apparatus changed, did the repair preserve the function the claim needed and remove only the displaced apparatus? Name each outside definition, constraint, or test by its supplying pattern and use a formal identity only for a live distinction or named reliance.
- Do the affected current consumers still receive the intended meaning and use? Resolve semantic, mechanical, or compatibility changes in the affected sources; report unresolved conflicts rather than creating a disposition for every unaffected consumer.
- Is the pattern's own
- Use preservation and guard selection are different decisions. Always compare the admissible uses of the old and repaired claims under their governing rules, including any expansion or narrowing. The
F.19plausible-reader test decides whether an explicit description, publication-use, or non-use guard deserves mention. A justified guard still undergoes the same before/after use comparison. UseF.19and the patterns that define or constrain the Method, Work, evidence, assurance, gate, status, decision, or unresolved role claim. ApplyCC-E19-0to dated Work.
When E.21 is active, its PrecisionRestorationProfile carries the quality result; E.19 does not duplicate it.
- Design-time and run-time both count. The same precision discipline applies to FPF pattern prose and to any reviewed publication text, worked slice, or performed-work exemplar when that text is being assessed for admissibility, guidance, reuse, gating, release, policy, assurance, or action-selection use.
- Report ordering (impact-first). In run outputs and remediation direction, give priority to ontology, semantics, modularity and SoTA findings in FPF-governed sections. Group low-signal formatting/typos into one compact tail finding unless they change meaning.
Add risk-driven profiles
PCP‑PRAG (Pragmatic utility & adoption) — Trigger: the pattern is Normative and claims practice guidance. Check these connected uses:
- Recognition and first action: an early
Use this whenor equivalent, a recognizable working situation, the cost of missing it, practical gain, first admissible action and ordinary non-use boundary. - Performable guidance: a worked application that makes the Method's own choices, intermediate results and continuation recoverable; useful Consequences and Anti-Patterns; and a clear adoption test. Select additional materially different applications when the declared use requires them.
- Subject and reader: a short account of the primary EntityOfConcern, relation or claim record and any minimal modeling lens needed by typed declaration content. Give early technical terms nearby plain glosses. When several reader situations are served, state the primary working reader, concern and viewpoint.
- Recognition and assurance: put first-use guidance before the declaration, modeling and review/check material needed for assurance. Keep its scope consistent with the recognition claim.
- Grounding and breadth. Connect explanatory SoTA rows to the working decisions they inform. Locate the grounds for the claimed range, including a general argument made within one example. Use materially different recognition situations when the intended readers must recognize a use under differences the available cases leave unresolved. F.16 supplies a compact worked-example format.
Use C.39.GA when an explanation gives a sound answer for a narrower case or class and promises a more general method. Locate the first inference the recipient would have to supply, restore its grounds and check the conclusions that use it at the promised reach. An argument developed within one example can suffice. E.21 supplies scope and case-coverage coordinates when the decision needs a quality value.
When admission or refresh includes precise-language repair, apply CC-E19-7a. It preserves practical guidance and the Plain/Tech relation under E.2 P-2 and E.12, with formal identity and dated-Work checks only under the conditions stated there. F.19 governs the whole-span repair; E.10 supplies compact cues and FPF routing.
For a broad cleanup across several patterns, or any cleanup that touches FPF-governed Problem frames, Problem sections, first-use recognition text, archetypal grounding, examples, or worked slices, check whether the didactic function was harmed. In inspect-repair-verify, restore the working situation, first useful move, and the definition, constraint, test, or other pattern contribution needed by the claim; in independent findings, record the exact harm and repair direction. A positive improved or preserved account is required only when another evaluation makes that value one of its substantive results, and it belongs in that evaluation.
PCP‑MOD (Modularity and abstraction-boundary discipline) — Trigger: the reviewed pattern or subset shows scope creep, mixes rules with different ranges of applicability, or conflates kinds or relations (for example, a pattern bundles universal core rules with frame-specific content and discipline-specific method semantics, or it mixes EntityOfConcern, Description, and Specification positions in one object).
Checks include:
- an explicit core vs extensions cut (universal invariants are factored into one stable “core”, and extensions reference it rather than re-stating or mutating it),
- no conflation of specialization, model refinement and dependency: apply the defining specialization rule when a kind or declaration specializes another; use B.5.MR for a claim of model refinement or abstraction, and
Usesfor the declared use dependency. State which relation the comparison requires, - no conflation of package-form, concrete pattern-to-claim contribution, and package-relation functions: Pack vs Kit vs Suite vs Family vs Bundle vs Cluster vs Profile vs Overlay vs Record vs Umbrella are not interchanged, and the review states carrier status, the definition, constraint, test, or other pattern contribution used, and the package relation, preserving those distinctions across stylistic changes,
- description-lane descriptions and their publications do not grow mechanism semantics; for an MVPK face or projected publication form, no-new-claim checks that it introduces no claim beyond the selected episteme and no-shadow-default checks that it introduces no undeclared default. Keep the selected episteme, optional projection/construction, face, publication form, publication occurrence, rendering, and carrier distinct. The selected episteme has
U.Viewmembership only when exact E.17.0 conformance independently obtains; face status, projection, profile selection, and compliance with these two checks establish no membership or truth, - slot-discipline hygiene for any ordered specialization set: SlotKind invariance is preserved and inherited operations do not gain new mandatory inputs (A.6.5 / A.6.1 specialization discipline).
PCP‑REFRESH (Staleness & compatibility refresh) — Trigger: there are signs of staleness, such as an outdated SoTA claim, a renamed or superseded relation, terminology drift, or an explicit refresh window in a current source-use, change, or decision record. Checks include:
- refresh-sensitive claims are identified and either (a) updated from the best current line of work for the problem with matching Solution changes, or (b) explicitly scope-limited and labeled as historical lineage. Source date, count, official status, or novelty alone does not establish current-best use,
- select living refresh only for a high-priority claim or pattern subset likely to change when new evidence or a changed neighbor appears. Monitor and reopen the smallest affected unit at a named trigger. When continued surveillance no longer justifies its cost, retain only a separately justified interval or trigger and any binding maintenance requirement; otherwise end that surveillance,
- Relations are updated to current pattern IDs; deprecations/renames are handled via explicit continuity notes (no silent relabeling),
- when a new or substantially revised pattern subset is being prepared for send or landing, inspect the related patterns, the concrete constraints or tests they supply, companion patterns, Relations entries, and monolith-backed pattern sections that may require aligned edits. Repair an in-scope mismatch or return it as a finding. The changed sources and the governing landing or release result show successful alignment. E.19 requires no separate pass entry,
- any long-lived companion, profile, check sheet, pattern-local companion row, review harness, or analogous selected non-pattern FPF kind-reference pair kept with the reviewed pattern or subset states its use question, the concrete pattern contribution or selected non-pattern FPF kind-reference pair it serves, admissible companion-only use, one real breakage if absent, and demotion or deletion condition when no such breakage exists.
- when the refresh causes Δ‑2/Δ‑3, verify that the governing change or decision result carries its actual-effect Delta-Class, actual dependent reach, and any DRR, focused verification, source-refresh, or F.9 consequence that the changed use requires under E.15, F.15, and F.9; repair or report an omission rather than copying a successful account into E.19,
A reviewer may override a trigger in rare cases. Override a triggered profile only when the risk that triggered it is absent here and a compensating check covers the live concern. When the override changes an admission, refresh, or other governing decision, place its reason in that decision basis. Otherwise E.19 requires no separate positive override account.
PCP‑NORM (Normative guidance integrity) — Trigger: the pattern introduces or changes normative requirements, introduces new conformance items, or shifts downstream requirements. Checks include:
- Delta‑Class (Δ‑0…Δ‑3) and impact radius are explicit (what breaks, who depends on this),
- requirements are testable in principle, scoped, and non-contradictory,
- downstream patterns cited in Relations are compatible with the new guidance.
- for a Δ-2/Δ-3 change, apply E.15's actual-effect and material-decision test to the DRR requirement; a new normative pattern retains its DRR. When a DRR is required, it cites the applicable actionable PQG findings or the repaired candidate and focused verification, according to the selected review form; pointers suffice.
PCP‑SOTA (Evidence and SoTA alignment) — Trigger: the pattern’s Solution asserts “best practice”, “state-of-the-art”, or introduces new synthesis claims. Checks include:
- each “best practice” claim or SoTA claim in the Solution has a stated link to SoTA‑Echoing rows (or to SoTA Synthesis Pack identifiers when used), so readers can recover its grounds. Those rows identify best-known current practice rather than popularity alone,
- the selected SoTA practice or source set answers the declared working problem and the relevant domain or practice tradition. Justifying package placement, naming or clustering alone is insufficient,
- each SoTA row changes at least one FPF-governed outcome for the pattern: what the user may do, a source-supported applicability or reliance limit, which FPF pattern application must be named, or a claim's eligibility for a named release, policy, assurance, gate, action-selection, or adjudication use. An explicit rejected reading follows F.19's grounded-guard test,
- novel synthesis retains its evidential status. Frame it as a scoped hypothesis with explicit limits, or substantiate the stronger claim through its source claims, support and conflict relations, shared dependencies and applicability under
B.1.3. Admission as normative guidance requires those grounds and limits. UseG.2when the receiving use requires broad, refreshable SoTA harvesting and its handoffs. Pack registration alone does not establish the synthesis as best-known practice. A merely explanatory SoTA note that leaves the FPF-governed sections untouched is non-conforming, - where traditions disagree substantively, the pattern makes the disagreement visible and states whether it adopts, adapts, or rejects each relevant source idea instead of silently selecting one tradition,
- retrieval or benchmark methods are used only when the relevant evidence relation is present; their dimensions do not become universal pattern-quality benchmarks,
- refresh‑sensitive claims (those likely to decay) are marked with scope limits, timespan notes, or lineage labeling when appropriate.
PCP‑BRIDGE (Cross-context or cross-plane reuse integrity) — Trigger: the pattern imports claims, terms or norms from another context, discipline or reference plane. Checks include:
- explicit Bridge usage where required (no silent identity by spelling),
- Congruence and loss are made explicit where applicable,
- any cross-plane reuse is explicitly acknowledged and its penalties do not leak into unrelated assurances.
PCP‑SUITE (Mechanism-suite integrity) — Trigger: the reviewed pattern or subset introduces or revises a suite-level Description that enumerates multiple distinct mechanisms (e.g., MechSuiteDescription or a suite specialization) and/or changes suite requirements, conformance pins, or suite protocols.
Checks include:
- the suite remains a Description-level object: it enumerates member A.6.1 declaration epistemes through exact
MechanismDeclarationRefvalues under A.6.7:4.1 and declares shared requirements/pins, but does not define mechanism blocks (OperationAlgebra,Transport,Audit, …) and is not used as a mechanism node, - membership has set semantics:
mechanismscontains no duplicates, and member order has no meaning. Express any intended ordering only insuite_protocols, - each step of a supplied
suite_protocolsresolves to its selected member edition and declaration-local operation under A.6.7 WF-MS-2. Test the handover the suite promises. One jointly admissible, accepted transfer can establish possibility. A guarantee must cover the outputs the producer may deliver under the agreed conditions. For alternatives listed separately, check their membership without requiring a joint execution, - the suite is not a family of implementations: it MUST NOT be encoded as a
MechFamilyDescription(families remain “many realizations of one mechanism”, not “many mechanisms”), - the suite creates no transport exception. Apply
A.6.7:4.2according to the claim. An F.9 semantic correspondence needs its obtaining Bridge, a C.3.3 kind correspondence needs its own basis, and a separately claimed plane relation remains governed by the rule defining that plane relation. A changed entity, edition, plane or notation alone creates none of those relations. Retain every activated channel's references and receiving-use conditions underG.Core:4.2.3. Loss assessment follows the receiving-use rule, numerical assurance penalties satisfy A.6.7:4.2(2), and the suite carries pins rather than CL/Φ/Ψ/Φ_plane tables, - CG/CN authority pins remain explicit references to the single governance card and legality gate: if suite protocols include numeric comparison/aggregation/scoring, they cite
CG‑Spec(SCP + Γ-fold + MinimalEvidence) and (where applicable)CN‑Spec, rather than duplicating “local CG‑Spec-like” content, - suite protocols contain no hidden tails: if UNM/UINDM/ULSAM are required, the protocol expresses them as explicit
Usessteps and suite audit requirements cite the chosen mechanism ids/refs (no “implicit normalization/aggregation inside score/compare/select”), - gate separation is preserved: mechanisms and guards use tri-state
GuardDecision := {pass|degrade|abstain}and MUST NOT publishGateDecisionorDecisionLog;blockremains gate-level only (OperationalGate(profile)), - defaults remain single-sourced: portfolio mode, dominance regime, and unknown/failure behavior are either pinned in
TaskSignatureor one policy-assignment record, or not claimed; the suite does not define competing defaults, - when the suite claims reusable outputs, publish/telemetry is an explicit external step of the surrounding protocol, governed by the rules for that use: G.10 for Part-G pack shipping and E.24.PUB for episteme availability, form or carrier claims. Under
A.6.7:4.4, it is not aProtocolStepin the suite-member sequence or a hidden tail inside selection.
PCP‑P2W (Planned baseline & slot-fillings seam integrity) — Trigger: the reviewed pattern or subset introduces or revises planned-filling content in one exact U.WorkPlan against an exact governed declaration member, including a publication or view of that content.
Apply the planned-filling rules in A.15.3 to the changed plan content and its affected consumers:
A.15.3:4.0–4.4govern declaration-local PlanItem content, declaration and member recovery, intended-performance and planned-value designation, target-declared cardinality, and positive intended-use meaning. Use the correspondingCC-A15.3-01andCC-A15.3-03…09questions.A.15.3:4.2,4.5, and4.6govern conditional reference/policy pins, independently established actual use, baseline-preserving comparison, and read-only publication. UseCC-A15.3-11…14for these uses.A.15.3:12a–12bsupply the ordinary A.15.2 plan-content exit and the exact missing-source blocker when reusable typed use is needed but cannot be supported.
The declaration's own pattern defines member meaning and actual-use predicates; A.15.2/A.15.3 define the planned intention. Under those rules, review the use that changed, retaining the exact declaration and WorkPlan editions on which that use relies.
PCP-TERM (Terminology & naming protocol) — Trigger: the pattern introduces new terms, new U-kind pressure, new governed value names, new “unified names”, redefines existing labels, leans on FPF-governed phrases whose head kind or qualifier claim kind or admissible-use boundary is not yet restored, or uses FPF-governed trigger wording as if the word itself carried the needed kind. Checks include:
- the “mint vs reuse” decision is explicit when a term is introduced or changed,
- naming follows the local-first naming protocol and avoids scope smuggling (role-word meanings, metrics, or stages baked into labels; overloaded words used as terms with a local sense). Remediation SHOULD use F.18 when its durable-name use condition applies,
- when
F.18winner selection andA.6.Pfollow-through are both needed under their respective use conditions, treat them as one chain: inspect the candidate heads or phrases, kind conflicts, lexical conflicts, selected wording, and survival of the repaired phrase; repair a broken chain or identify the defect. A successful chain needs no pass entry, - use the semantic-area cues in
E.10:0.2with F.19's whole-span reading. The accepted sentence itself or its governing declaration must make the relevant object, value frame, relation, work, authority reference, pattern application, publication kind, companion function, or conformance claim recoverable; repair or report any case where it does not, - for unresolved generic heads or claim-bearing qualifiers, and for a subsequent comparison, escalation, downgrade, or other use that depends on how the phrase is interpreted, apply
F.19:4's precision-before-coarsening rule, - when repaired wording still carries an architectural claim kind or admissible-use boundary, verify that the resulting primary
EntityOfConcern, first useful move, outside work, and anyE.10.ROLEdisposition or package-form decision remain recoverable in the repaired text or the decision that set the boundary; repair or report a mismatch, and - source-side old wording and continuity rules are respected.
PCP‑DEONT (Deontic clause hygiene: RFC keywords) — Trigger: the pattern conflates admissibility/validity constraints with deontic obligations (for example, by using RFC keywords where a non-deontic Invariant: predicate is required).
Checks include:
- Deontic requirements are expressed with RFC-style keywords (see H-8);
- obligations are not smuggled into prose as informal imperatives. Admissibility/validity constraints are stated non‑deontically as
Invariant:/Well‑formedness constraint:predicates and referenced from the Conformance Checklist when enforceable. - Subject discipline for RFC keywords. If a sentence uses RFC keywords, its grammatical subject MUST be an agent or a published record or model whose required content is being constrained. State modeled-world admissibility or validity requirements as
Invariant:orWell-formedness constraint:predicates and reference them from CC items when needed, under E.8 H-8 andCC-SG.4.
PCP-ENTRY (Pattern-entry discoverability and entry-orientation changes). Trigger this profile when a change materially affects how a reader recognizes, selects, rejects or reclassifies a pattern body that directly governs the case, an applicable projection function, a set of patterns to compare on entry, a Problem-frame recognition signature, an expanded entry-disambiguation case or a lexical-query cue.
Trigger classes
PCP-ENTRY is a risk-selected editorial profile in the PCP family. Choose one lead profile for the change and add others for the specific failures they address.
Classify the change using the following cases.
-
Trigger class 0 — micro-edit punctuation, formatting, typo repair, grammar, or meaning-preserving compression with unchanged pattern-selection effect. No
PCP-ENTRY, no compact pattern-local note, no evidence mode, and no parity scan are required. -
Trigger class 1 — local recognition wording repair one improved
Use this when,Not this pattern when, or one removed sequence-implying phrase with unchanged candidate-pattern set and unchanged governing-entry or applicable-projection-function boundary. Only the four-question core check is required. -
Trigger class 2 — substantive entry, companion, or projection change This class covers a new or changed README scenario, ToC query cue or
E.11entry-distribution locus, including a worked comparison, pattern or applicable projection function newly used for entry. It also covers a changed boundary with a wrong pattern, a governing entry or an applicable projection function, a changed effect on local first-pattern selection, or a substantive change to a lexical-query cue. Run the core check and add the checks needed for unresolved entry risks. Reuse overlapping evidence. A compact pattern-local note is required only when the rationale is needed as described below. -
Trigger class 3 — multi-companion-function or high-risk public entry change A change affects several selected projection or companion functions together, rewrites a public entry, changes a frequently misclassified entry-recognition function, or introduces a set of patterns to compare on first entry. Run the core check and add the risk checks the case warrants, such as parity, wrong-pattern, public-entry or expanded-entry-disambiguation-case adequacy. Stop when the live risks are covered.
-
Trigger class 4 — retrieval-facing, observed-failure, or measured-improvement change one retrieval-facing companion or projection function changes, one observed misretrieval or repeated search failure is being repaired, or the patch itself claims measured discoverability improvement. One selected evidence mode may be required, but benchmark-style reporting is not the default.
-
Trigger class 5 — normative authority, kind, or durable-name change one entry-selection split, stable-name settlement, label-family change, or other normative architectural rewrite is in scope.
DRR,PCP-TERM, andPCP-MODare the lead decision or review profiles as applicable;PCP-ENTRYreviews only the entry-facing effects.
Ordinary changes outside this profile include the following.
-
punctuation, formatting, and typo fixes;
-
meaning-preserving prose tightening;
-
one bare mention of a pattern without changed entry-selection effect;
-
local wording repair that preserves what the reader can correctly recognize at first entry, candidate-pattern set, boundary of the governing entry or applicable projection function, and membership of the first-entry pattern-comparison set.
PCP-ENTRY reviews entry effects alongside any independently applicable PCP-PRAG, PCP-MOD, PCP-TERM, PCP-NORM or other profile.
It asks what changed in pattern selection, first-use recognition, the patterns compared at first entry, the boundary with a tempting wrong pattern, Problem-frame recognition, an expanded entry-disambiguation case, a lexical-query cue, or agreement in meaning among the affected companion or projection functions.
The default review consists of one core check with four questions.
-
No workflow implication Entry text does not imply mandatory sequence, control transfer, handoff, or publication, carrier, or record sequence unless another governing entry or applicable projection function explicitly governs that semantics.
-
Governing-entry boundary preserved Entry, index, and lexical-query companion functions do not redefine the direct pattern body's
ProblemorSolution. -
First honest entry-recognition function preserved The change does not make the first entry-recognition function or case signal misleading.
-
No duplicate high-detail companion or projection function Does the change create a stale echo or a second high-detail companion or projection outside the pattern body or applicable projection function already named for the claim? It must not.
Limit review effort to the concern the change affects.
Learning-order edits do not trigger PCP-ENTRY unless they also change
candidate-pattern set, governing-entry or applicable-projection-function boundary,
first honest entry-recognition function, or first-entry pattern-comparison-set membership.
Lexical-only edits do not trigger extra entry-review scope unless they change
pattern-selection effect or entry recognition.
Retrieval fixtures are not required unless retrieval-facing behavior is
explicitly claimed, one machine-consumed projection is in scope, or one
observed misretrieval is being repaired.
When the risk warrants more than the core check, you may add only the relevant checks.
- one parity check when more than one pattern-entry discoverability-bearing projection changes;
- one wrong-pattern check when misclassification is observed or independently plausible for the intended reader under F.19's grounded-guard test;
- one lexical check when subject-language divergence is substantive;
- one expanded-entry-disambiguation-case check when a worked E.11 entry comparison changes or one high-risk first-entry pattern-comparison set still lacks depth;
- one public-entry check when coarse public entry wording substantively changes entry-selection effect or carries high public-entry risk;
- one retrieval check when the change is retrieval-facing or repairs one observed retrieval failure.
For a substantial discoverability change, keep one compact pattern-local note only when the governing decision needs its rationale. Use the current DRR, PCP result, patch note or other record of that decision.
One stated rationale can suffice when inspecting the governing entry or applicable projection function, partitioning the companion or projection functions, or repairing local wording already controls the risk.
It is not a separate review record unless the change is high-risk, disputed,
public-facing with substantive entry risk, or retrieval-facing.
When needed, the compact note names only the changed companion or projection function, the affected first-entry pattern-comparison set or pattern, the changed first-use recognition function or signature, and the entry or applicable projection function governing the claim or projection. Name the selected check if there is one.
Empirical evidence is required only when the change is:
- high-risk;
- disputed;
- retrieval-facing;
- repeatedly misclassified;
- public-facing with substantive entry-selection change, repeated failure, or one measured-improvement claim;
- or itself claims measured discoverability improvement.
Select PCP-ENTRY-E4 only for a stated retrieval claim, a machine-consumed projection in scope, or an observed misretrieval being repaired.
Public-facing changes with substantive entry-selection risk usually select PCP-ENTRY-E1.
Lexical-hook changes usually select PCP-ENTRY-E3.
Changes across multiple projections or companion functions usually select PCP-ENTRY-E5.
Observed search or query failures usually select PCP-ENTRY-E6, optionally
together with PCP-ENTRY-E3 or PCP-ENTRY-E4 when the failure is lexical or
retrieval-facing.
Select only evidence modes needed for the changed entry risk. An unselected mode requires no result row or durable disposition. Selected evidence modes may include:
-
PCP-ENTRY-E1 — cold-reader recognition or pattern-selection task Given a real case signal, what entry, set of candidate patterns or justified stopping point can the reader identify? One short task can establish the entry, admissible set of candidates or justified stop this reader selected from this signal with the material provided. If the prompt supplies the intended entry, the response shows what the reader can do with that cue. For a claim of improved discovery, compare what readers could find with the old and new entries on comparable tasks under conditions suited to the claim. Identify the readers and signals covered and account for prior exposure when it can affect the choice. Ask about the alternative in item 2 only when an observed choice or independent local cues make it plausible for the intended reader and the distinction changes selection or use. Otherwise omit item 2.
-
PCP-ENTRY-E2 — wrong-pattern and wrong-entry trap For an observed or independently plausible misclassification, can the reader distinguish the intended pattern, entry, or family from that alternative? Use cues from the problem and its subject. State the rejected alternative only when F.19’s grounded-guard test warrants it.
-
PCP-ENTRY-E3 — lexical query check Does subject-domain phrasing retrieve the governing entry or applicable projection function without uncontrolled aliases?
-
PCP-ENTRY-E4 — retrieval or
RAGfixture Does retrieval recover the governing entry or applicable projection function under exact-ID or keyword phrasing, under semantic paraphrase phrasing, and under projection-vs-governing-entry ambiguity, while keeping retrieved companion material, source faithfulness, stale echoes, and post-rationalized citation-like material distinct from the applicable direct pattern body? Retrieval returns the governing entry or intended projection cue before one stale echo, and answer-to-governing-entry faithfulness remains intact. When thin echoes are used, check that they carry a governing-entry reference. -
PCP-ENTRY-E5 — companion-or-projection function parity check Check that the governing entry or applicable projection function is unique. The changed companion or projection functions must agree on first-use recognition, the wrong-pattern boundary and projection-only status. None may introduce a claim beyond the use admitted in the Core pattern. Wording and examples may differ. Include any stated absence in the comparison; rows may differ too.
-
PCP-ENTRY-E6 — observed failure or query-log capture Under the conditions in which the misretrieval, wrong-pattern loop or repeated query miss was observed, does the repair remove the failure?
Reference cases for regression checks and worked examples
Select a case that exercises the changed entry risk. Cases 1–4 specialize E.11:5.10, E.11:5.8, E.11:5.12, and E.11:5.9 respectively. E.11 governs the entry-distribution use, and the direct subject patterns govern the recovered claims. Cases 5–6 add search and retrieval stress under PCP-ENTRY-E1 and PCP-ENTRY-E4. Another relevant E.11 case may be used. Select empirical evidence under PCP-ENTRY. Unselected cases need no run or absence note.
The tempting_wrong_pattern_or_wrong_relation column is conditional on an observed or independently plausible reader mistake that changes selection or use. Leave it unused when that condition is absent; keep the case's positive recognition and admissible stop.
In case 3, evaluate episteme identity separately under C.2.1: its discriminators are claim content, EntityOfConcern, and effective reference scheme. Changed wording or audience alone leaves that identity unchanged; a changed discriminator identifies another episteme. A.6.3.CR permits claim-preserving or explicitly loss-declared same-EntityOfConcern re-expression under its own use conditions.
Selected cases test:
- entry-recognition consistency;
- wrong-pattern or wrong-entry rejection;
- admissible entry-stop honesty;
- lexical-query discipline;
- thin-echo retrieval hygiene;
- and governing-entry and projection separation in the changed entry text.
When one empirical or retrieval evidence run is selected, keep recoverable the facts needed to understand its question and result. A structured result may use the following field names, including only those needed by that run:
For a retrieval check, distinguish what was retrieved, whether the answer is faithful to its source, and whether a stale echo was returned. Ordinary edits require no benchmark report.
Use the retrieval questions in PCP-ENTRY-E4, including its conditional
thin-echo reference check.
Ordinary local guidance stays in prose. It does not require a stable governing-entry reference by default.
Additional questions for the selected review concern
Open a common hardening question when the concern has FPF-governed use, is disputed, or the reviewed pattern or subset expressly calls for it. Inspect the relevant source and the reviewed loci. In inspect-repair-verify, repair any defect and verify the affected use. In independent findings, record the defect and repair direction. When the question reveals no defect, make no durable absence or pass recital.
Use these questions only for the selected review concern:
-
Usability and working-reader fit. Open this when first-reading recognition text, assurance text, first-minute working-reader usability, practical payoff, worked slices, primary-reader fit, or
E.8/E.12/E.13/E.14/E.17.*/F.16checks can change the admission or refresh result. If a separate evaluation assigns a value, use that evaluation's result rather than copying it into E.19 findings. -
Scenario, anti-case, and utility-fit source set. Open this when a scenario pack, anti-case corpus, pilot bank, utility tree, fitness catalog, or analogous source is relevant to the case or substantively disputed. Record only a missing, misused, or failing source/case as an E.19 finding.
-
Packaging, concrete pattern contribution, package relation, and shipping fit. Open this for a publication, pattern-contribution, or package-relation claim. The changed sources and governing publication or release result carry successful alignment; E.19 repairs or reports a mismatch.
-
Domain-tightened profile depth. Open this when a domain-specific note tightens a selected profile. Apply its questions; do not add a second account of positive results.
-
Accepted-decision or accepted-source-material carry-through. Open this when the reviewed pattern, subset, or current change is claimed to implement an accepted
DRR, repair findings, intake material, architecture source material, or other accepted source material named by value. Inspect each independently applicable decision against the reviewed loci and the concrete pattern, claim, companion, result, or accepted source that carries it; require exact predicate or definingClaimGraphidentity only when that decision or the named reliance needs it. Repair or report partial, missing, wrongly rejected, wrongly routed, or wrongly classified carry-through. The accepted source remains the decision source; E.19 does not duplicate decisions that are expressed sufficiently, inherited unchanged, correctly absent, or outside the reviewed subset. AnE.17.ID.CRcomparative review unit,PublicationUnit, publication form or face, source-pinned interpretation case, source material, or project-side review relation retains its own kind in that comparison. -
Decision-useful advice and evidence demands. Open this when a pattern's advice, feasibility or evidential burden is in question and can change the review result. Use
C.11.DUAto recover the demanded work, compare its attainable contribution and burden, and examine a disputed requirement's merits and force. Follow the selected method through its result, prescribed fields and stopping point: retain the useful basis and any explanation the decision or recipient needs, while inactive inquiry creates no omission account. A worthwhile evidence demand remains. ApplyF.19to wording that needs repair; this question adds no universal profile or second language pass.
For PCP-ENTRY, the ordinary compact pattern-local change note remains enough when the governed discoverability decision requires one; no separate E.19 account is required for checking the profile.
Pattern-Edition Use-Value Replay
Use this replay when an exact candidate pattern edition changes materially under E.8:4.1.2. Run it once on the stable candidate before acceptance or landing. Start with the bounded E.8 loop over the actual predecessor and proposed prose, then open only each affected prior-edition or candidate-only use whose result can differ, pinned to its exact basis and changed locus.
Treat a change as mechanical only when the smallest relevant comparison shows that every materiality value named in E.8:4.1.2 is preserved. A genuinely bounded local semantic edit opens only its affected use probe and changed wording group; the physical size of the rewrite does not establish its scope.
When the candidate keeps, merges, removes, profiles, reuses, externally supplies, or omits a narrower contribution, apply the same-situation decision in E.8:4.1.3. If reuse or a gap answers the working question, establish what is returned: a result of its own kind from the supplying product, a MethodDescription reference, direct-source evidence, or a named unavailable result. For an external result, verify the exact result and supplying product, receiving use, practical discovery route, material currentness or availability, and the statement that it remains outside the receiving framework; state maintenance only when it changes that use. Otherwise the package still has a gap or omission.
When the resulting stable set materially changes a promised problem family, verify the current E.4.DPF.DA D12 judgement for the resulting exact edition required by E.8:4.1.3. Reuse a matching current result when the exact edition, promised families, declared use, relied-on results, and relevant conditions did not change; E.19 asks for neither a duplicate package evaluation nor evidence that a revisit occurred.
Judge each affected use probe separately when its result can differ by exact predecessor or candidate-only basis, working use or relying work, expected first useful result, boundary, necessity, or evidence mode. One review may contain probes from both bases.
A grouped verdict such as uses preserved or added or usability preserved cannot substitute for those judgements. Inspect and repair a failed probe in repair mode, then verify it. In independent review, record only regressions, insufficiencies, invalid transfers, unsupported decisions and blockers. Neither form requires a per-probe progress record. When E.8, E.21, or another governing evaluation requires reusable dispositions or values, keep them in that evaluation's result rather than copying them into E.19 findings.
Changed-wording check inside each affected prior-edition probe. Keep the selected use probe as the outer unit. When a predecessor-bearing candidate materially rewrites a normative sentence or inseparable sentence group that carries the governed extension, action discriminator, first useful result, stop, or neighboring-pattern exit, judge that wording group under the dispositions below before closing the use probe. Keep sentences together only when they serve one reader task and must receive one disposition; split them when their extension, action, result, or route can differ.
For each changed wording group:
- pin the old and candidate wording and the exact use it serves;
- state in plain language the subject, concrete action or choice, visible result, and any stop or exit the use needs;
- compare the old and candidate head and modifiers, modal force, admitted referents or actions, applicability boundary under its governing rule, and local interpretation burden. Check every widening or narrowing against that rule and the accepted change decision, independently of the reader-plausibility question in the next step;
- if independent local evidence makes one rival reading plausible to the intended reader and its treatment can differ, test that case. Use the plausible-reader test to decide whether an explicit guard contributes to the final wording; the extension comparison remains required. Do not invent an alien or excluded case to fill the review form; and
- apply the differential disposition.
preservedrequires no unauthorized widening or narrowing and no greater decoding burden: a reader must not need campaign memory or an ontology-development memorandum to recover the action.
For a new action-guiding paragraph with no predecessor, do not invent history or a foil. Verify that the local wording exposes a recognizable situation, concrete action or choice, visible first result, and any independently grounded applicability or neighboring-pattern boundary needed for use.
Formatting, typo, link, citation, or exact-reference corrections remain mechanical when the smallest comparison proves that no E.8:4.1.2 materiality value changed. A bounded semantic edit checks only its affected wording group and use probe. Reuse an earlier hunk or language result only when the object and compared editions, changed scope, and assurance question match this extension, modal-force, applicability, and interpretation-burden test; idea presence or broad-use preservation is not enough. Complete the review in the same increment on the stable candidate before acceptance or landing. It requires neither per-edit review nor a new ledger, and the first finding does not trigger a handoff.
Prior-edition differential. For one candidate pattern edition × one prior-edition use probe, distinguish the applicable disposition when the governing decision needs it:
A use classified as unsupported historical residue before replay receives no differential disposition and supports no compatibility claim. New evidence of a valid old use reopens that classification instead of restoring wording silently.
A required regressed probe prevents a positive conclusion, but it does not stop inspection of the remaining independent probes.
Candidate-only adequacy. Review one candidate pattern edition × one new intended-use probe against its exact candidate-only basis, never against invented history. Distinguish these outcomes when the governing decision needs them:
A missing candidate-only decision or basis is absent or insufficient; it never licenses a fabricated prior edition. Absence for a required new use prevents a positive conclusion but does not stop the other independent probes. Missing optional breadth alone does not block a positive conclusion. It provides no grounds for breadth, transfer or exceptional-expression claims. If no exact new intended use is selected, no candidate-only check opens.
Replay the positive Solution separately. Judge the following over the candidate edition when their answers can differ:
- the governed subject;
- the recurring problem and ordinary failure;
- an executable proposed move;
- a first useful result rather than completed review apparatus.
When the Solution uses a boundary or guard. Judge each guard for which independent local evidence makes the exact rival reading plausible to the intended reader and whose presence changes action. Check whether any remaining guard merely supplies the outline that the positive Solution should state directly. Refine this question by boundary whenever boundaries can pass, fail, or route independently. This guard check leaves the rule-defined extension comparison above intact.
When the Solution concerns a Method, work, or world-side change. First distinguish content that defines, constrains, tests, or guides from an assertion of one actual occurrence. Method guidance alone triggers no fictive performer or Work. When an account asserts dated U.Work, verify the §4 actual-Work account; when it asserts a world-side change, identify the change relation, the pattern that defines it, and the things it relates. Keep the intended-reader position, any qualifying A.3.2 method-description episteme, actual performer, Work, and problem-facing result separate. A literal dated-Work claim is defective if an episteme, checklist, plan, prose, or intended-reader or representation position performs Work. Judge ordinary metonymy through F.19.
Follow the short first-use rendering's action and result logic against a concrete situation. Finding labels such as situation, move, result, or stop does not show that a reader can use the instruction. Repair each failed item or record it as a finding that identifies the failure and how to repair it; do not replace the replay with one prose-quality impression.
Replay each triggered enumeration or coordination under F.19. Apply its contribution, coordination/list, and foregrounding rules in F.19:4, including the return to a defining or testing pattern when an FPF kind, relation, or structure remains hidden. Review a member separately only when its membership or contribution can fail independently or require a different repair. An unchanged series covered by the same rule needs no positive entry. A blanket all lists are coherent conclusion cannot replace the reading.
Desk replay is the ordinary way to check whether a changed passage still warrants an earlier use or gives the reader a new, usable choice. Examine the assumptions and the reasoning through the consequential branches. If the uncertainty concerns the words, ask a fresh reader to recover the choice or boundary from the passage with the material available in ordinary use. If it concerns what practitioners do, examine their work under the claimed conditions. A competent reader may also supply a valid proof or counterexample. Check that argument on its own grounds.
For a claim that a rule works throughout a stated domain, use a derivation, an exhaustive check of a finite domain, or a division into cases with grounds that they cover the domain and that the argument holds within each. Numerical agreement or a successful reading on one case cannot, by itself, establish the general claim. For a proposed application, a valid argument and established conditions at the site can warrant using the rule. To report what actually happened, select observations of the work. To attribute an improvement to the rule, select a comparison suited to the practitioners and conditions claimed. A practitioner survey, case study or action-research cycle can help where those observations answer the question. When the claim concerns the organization of a whole, establish the relevant relations between its parts. For another level of model abstraction, establish which assumptions and answers carry across. The same argument can cover several levels when its conditions hold at each. Select further work according to the consequence of being wrong. E.21 owns ordinal values and PatternQualityStatus for a full evaluation.
Decision outcomes
Complete the selected review scope before making an admission, refresh, or return-for-repair conclusion. A first defect or already-negative conclusion does not end the search for other independently obtainable findings. If a condition makes the remaining questions impossible to judge truthfully or safely, report a partial result and identify the unexamined scope and the condition that prevents its review. Before using a reader's result for a wider finding, compare the task as given to that reader with the work the pattern promises. A reviewer who supplied a choice of parts can use the result to judge how well the reader applied that choice. If choosing the parts and establishing their joint suitability are also promised, recover the grounds for those judgements separately.
Inspect, repair, and verify. Complete every in-scope review application, repair every defect through the relevant authoring work, and perform focused verification over the affected questions. The changed pattern edition and focused-verification claims are the substantive evidence and remain distinct from the work that produced them. Constitute one aggregate E.19 result episteme only when a receiving admission/refresh decision needs it. Record only an unresolved blocker, a decision outside current authority or work that must transfer. Completed repairs need no parallel list.
Independent findings. Leave one compact C.2.1 findings-result episteme or handoff file containing all actionable in-scope defects and blockers, ordered by semantic impact, with repair direction precise enough that the author need not rediscover the diagnosis. If the selected questions reveal no defect, create neither an empty pass report nor positive checklist recital. The findings result is not the dated review work or an authority-bearing admission decision.
If a governing admission, refresh, E.21, DRR, publication, or release decision requires a durable conclusion or value, use its existing result. That result may cite E.19 findings or the repaired candidate; it does not turn per-question positive outcomes into a second review record.
Precision-remediation order. When a defect sentence combines a generic head, a claim-bearing qualifier, and pressure to compare values on mixed criteria, remediation SHOULD follow the precision-before-coarsening rule in F.19:4. That rule supplies the head/qualifier/comparison order and the recoverable precise interpretation required for a later Plain, didactic, or coarsened restatement.
Kind-restoration verification. A wording, naming, or F.19 phrase-level repair does not succeed merely because the old trigger word disappeared. Compare both versions for object kind, relation or claim kind, admissible use and scope. If the repair narrows, widens, splits, or changes them without an accepted decision, repair it or keep the defect unresolved. The repaired object, focused verification, or governing decision carries this evidence; E.19 does not require a per-repair pass account.
Ordering and effort. Put ontology, semantics, modularity, and SoTA defects in FPF-governed sections before compact low-signal formatting findings. If semantic defects are present, address them before mechanical edits; formatting and micro-typos must not dominate the work by volume.
Archetypal Grounding — perform the selected review
The situations in section 0.2 help a reviewer recognize which questions to select. The following constructed reviews show the comparison, conclusion and return for two different kinds of defect.
Review and repair a source-change instruction
E.21:5.0 supplies the complete compact pattern EX.1@source-pin-1. It asks the user to record a source edition, the rule used and its effect on a decision, then concludes: "Reopen the decision whenever the source publishes any new edition." The present E.19 question is a bounded refresh review of that last instruction, read with the whole quoted pattern. The relying use is to keep a decision connected to its source grounds.
The reviewer selects PCP-BASE and PCP-REFRESH. The edition-change instruction can repeatedly create work and can determine which decisions receive attention. The reviewer therefore examines the trigger condition. The small scope permits inspect-repair-verify within the reviewer's authority.
The reviewer compares the trigger with the reliance described earlier in the pattern. In one constructed new edition the title changes while the used rule, applicability and limitations remain the same. The instruction reopens a decision whose stated ground is unchanged. In another edition the relied-on maximum inspection interval becomes shorter, affecting the chosen interval. Here reconsideration is needed. The difference identifies the defect: publication of an edition is being used in place of a material change to the decision's basis.
The reviewer replaces the last instruction with:
When a source changes, compare the used rule, its applicability and its stated limitations. Reopen the relying decision where that comparison changes its grounds. Retain decisions whose grounds remain applicable. If the used rule or edition cannot be recovered, apply the existing stop-and-retrieve instruction before relying on it.
Focused verification applies the two changes to the revised instruction. The title change now calls for preserving the recovery route while retaining the decision. The shortened interval sends the affected decision back for reconsideration. The source-identity and stop-and-retrieve contributions remain intact. The review can close this bounded refresh question with the repaired instruction and that comparison. E.21's evaluation of the deliberately defective quoted edition remains an evaluation of its original text.
Keep a pattern-quality conclusion within its authority
A different candidate adds this claim to its otherwise developed review guidance: "An admitted deployment pattern authorizes the team to deploy the system." The requested review concerns this new authority claim; the preceding account already explains the Method and its conditions.
PCP-BASE identifies a change from the quality of guidance to permission for a particular operation. PCP-NORM is selected because the new sentence asserts authority. The reviewer locates no project decision or authorization relation in the candidate's grounds: the supplied conclusion concerns the pattern edition. The claim therefore lacks the contribution that would permit the asserted deployment.
Under independent findings, the reviewer leaves the candidate unchanged and returns the exact repair: state that the admitted pattern supplies guidance for the deployment work, and identify the project's governing authorization before claiming permission to perform that work. The author replaces the sentence with: "Use the admitted pattern as guidance for preparing deployment. The project's responsible authority decides whether this deployment may proceed under its governing conditions." Focused rereading preserves the useful Method guidance and separates the permission decision and its needed basis.
This defect requires a different comparison from section 5.1. There the relation from source change to decision grounds selected the refresh; here the subject and authority of the returned conclusion select its permissible use. Both reviews choose their depth from an actual claim, carry out the selected comparison and complete a repair or findings return. Their reasoning is available for analogous questions under the applicable profiles; observation of transfer by other reviewers requires those actual uses.
Bias-Annotation
These risks can arise when reviewing any FPF pattern or cluster.
- Governance bias (Gov): reviewers may over-prioritize compliance signals and under-prioritize teaching value. Mitigation: PCP‑BASE checks didactic grounding and internal coherence and prioritizes ontology and semantics.
- Architecture bias (Arch): internal package architecture can displace the problem-owning domain or practice.
Mitigation: test EntityOfConcern, narrowed branch, and practical payoff against the domain/practice question and relevant SoTA under
CC-E19-7. - Epistemic monoculture (Onto/Epist): SoTA‑Echoing can become single-tradition name-dropping. Mitigation: use multiple traditions when the question or claimed breadth requires them; make substantive disagreement visible. Use F.18 for neutral durable naming when its use condition applies.
- Pragmatic bias (Prag): a pattern can be “correct” yet unusable.
Mitigation: consequences and anti-patterns remain mandatory sections, surfacing material costs or limitations and grounded misuse or application boundaries under
E.8; an already established boundary may be referenced. - Didactic bias (Did): narrative quality can be mistaken for truth. Mitigation: conformance and SoTA‑Echoing sections bind claims to explicit requirements and lineage.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
Patterns are both teaching publications and normative guidance publications. A specification that grows without explicit quality gates can become a patchwork: locally good, globally inconsistent. A profile-based gate combines a short common baseline with depth selected for the live risk and pattern kind.
The baseline profile protects cross-pattern comparability and editorial sanity. Risk-selected profiles keep depth where it matters: norms, SoTA claims, cross-context reuse, terminology changes, staleness refresh, and reader fit. A pattern that is admissible in package terms but speaks to the wrong reader is still a review defect.
SoTA-Echoing — problem-first comparison of review approaches
Working trade-off. For one FPF pattern edition, detect semantic, ontological, practitioner-use, and source-currentness defects without making an ordinary review cost more than its likely harm warrants. Design inference from the comparison. Combine narrowly targeted tools, human review of contextual defects, subject reasoning for the range of a rule, practitioner evidence for transfer in use, and living-guidance methods for volatile claims. The table gives the contribution and limits of each source.
Evidence binding. If a current SoTA Synthesis Pack answers this exact review or refresh trade-off, cite it and keep this section consistent with it. Otherwise, use the source contributions below for the named review question and decision; source identity alone does not establish the quality of that decision.
Selected current front. Ordinary E.19 use combines a lightweight stable-candidate review with cheap bounded automation, then scales review depth by likely harm, novelty, reuse breadth, and source volatility. It retains independent findings as a review form, tests breadth with practitioner evidence only when the claim warrants it, and refreshes the smallest triggered unit. The selected trade-off avoids running questions for absent risks while retaining human judgement of semantics, practical use, and current-source decisions. Practitioner studies and continued surveillance add effort only when the claimed breadth or currentness warrants it.
Reopen this choice when another approach demonstrates equal or better detection of those four defect families at lower comparable effort.
When a receiving use requires a reusable E.19 result, it states the review claim for one FPF pattern edition or subset. Its EntityOfConcern, ClaimGraph, scope, applicable profiles, findings or aggregate cleared boundary, conclusion, and reopen condition state that pattern-quality claim.
A project-side claim beyond the stated pattern-quality result needs its own grounds and any evidence, assurance or decision required by the relevant project rule. Review, repair, verification, result publication, admission or refresh decision, and project-side reuse remain separately recoverable.
Reopen the affected result claim when a change to the reviewed text, accepted source, SoTA grounding, related pattern contribution, selected companion or projection function, profile trigger, review boundary, or claimed downstream use can change its conclusion or applicability.
Relations
-
Builds on:
E.8(authoring conventions; canonical section order; SoTA-Echoing authoring requirements)F.19(common whole-span precise-language repair, plausible-intended-reader guard test, coordination and list contribution, and local semantic reread after wording change)E.10(compact lexical cues and exact FPF routing)E.10.ARCH(deep-only ontological recovery architecture afterF.19andE.10leave an unresolved FPF object or relation)E.9(design rationale records for changes that affect semantics)E.9.DA(content-first adequacy check for one exact DRR before pattern drafting or host amendment. An ordinary bounded check returns precise findings or repaired text; a full coordinate result and exact assessment identities are added only when explicitly requested or used by a named later reliance. An E.19 finding may expose an upstream DRR defect, but an E.19 pass, return, or absence is not E.9.DA evidence.)E.22(improvement-oriented quality-evaluation question framing; distinguishes floor blocker review, exceptional-improvement review, Pareto trade-off inspection, open-question discovery, and absorption impact before an E.19 review result is formed.)E.23(repeated quality-improvement method; an E.19 profile can supply questions and findings inside such a loop, but E.23 governs repeated absorption, object-under-improvement re-evaluation, method-family selection, and stop, continue, switch-method, open-new-frame, or hold decisions.)E.15(change between exact pattern editions; actual-delta classification, affected-reach repair, predecessor preservation, and proportionate verification)A.6.5(slot discipline; SlotKind/ValueKind/refMode invariants)A.6.P(direct relation and participant recovery when the claim remains unresolved)
-
Coordinates with:
A.13andA.15.1(exact actual-performer recovery and independent review, repair, or verification Work);A.2,A.2.1, andF.6only when local classification or precise assignment-bound attribution is expressly consumed;A.6.1separately defines check applications and bindingsA.3.2andE.10.ROLE(the fullU.MethodDescriptionmembership test and recovery of an ambiguous source role without forcing a system-role kind, assignment, participant, or representation position)C.2.1(finding, focused-verification, aggregate review-result, and optional record epistemes)A.10for an implicit or required-replayable source-to-use account;B.3for a selected assurance claim about an E.19 resultF.10andE.24.PUB(status use/interpretation and publication occurrence/form/carrier; neither is review work or admission authority)F.8(mint vs reuse decisions)F.18(local-first naming when one durable reusable name is needed)F.9(cross-context alignment discipline)F.15(conceptual harness and regression framing)E.17(MVPK publication and face discipline; an MVPK face, projected publication form, projection/construction, publication occurrence, rendering, and carrier remain distinct)E.17.0(independent conformance required before the selected episteme hasU.Viewmembership; E.19 profile checks and no-new-claim/no-shadow-default compliance create no membership)E.11(pattern-entry discoverability discipline, forPCP-ENTRYonly as a review hook, not as a semantic prerequisite)E.13(pragmatic utility and proxy-to-value alignment when a pattern-quality pass, score, coordinate value, checklist result, benchmark, projection signal, or release posture is being used as value evidence)E.21(scoped pattern-quality characteristic space, coordinate evidence discipline,PatternQualityStatus, and stop condition; E.19 findings may become evidence only through the exact E.21 assessment application. Final coordinate values andPatternQualityStatusbelong to a separate E.21 result episteme, not the E.19 profile or result.)A.6.7(MechSuiteDescriptionsuite-level semantics)E.20(mechanism-introduction and governing-definition changes when its trigger applies)A.15.3(SlotFillingsPlanItemP2W planned-baseline seam)G.11(refresh/decay orchestration principles, where applicable)
E.19:End
Mechanism Introduction Protocol: Introduce or Revise FPF Mechanisms
Type: Architectural pattern Status: Stable Normativity: Normative
Problem frame
FPF is open-ended by design. Authors can add new U.Mechanism definitions, suite compositions and wiring modules informed by the best-known practice for the problem. Introducing or revising a mechanism often affects several subject patterns, specification sections and extension blocks across Parts A/E/F/G. They can drift apart in several ways:
- semantics appear in the wrong governing locus (e.g., Part G wiring starts carrying mechanism meaning),
- suites degrade into “meta‑mechanisms” or hidden gates,
- planned baselines in exact
U.WorkPlancontent are conflated with dated performed Work, - token drift breaks public references, or
- the corpus accumulates dangling references and non-normative drafting commitments without a governing definition.
This protocol tells authors how to assign each change to its governing definition. Making those definitions and the extension points visible keeps repeated introductions consistent with the kernel.
Use this when. Use E.20 when a proposed FPF change introduces or revises mechanism meaning, suite denotation, suite closure, suite obligations, suite pins, suite protocol semantics, planned-baseline pins, wiring semantics, governing-definition assignment, or what a citeable token denotes.
First useful move. Classify the edit with MIP trigger triage: MIP not triggered, local wording or alias-docking only, or MIP-run manifest required. If a manifest is required, name a single governing definition for each changed item before writing the pattern text.
When the change also involves a transformation flow. Use E.18 for transformation-flow structure, paths, valuations and crossings, A.20 for internal step validity, A.21 for a gate profile application and decision, and E.20 to assign mechanism meaning to its definition. Open each of the other three only when its own claim is present.
After a wording or publication correction. If E.20 rules out a misleading word, face, alias or source label, name the governing definition, canonical location, alias-docking relation or non-trigger stop that remains available to the author. Classifying the vocabulary or publication face alone leaves the authoring move unspecified.
Subject and relation separation. Keep the selected transformation-flow structure, paths and crossings (E.18), its mathematical description (E.18.2), MVPK publication faces (E.17), internal CV status and witness (A.20), gate profile application and decision (A.21), evidence or provenance relation (A.10/G.6), work plan or work occurrence (A.15), and mechanism-definition assignment (E.20) distinct.
An MVPK face, DecisionLog, evidence value, provenance reference, MIP manifest, or work witness does not supply another subject's project-side value unless an exact dependent-use assertion and its defining or constraining ClaimGraph establish that relation.
Smallest affected locus. Start with the smallest affected PathSlice or crossing in E.18, CV step in A.20, profile application and bounded action decided in A.21, or governing mechanism definition in E.20. When that locus is sufficient, keep the change there, without widening it to the whole flow or unrelated claims, loci or EntitiesOfConcern.
Do not merge these pairs. Keep CV.Status distinct from GateDecision, E.18 Check locus distinct from GateCheckKind, MIP manifest distinct from DecisionLog, ViewpointMap distinct from graph semantics, PathSlice distinct from a work run, and GateProfile=Lite distinct from PublishMode=Lite.
What the manifest establishes. A complete MIP-run manifest is not a substitute for the governed mechanism result. The cited U.Mechanism episteme must recover <content, EntityOfConcernRef, effectiveReferenceScheme> and the A.6.1 content needed by the receiving use. Realization, refinement, bridge, evaluation, evidence-use, and publication relations remain neighboring claims under the patterns that govern those relations.
What goes wrong if missed. A suite can start defining mechanism meaning, declaration-local WorkPlan rows can start carrying enactment witnesses or gate decisions, a wiring module can carry kernel semantics, or a token rename can break citations while looking like harmless cleanup.
What this buys. Assign the change to the right governing definition. This keeps mechanism, suite, planning, wiring and lexical continuity distinct.
Not this pattern when. For a currentness, typo, reference or old-label cleanup, check whether it changes any semantics or citeable-token denotation. If it changes neither, record the current governing locus and stop. If the question concerns runtime gate passage, gate decision, approval, suite-as-mechanism, plan-as-enactment, or performed work, use the applicable gate, suite, planning, or Work pattern for that question. A MIP-run manifest is not a runtime gate, gate passage, approval packet, or binary pass/fail decision.
Problem
Without an explicit authoring protocol, introducing a mechanism or mechanism family can cause five problems:
- Governing-definition ambiguity causes partial changes: a suite enumerates a new
MechanismDefinitionRef, but that designator has no resolvable A.6.1U.Mechanismepisteme or resolves only to a card-shaped placeholder without mechanism identity and content. - Boundary erosion occurs: suite descriptions start to define mechanism semantics; method wiring starts to redefine kernel meaning; publication/telemetry becomes a hidden tail.
- Plan/enactment confusion appears: planned slot fillings start to carry launch values, witnesses, or gate decisions.
- Terminology drift breaks citations: renames happen silently; tokens fragment across registers; downstream references become unstable.
- Review becomes non‑local: reviewers must piece together each introduction across several patterns, making training, review, and refresh unreliable.
Forces
Solution — the Mechanism Introduction Protocol (MIP)
Terminology note (disambiguation)
MIP maps authoring changes to their governing definitions. Its manifest records the assignment and, when a correction changes use, what must not be inferred from suites, plans, wiring, aliases or gates. A manifest is not an approval packet, gate, runtime decision or pass/fail result.
MIP governs how changes are assigned to their governing definitions, not how systems execute.
MIP trigger triage. Classify the proposed edit before requiring a manifest:
- MIP not triggered: currentness, reference, typo or old-label cleanup that changes no mechanism, suite, planned-baseline, wiring, governing-definition, or citeable-token semantics.
- Local wording or alias-docking only: wording clarifies an already-governed mechanism relation, or
F.18alias docking preserves citeability of an old token without changing what the token denotes. - MIP-run manifest required: the edit changes mechanism meaning, suite denotation, suite closure, suite obligations, suite pins, suite protocol semantics, planned-baseline pins, wiring semantics, governing-definition assignment, or what a citeable token denotes.
Only the third outcome uses the manifest in E.20:4.2. For the first two, name the current governing locus or alias-docking relation when the text will be published. When the only result is no denotation change, the published content should not carry MIP-run vocabulary except as a short non-trigger note.
Mint vs reuse
Mints:
- MIP — Mechanism Introduction Protocol (this pattern).
- MIP-run — an authoring event that applies this protocol to a change set, captured as a short manifest (recorded as a DRR-linked change record or an equivalent citeable change record).
Reuses:
- A.6.1
U.Mechanismepistemes, theirMechanismDefinitionRefdesignators, non-mechanism reference-reservation stubs, suite descriptions (MechSuiteDescriptionand specializations), exact A.15.2U.WorkPlanepistemes and their declaration-local A.15.3 planned-filling rows, alias docking (F.18), RSCR triggers (G.Core), and PQG profiles (E.19).
Step 1: Classify the introduction
First establish what a user of the declaration or suite must be able to obtain after the proposed change. Trace that result through the operations and handovers it needs. Which current rule excludes a required use, admits a forbidden one, or leaves a decision to be made? A new binding can enable a use already covered by the operation laws. A promise about several operations together may require changing the suite protocol while all member declarations remain valid. If the user requested only a correction to an existing reference, its resolution can settle the matter.
Classify each change by its governing definition:
- New declared operation family or archetypal grounding. The
EntityOfConcernRefnames an operation family not previously declared at the selected governing locus. - New mechanism declaration or semantic edition. One A.6.1
U.Mechanismepisteme receives new identity-bearing content or a new effective reference scheme. - Neighboring mechanism-relation change. A realization, refinement, conservative extension, equivalence, bridge, evaluation, evidence-use, or publication relation changes while the mechanism content does not.
- Suite change (membership, obligations, spec pins, or suite protocols).
- Planned-baseline change (new or revised declaration-local planned-filling rows inside one
U.WorkPlan, or changes to their pins). - Wiring change (new or revised Part-G extension modules, SoTA method packs, or selectors).
- Terminology migration (renames, token splits or merges, or register changes).
- Deprecation, supersession, or retirement (status change, successor relation, and preserved citeability; apply E.20:4.9.1).
Mechanism-kind boundary. MechanismDefinitionRef is a designator. Minting it neither creates a U.Mechanism episteme nor admits a new U-kind. A new U-kind claim requires E.24.UK; a new mechanism episteme must satisfy A.6.1 identity and content; a new transformation-flow structure requires E.18.
A.6.1 compatibility. Mechanism identity is <content, EntityOfConcernRef, effectiveReferenceScheme>. Identity-bearing content comprises direct subject and range fields, OperationAlgebra, LawSet, AdmissibilityConditions, Applicability, and an optional SignatureManifest when dependency replay matters.
An operation index may be derived from the declaration-local operationDesignator values; it is not another content group. Each operation's arguments and results remain A.6.1 ArgumentDeclaration and ResultDeclaration content. A.6.5 SlotSpecs remain exclusive to a RelationSignature for an already governed direct relation. Realization, refinement, extension, bridge, evaluation, evidence-use, and publication relations are governed separately.
New-declaration criterion. Treat a change as a new declared operation family when EntityOfConcernRef changes. Treat changed mechanism content or effective reference scheme as a new semantic edition. A changed neighboring relation alone does not create a new mechanism identity. It may reopen reliance on the current declaration.
One MIP-run MAY span several classes. The author SHALL assign each changed item to its governing definition below.
Step 2: Declare the governing-definition assignment map (mandatory)
For every new or modified change item, the author SHALL name exactly one governing definition and make the change there. In FPF, that governing definition is a citeable, patchable PatternId, PatternId:SectionPath, PatternScopeId = G.x:Ext.*, or DRRId (E.9). The core MIP-run manifest in a citeable change record is limited to:
- each changed item,
- its governing definition,
- its canonical location (expressed as
PatternId:SectionPath,PatternScopeId, orDRRId, not as prose).
Conditional manifest fields appear only when the corresponding claim is present:
- a forbidden reading or move only when its independently grounded correction changes use under F.19,
- the change class(es) from E.20:4.1 when needed to disambiguate the assignment,
- new or changed citeable tokens, including a
MechanismDefinitionRefor a public operation, argument, or result designator, when token denotation or citeability changes, - the actual-effect Delta-Class (
Δ-0toΔ-3) and affected-reach estimate from E.15 when the run is plausiblyΔ-2orΔ-3, - intended RSCR trigger types when a refresh or regression-wiring claim is present, and
- the PQG (E.19) profile set when the run crosses an E.19-governed review boundary.
Part-G wiring. If the canonical location is a Part‑G wiring module, it SHALL be cited as a PatternScopeId (G.x:Ext.*) and the module SHALL declare GoverningPatternId. The module supplies only bindings. Its cited pattern governs meaning.
Canonical governing-definition map (normative):
When a change has no named governing definition. Any proposed change that cannot name a governing definition from the table above SHALL be treated as a non-normative drafting note or candidate intake and SHALL NOT be relied upon as an FPF architectural commitment. Such material may exist only in a source note marked non-normative until assigned to its governing definition.
Step 3: Resolve the designator before dependent use
When a change introduces a MechanismDefinitionRef, create one resolvable target at the subject-pattern locus before another declaration cites it. Distinguish two target states:
- Reference-reservation stub. This is a draft authoring episteme, not
U.Mechanism. It reserves the designator, names the intended operation-family EntityOfConcern, cites the subject pattern, and lists the missing A.6.1 identity or content needed for introduction. A publication may expose the stub as a candidate. A suite may cite it only in a position designated for candidates; the stub cannot satisfy admitted suite membership, closure, planned-baseline, wiring, gate, reuse, or import claims. - Introduced mechanism episteme.
MechanismDefinitionRefresolves to one A.6.1U.Mechanismepisteme with recoverable identity and sufficient content for the receiving use. Only this state can fill a position whose ValueKind isU.Mechanism.
A card, table row, file, or register entry may publish either state. Its layout and publication identity do not determine which state obtains.
Step 4: Complete mechanism semantics
An introduced mechanism has the A.6.1 identity tuple:
Its minimum semantic content for ordinary reuse names:
- direct
SubjectKindandRangedValueKind, withResultKind,SliceSet, andExtentRuleonly when current; OperationAlgebrawith one A.6.1OperationDeclarationper reused operation and one declaration-localArgumentDeclarationorResultDeclarationfor every typed argument or result position, including its meaning, exact ValueKind, binding designation rule, binding predicate, and any semantic cardinality;LawSet;AdmissibilityConditions;- Applicability through exact claim scope, selected time, reference plane when current, and mechanism-specific conditions;
SignatureManifestonly when actual imported or provided declaration content must replay.
An operation index may be derived from the declaration-local operation designators for retrieval; it is not another content group. Argument and result declarations remain inside their A.6.1 operation declaration and never become A.6.5 SlotSpecs.
Refinement, conservative extension, equivalence, bridge use, mechanism realization, evaluation, evidence use, method use, dated work, description, representation, and publication remain neighboring objects or relation occurrences. A MIP-run names their subject patterns. It does not copy the neighboring objects or relations into the mechanism declaration.
Create a new semantic edition when content, EntityOfConcernRef, or effective reference scheme changes. Keep the current edition when only a neighboring relation occurrence or publication changes. Use the current numbered A.6.1 conformance checklist. E.20 has no separate checklist-ID family.
If a suite or family claims shared operation-member vocabulary across several mechanism declarations, apply E.20:4.5.
Step 5: Keep operation-member names consistent across a suite
Use this step only when a suite or family claims that several mechanism declarations use a common vocabulary for operations, arguments or results by design. Repeated spelling by itself does not establish that claim.
-
The suite-subject pattern SHALL name one citeable vocabulary locus and the member mechanism declarations to which the shared terms apply. That vocabulary coordinates names only; it creates no
OperationDeclaration,ArgumentDeclaration,ResultDeclaration, ValueKind, binding predicate, or actual binding. -
Each member mechanism SHALL declare every current operation, argument, and result locally under A.6.1, including its exact meaning, ValueKind, designation rule, binding predicate, and cardinality. A cited shared term or equal spelling imports none of those semantics.
-
When a public shared term is introduced, renamed, split, or merged, update the shared vocabulary locus and every affected declaration or alias route. When only one declaration changes meaning, keep the change local unless the intended shared denotation also changes. Apply E.20:4.9 whenever citeability changes.
The shared vocabulary keeps the suite term consistent. Each A.6.1 operation member retains its declaration-local semantics. The vocabulary supplies neither an operation position nor an actual application binding.
Step 6: Suite integration (if the mechanism is a suite member)
If the introduction changes a suite (MechSuiteDescription or specialization):
- Membership set semantics (WF‑MS‑1).
mechanismsis a set: duplicates are nonconformant and list order carries no semantics. - Ordering is only in protocols. Express any order that matters in
suite_protocols. - Protocol closure (WF‑MS‑2). Each
ProtocolStepresolves to a selected member edition and a declaration-local operation. Alternatives listed as members must meet the suite's membership and edition rules. For a joint use, establish which operations and results must be compatible in the same configuration. Where one supplies another, use A.6.7 to check the handover. One feasible handover establishes a possible route. For a promise to handle every admissible output, determine who can select a branch, when the selection binds and what that participant can then distinguish. The selected branch must work for all outputs still possible on the information available at that point. A single receiving operation may already accept them all. Where branches join, follow their inputs back to the same permitted configuration. Two acceptable results belonging to different orders or editions can pass their local checks and still fail to provide the promised combined result. For a dated delivery check the actual handover. For an agreed risk allowance use its criterion. - No hidden tails. Required stages (e.g., normalization/aggregation/Γ‑fold) are explicit protocol steps; do not hide them inside other steps.
- Guard/gate separation. Suites and mechanisms SHALL NOT publish
GateDecision/DecisionLog.AdmissibilityConditionsand tri‑stateGuardDecisionremain governed by the mechanism definition;OperationalGate(profile)acceptance thresholds and pass/fail criteria remain gate/acceptance concerns. - Suite is descriptive only (WF-MS-3/4). A suite states membership, obligations, pins, and suite protocols. It does not restate
U.Mechanismidentity-bearing content. Any publication or telemetry continuation remains outside the suite protocol and requires its own exact publication or flow assertion and predicate.
Kernel stability rule (recommended). When a change would add a required stage to a kernel suite, prefer a suite variant to changing kernel membership. If mutation is unavoidable, pair it with terminology continuity (E.20:4.9) and RSCR triggers (E.20:4.10).
Step 7: Planned baseline & P2W planning-to-work boundary (if planning changes)
If the mechanism introduction changes what one U.WorkPlan pins, the WorkPlan edition is the identifiable planning object. Examples of pins that may change include selected comparator specifications, method descriptions, a time selector and guard pins.
- For a future use of an existing declaration member, introduce or revise
SlotFillingsPlanItemrows as declaration-local ClaimGraph content inside that WorkPlan edition under A.15.3. Use ordinary A.15.2 plan content when only a method or suite choice is being stated. - Give no row an independent kind, record identity, edition, specialization lineage, canonical target, or successor relation. Changing identity-bearing row content changes the WorkPlan's claim content and is handled as a WorkPlan-edition change under C.2.1 and A.15.2.
- Keep the declaration-local planned-filling content planning-only:
- positive intentions for planned values or designations, using the selected member's ValueKind, designation rule and semantic cardinality under A.15.3;
- no launch values;
- no
FinalizeLaunchValueswitnesses; - no gate decisions or decision logs; and
- conditions, including time, only when they change applicability or value selection, as specified in A.15.3:4.2; retain material edition pins under A.15.3:4.6 rather than implicit “latest” or “current” resolution.
- Resolve each row through the declaration edition and member selected under A.15.3:4.1–4.4. For an operation argument or expected result, cite the A.6.1 mechanism edition, operation designator and
ArgumentDeclarationorResultDeclaration; no Description wrapper is required. If the member's meaning or actual-use rule is undefined, returnmissing-governor. - When a receiver needs one row, cite it only through the WorkPlan edition and its stable local-content locator. The locator does not make the row independently resolvable.
The WorkPlan states the planned fillers. Enactment supplies witnesses for the actual runs.
Step 8: Wiring & SoTA updates (keep method evolution out of kernel)
For a method, comparator, selector or other SoTA-sensitive choice, retain the relevant source contributions and the editions used. A sufficient direct source reference can support the choice.
- Use a G.2 synthesis pack when the receiving synthesis or Part-G use needs that result. Its handoffs reach the subject definitions used for the choice: G.3 for CHR, G.4 for CAL and G.5 for registry/dispatch. The pack preserves grounds and references; those definitions and admission conditions remain applicable.
- If a WorkPlan changes, pin the chosen source or pack refs in that WorkPlan. Use declaration-local rows only for existing members under E.20:4.7; wiring consumes the selected planned values. It must not override those values without making the change visible.
- When flow/task binding or dispatch is current, put that logic in wiring modules (
GPatternExtension) with an explicitPatternScopeIdand declared subject pattern. - Wiring may bind, select, dispatch or cite method packs. It may not redefine the mechanism's identity-bearing A.6.1 content. When wiring names a bridge, realization, evaluation, evidence-use or publication claim, apply the pattern governing that relation.
- If a source update changes the mechanism's signature or laws, the author SHALL change the mechanism-subject pattern under the A.6.1 definition template and SHALL emit the affected RSCR triggers under E.20:4.10.
Step 9: Terminology continuity (alias docking)
If the introduction renames any public token or changes canonical naming:
- Use lexical alias docking (F.18) so old tokens remain citeable.
- Update registers and twin labels per lexical discipline.
- To avoid silent rewrites, the MIP-run manifest SHALL state the alias relation and the successor relation.
Deprecation / supersession / retirement (preserve citeability)
If the change class includes deprecation, supersession, or retirement (E.20:4.1 #8), the author SHALL preserve reference continuity and state the status change:
- Preserve each identifiable target. A deprecated
U.Mechanismepisteme, reference-reservation stub, suite description, exact WorkPlan edition, or wiring module SHALL remain resolvable at its canonical location. Deprecation MUST NOT remove it and break citations. A declaration-local planned-filling row is not another canonical target. - Keep the public token citeable. A deprecated token such as a
MechanismDefinitionRef, suite token, WorkPlan token, public local-content locator, or wiring token SHALL remain citeable. If a successor token or name is introduced, alias-dock the old token under F.18 (E.20:4.9). A local-content locator still resolves only through its WorkPlan edition and creates no independent row identity or edition. - Declare a successor or state that none is current. Apply that obligation to the deprecated mechanism episteme, reference-reservation stub, suite description, WorkPlan edition, wiring module, public locator, or alias under its direct supersession or deprecation pattern. A changed planned-filling row contributes to changed WorkPlan claim content; it has no separate successor relation.
- Update the definition that owns each change. Make each needed change to suite denotation, closure, obligation, pin, protocol semantics, WorkPlan content, or wiring semantics at its definition locus in E.20:4.2. Prefer a suite variant to an unannounced change in kernel membership.
- Emit RSCR triggers. Deprecation or supersession SHALL emit typed RSCR triggers and extend the regression envelope (E.20:4.10), including checks for dangling references and alias coverage.
Step 10: RSCR triggers + regression envelope
The author SHALL emit typed RSCR triggers through the RSCR subject pattern when a MIP-run changes any of the following:
- mechanism signatures,
- suite membership/protocols,
- planned baseline pins,
- shared operation-member vocabulary or declaration-local operation, argument, or result designators,
- terminology/alias docking that changes citeable tokens,
- or other reference loci
The author SHALL extend the regression envelope over the affected uses:
- reference resolution for changed citeable references;
- membership set semantics and protocol closure when a suite changes, with compatible conditions and usable handovers for any promised joint use;
- guard/gate separation when that boundary changes; and
- the P2W planning-to-work boundary when planned-baseline content changes.
Trigger identifiers. Trigger kind identifiers (e.g., RSCRTriggerKindId) SHALL be selected from the RSCR trigger catalogue governed by G.Core. Authors SHALL NOT create ad hoc trigger kinds (“reason kinds”) in individual patterns or modules.
Recommended manifest contents. The MIP-run manifest SHOULD list emitted trigger types and changes to the regression envelope as checkable items.
Step 11: Apply PQG profiles (E.19) and close the run
Every MIP-run SHALL be reviewed using PQG (E.19) with:
- PCP‑BASE always, and
- the triggered profiles implied by the change class (at least):
- PCP‑SUITE if any suite locus changed,
- PCP‑P2W if any planned-baseline locus changed,
- PCP‑TERM if any new terms/renames are introduced,
- PCP‑SOTA if SoTA packs are introduced/modified,
- PCP‑NORM if the run introduces/changes normative requirements or conformance items,
- PCP‑DEONT if RFC keyword clauses are introduced/modified (or if invariant/predicate vs deontic form is ambiguous),
- PCP‑BRIDGE if cross-context reuse, crossings, or bridges are introduced or changed,
- PCP‑REFRESH if refresh-sensitive claims (SoTA lists, “current practice”, enumerations) are touched,
- plus any applicable modularity / boundary / normativity profiles required by the delta.
MIP-run outcomes (normative set). A reviewed MIP-run SHALL be closed as one of:
- Proceed (single change set).
- Proceed via governing-definition split (mandatory when semantics were placed under the wrong governing definition; split the change so that each edit is made at its governing definition).
- Proceed via suite variant (preferred when kernel stability is threatened by adding new required stages).
- Block with explicit missing condition (the mechanism semantics are insufficient, a reservation stub exists, and a DRR tracks the condition for completing the mechanism declaration).
- Reject (violates invariants such as suite-as-gate, plan-as-enactment, or governing-definition ambiguity).
Archetypal Grounding
Optional citation, no suite or mechanism change. A suite description adds only an optional citation of an already-introduced U.Mechanism episteme by its MechanismDefinitionRef. Its admitted membership, denotation, closure, obligations, pins, protocol semantics, and relied-on neighboring relations remain unchanged. Name the suite-governing locus and stop. This citation opens neither a new mechanism declaration target nor a MIP-run manifest. A membership addition instead uses the suite-change branch.
Bias-Annotation
This protocol emphasizes stable definitions and recoverable dependencies. Select its detail by the change: an optional citation can stop at its governing locus, while changed membership, laws or admissibility require their own subject repair.
Conformance Checklist (normative)
Conformance use. This checklist tests the governing-definition assignment guidance already stated in the Solution. It is not the first entry text for ordinary use or a mandatory full-corpus check; an item is applied only when its corresponding trigger triage, manifest, declaration target, suite, planning, wiring, lexical, RSCR, PQG, or deprecation move is present. Before applying an item, name the Solution guidance it tests. When that use is absent, treat the item as orientation only or as inapplicable.
Conformance groups. Ordinary E.20 use starts with trigger triage and stops at the current governing locus when no denotation or mechanism-meaning change is present. Manifest-core items apply only when a MIP-run is triggered. Publication and assurance items apply only when citeability, reference-reservation stubs, alias docking, RSCR, PQG, or deprecation continuity is part of the current claim. When a claim concerns crossing, launch or work enactment, use the gate, planning, or work loci and keep E.20 to governing-definition assignment.
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits
- Mechanism introductions become trainable and reviewable because authors can apply the same rule to assign each changed item to its definition.
- Assigning each mechanism meaning to one subject pattern reduces drift. Its semantics stay with that definition.
- Keeps suites descriptive and the P2W planning-to-work boundary inspectable.
- Supports SoTA evolution without destabilizing kernel meaning.
Costs
- Introductions use more explicit assignment records (governing-definition map, PQG coverage).
- Some changes will be split into multiple governed edits, which takes additional authoring work.
- Kernel stability discipline can feel “slow” when a team wants a quick mutation.
Rationale
Mechanism declarations are high-leverage epistemes: a small change can affect suites, planned baselines, wiring modules, evaluations, and evidence uses. Without a protocol, the corpus tends toward semantic duplication across governing loci, so a reader cannot recover which declaration or neighboring relation changed.
Authors can follow the protocol without tooling. The stable mapping from each change to its governing definition supports later review and refresh.
SoTA-Echoing
Relations
Builds on:
- E.8 (pattern structure and normative authoring discipline)
- E.10 / F.17–F.18 (lexical registers, twin labels, alias docking)
- E.19 (PQG/PCP profile-based review)
- E.15 (change between exact pattern editions, actual-delta classification, affected reach, and edition continuity)
Coordinates with:
- A.6.1 (
U.Mechanismdefinition template governance) - A.6.7 (
MechSuiteDescriptionintegrity) - A.15.2/A.15.3 (exact
U.WorkPlanidentity and declaration-local planned-filling content) - E.18 (
TransformationFlowStructurevalues that cite planned baselines) - G.Core (RSCR trigger catalogue)
- G.2 (SoTA synthesis packs)
- G.x:Ext.* (wiring modules via
GPatternExtension)
Constrains:
- Any change set that introduces or revises mechanisms, suites, planned baselines, or wiring in a way that changes citeable loci.
E.20:End
FPF Pattern-Quality Evaluation CharacteristicSpace
Type: Pattern Status: Stable
Problem frame
Use E.21 to evaluate the quality of an authored FPF pattern edition or bounded version for a named use. Its result can guide ordinary practitioner use or supply input to authoring, landing, release, external review, high-assurance reuse, canonization or another stated pattern-quality use. E.21 declares the characteristic space, evaluation specification and result rules. The evaluator applies the questions to that edition or version under the required ClaimScope.
If the pattern fails the required use, the result episteme states repairBeforeUse, holdForArchitectureDecision or refreshNeeded. A different use needs its own evaluation frame and cannot rescue the original result.
Not this pattern when the evaluated object is one DRR, an FPF-level corpus object, a single wording repair, a source-use decision, or a project-side evidence, assurance, gate, release, safety, compliance, work, or decision claim. Use E.9.DA for a DRR, E.2.DA for an FPF-level corpus object, F.19 for a wording repair, and the pattern governing a source-use or project-side claim for that claim. Open E.10 or a named precision-restoration neighbor for an unresolved FPF-specific meaning.
First useful move. Recover the work expected of the pattern's reader from the governing request. What will the reader receive, what must they choose or construct, and for whom? Name the edition, required use, scope and qualification window. Follow that work from the working situation to the result a reader can obtain with the guidance. Use this comparison when assigning each coordinate its evidence-based value and adjacent-value rationale, then form the aggregate result episteme.
floorEvaluation changes only the declared floor and expected cost of obtaining evidence. An E.21 result retains the required ClaimScope, full coordinate set, rationales, and PrecisionRestorationProfile. Evaluate fragmentary, ill-formed or weak pattern text under the required scope. For an established weakness, record low coordinate values, repair status, an architecture hold or refresh, as applicable.
What goes wrong if missed. Reviewers can mistake taste, checklist closure, source count, review or landing state, or text length for pattern quality. Short patterns can pass while missing mature content; long patterns can pass while hiding the first user-facing action; semio material can take over a non-semio pattern.
What this pattern provides. The result makes one scoped, non-arithmetic PatternQualityQBundle claim about one exact pattern edition. It gives the complete coordinate set, evidence basis, adjacent-value rationales and a visible stop, repair, hold or refresh status.
What the evaluation is about. Its primary EntityOfConcern is one exact authored FPF pattern edition or bounded version checked under one declared quality scope and qualification window. Keep the quality questions, evaluator, coordinate claims, aggregate result, evidence use, any admission decision, and later repair distinct. Use the dated-Work account in E.21:4 and CC-E21-0 only when a later claim needs a dated assessment-Work account.
Problem
FPF patterns need an evaluation that goes beyond a style checklist while requiring less work than a project assurance audit. Earlier review habits produced two opposite failures:
- Too weak. A reviewer marks a pattern "ready" because no blocker is obvious, because it landed, or because headings exist.
- Too heavy. A reviewer adds more warnings, evidence cards, source rows, boundary notes, and process residues until the pattern becomes harder to use.
E.21 addresses both failures by evaluating the quality of the pattern on every required coordinate. The coordinates ask whether the pattern is usable, coherent, current, precise, affordable, mature enough for its claim, and safe from proxy improvement.
Forces
Solution
E.21 declares the FPF pattern-quality U.CharacteristicSpace, its object-specific A.19.ECS evaluation specification, ordinal scale, complete result-shape rules, the local non-arithmetic PatternQualityQBundle result payload, and local result-status meanings.
An evaluator applies these questions to the pattern and assigns its coordinate values. Evidence use, assurance, admission, and later repair have their own objects and relations below.
Apply the evaluation in six steps:
- configure the checked pattern edition, scope, use, reader, window, characteristic space and specification, and evidence basis; include the exact semantic evaluation Method only when its identity or actual assessment Work is asserted;
- let the evaluator apply the specification; add the dated-Work account below only when the result deliberately asserts assessment Work, and add an
A.6.1application only when the conditional rule below is independently satisfied; - constitute every coordinate-result claim with
ShortRationaleand the aggregate result episteme; - assert the local
PatternQualityStatusin that result; - state its stop, repair, architecture-hold, or refresh condition; and
- when improvement is requested, return distinct finding or proposal claims without changing the coordinate result into a work plan.
For one pattern-quality evaluation, keep each object and relation asserted in the selected ordinary or Work-bearing form independently recoverable:
- one exact authored FPF pattern edition or bounded version as the checked object;
- the declared
ClaimScope, working reader, intended receiving use, qualification window, evidence basis, and evaluation configuration; - the selected
U.CharacteristicSpace, this E.21 evaluation-specification episteme, every coordinate/scale binding, and the local result-form and status-value rules; - when exact Method identity or actual assessment Work is asserted, one separately identified semantic evaluation
U.Method; - when actual dated assessment
U.Workis asserted, first recover every evaluator-performer's A.13 core for the assessment action. A.15.1 then independently admits the Work from its performance history, enacted Method, temporal extent, and one obtaining locally declared relation to the containingU.System, under the exact system boundary and qualification window. Add F.6 only when the evaluation account also needs precise assignment-bound attribution, using the same obtaining A.13 assignment. A compact account may omit an identifier unused by the receiving claim only when every relation it consumes remains recoverable; - every coordinate-result claim, their same-bearer non-arithmetic
PatternQualityQBundleClaimGraph payload, and one C.2.1 aggregate pattern-quality-result episteme when a durable result is needed; - witnesses, comparator/source/case refs and actual support for the coordinate claims; A.10 source-to-use relations and bounded reliance results when the account is implicit or must remain replayable; and any B.3 result for an actual named assurance claim;
- an optional evaluation-record episteme that packages those refs;
- the local
PatternQualityStatusvalue and any separate F.10 status use/interpretation, E.19 admission or refresh decision, project gate or authority decision, publication, and currentness relation; and - later E.23 improvement or other repair work and its changed pattern edition.
A.6.1 enters only when a separately admitted U.Mechanism declares the exact operation that was used and the receiving claim needs that application occurrence or its bindings. Identify the mechanism and operation. The account must include that operation's ApplicationPredicate, ApplicationIdentityRule, ApplicationExtentRule, argument and result declarations, declaration-local binding predicates, exact application occurrence, and actual declaration-local bindings.
Treat the checked pattern, configuration, coordinate results, and aggregate result as application inputs or results only when the operation declares those exact meanings and the corresponding bindings obtain. Otherwise omit PatternQualityEvaluationApplicationRef; the dated-Work and result accounts remain complete without it.
Any local evaluator system-role kind and independently obtaining System-classification judgment are optional separate claims. Route unresolved source role through E.10.ROLE.
Each coordinate-result claim is one quality ascription about the exact checked pattern edition. It keeps recoverable the bearer, effective ReferenceScheme, characteristic, scale value, evaluation rule or probe, comparison or calibration frame when used, U.ClaimScope, intended use, qualification window, ordinary assessing action or exact declared-operation application when separately asserted, short rationale, and evidence locus.
The complete same-bearer coordinate set forms the non-arithmetic PatternQualityQBundle payload carried by the aggregate result episteme. The evaluator system, evaluator viewpoint episteme if any, witness set, optional record, and receiving status or admission use remain separate.
If a pattern lacks frame, first move, source basis, mature comparison, or naming clarity, lower the relevant coordinates in the one E.21 result.
A bounded lexical, checklist, or automated smell screen may identify suspect loci and reduce search cost. Record the checked edition, covered defect family, and observed limits in EvaluationEvidenceBasis; the screen neither assigns a coordinate value nor establishes semantic completeness, practical use, or the aggregate result.
When candidate editions are compared, keep the declared use, reader, probes, and evidence conditions common where possible and expose missing or underrepresented evidence. These conditions let another evaluator replay the comparison. They do not turn ordinal coordinates into one score or establish agreement between evaluators without a study of that agreement.
An E.21 result evaluates one exact edition for one declared use. It does not validate the pattern universally. A stronger validation claim needs separately declared expert checks, observed applications or cases, or other fit-for-purpose research evidence. Missing actual-use evidence therefore caps only the coordinates whose stronger values require it.
Local names and kind settlement
Names are local to pattern-quality evaluation unless F.18 promotes a durable name. Each has only the direct function stated above; any later receiving use requires its own relation.
Evaluation configuration, application, result, and optional record
An unfinished table, prose summary, or record with missing coordinate claims remains assessment material. A complete E.21 result places every required coordinate claim in the result episteme; the objects named in E.21:4.1 retain their stated functions.
Ordinal scale, result row, and adjacent-value rationale
The values express ordinal judgements about pattern content. A C.16 measurement, average, percentage, maturity-ladder step, review vote or landing status answers a different question.
The result-bearing coordinate row has exactly this shape:
For values 1..4, explain why the lower adjacent value would understate the evidence and the higher adjacent value would overstate it. For 0, explain why 1 would overstate the evidence and what would raise the value or reopen it. For 5, explain why 4 would understate the evidence and what would lower the value or reopen it.
A two-column coordinate-and-value table, a narrative paragraph, a table whose comment lacks adjacent-value comparison, or a result whose value depends on unchecked external loci is not an E.21 result. Keep it as draft evaluation material until every coordinate has a ShortRationale row and the result names the EvaluationEvidenceBasis used for values that depend on source, comparator, corpus, projection, or worked-case evidence.
A ShortRationale can be brief. It must give its evidence. When the value depends on a source-currentness row, mature comparator, README scenario, ToC row, E.11 entry-distribution locus, including a worked comparison, card, retrieval cue, monolith section, worked slice, near-miss, or anti-case, the rationale names that locus by value or says that the locus was missing or unchecked.
"By value" means a recoverable section, row, case, checklist item, relation, source row, projection row, comparator id plus selected ingredient, or specific absent locus; a category list such as "entry, first move, boundaries, SoTA, checklist, relations" is insufficient.
An established absence of required content or support lowers the affected value under its declared anchor. An unperformed check or unavailable basis leaves the dependent value unestablished: name the missing judgement and retain the complete evaluation as draft material under the preceding rule. Preserve independently supported coordinate values. This evidence gap supplies neither a low property value nor an admissible-use result.
For SoTABindingAndCurrentness, source identity and currentness support traceability only. One completed canonical E.8:11 comparison supplies the required comparison basis; the evaluator assigns the value from that substantive comparison and its binding into the checked pattern.
Clear early wording, named neighbour relations or a well-formed set of fields alone do not justify a 5. To assign 5, establish exceptional expression for the declared use: reinforcing loci, a worked or otherwise replayable slice where the coordinate demands one, and no hidden cost or neighbour loss. When the evaluator cannot say why 4 would understate the evidence, assign 4 or lower.
When a coordinate's 5 meaning names a filled case, replayable slice, near-miss, anti-case, worked comparison, projection evidence, currentness basis, or selected-neighbour replay, absence of that evidence caps that coordinate at 4 even if the prose is otherwise strong. Do not hide the same absence only in CaseCountercaseAndTransferCoverage; lower every coordinate whose own 5 meaning needs that missing evidence. A 5 rationale names the reinforcing evidence loci that make 4 too weak.
For MaturePatternParityAndSelectedContentSufficiency, the rationale names a mature-pattern comparison set and the selected mature ingredients being claimed. For non-epistemic patterns, include at least one mature non-epistemic comparator when one exists—for example, a mature pattern about Work, Method, a system-role kind, a system-role assignment, direct-relation participation, a System, control, architecture, selection, engineering action, or another primary EntityOfConcern that is neither an episteme nor a publication. Route an unresolved source role through E.10.ROLE rather than treating the word as one pattern family.
Value 4 requires that the rationale locate the selected ingredients in the body or neighboring pattern that defines or constrains the claims; comparator IDs plus a generic "main ingredients are present" sentence are only value 3. The comparison is not a length target and not permission to copy semio apparatus.
For a 4 or 5 on MaturePatternParityAndSelectedContentSufficiency, include a compact maturity-discharge payload in the rationale or CoordinateEvidenceRefs: comparator=<pattern id>; selectedIngredient=<ingredient name>; currentLocus=<section, row, case, checklist item, relation, or neighboring pattern governing the claim>; missingOrLowering=<absent or weak ingredient, if any>. A category list such as "frame, first move, neighbour relations, CC, SoTA, relations" without current loci warrants only 3, even when the listed categories are plausible mature ingredients.
Use the question-to-ground and result-acceptance operation in E.22:4.3 within this evaluation.
Precision-restoration profile
Before assigning the coordinate table, apply the whole-span precise-language reading in F.19 and record one PrecisionRestorationProfile summarizing its quality effect. F.19 governs the reading, repair and local revalidation. E.21 uses their result to judge its existing coordinates. The reading asks which governed object, claim, relation, and reader use the sentence, table, section, or repeated content family serves in the pattern of concern.
Use F.19's evidence-selection and coverage conditions for recoverability. Name the kind of basis and its material reading conditions in EvaluationEvidenceBasis; use checkedLoci for the text and relations it covers. A clean profile retains those qualifications. If F.19 requires an actual reading that is missing, the dependent recovery claim lacks the required evidence. Apply E.21:4.3's missing-evidence rule to the affected coordinates and identify the missing judgement. Distinguish this evidence gap from a wording defect established by reading the text.
Use this compact shape:
The diagnostic fields below are optional. Retain a field when its finding, restoration choice, or bounded evidence changes the quality result. A clean result needs no separate clean entry for each field. Fuller profiles use the same field meanings. When a receiving form needs an explicit untriggered kindRestorationCheck disposition, it may use not triggered, ordinary prose, or no FPF-governed phrase changed with the checked locus; E.21 leaves this detail optional.
The overall effect is the strongest quality effect that any layer requires: clean, bounded local repair, coordinate lowering, or repair-before-use. Classify a new symptom under the relevant diagnostic layer or restoration locus and apply its effect to existing E.21 coordinates.
F.19 settles ordinary whole-span language questions. Open E.10, E.10.ARCH, or F.18 for an unresolved word, head, or name problem; for hidden candidate ontics and ontic-vs-description-vs-publication boundaries, use E.24.CD, E.24.PUB, or the concrete pattern content that defines or constrains the disputed object; claim, relation, evidence, Work, decision, assurance, publication, or pattern-application problems return to the pattern content that defines, constrains, or tests the disputed item. A pattern reference locates the content whose concrete contribution the evaluation uses. Exact predicates and ClaimGraph identity are required only when the evaluated claim or named reliance needs them.
E.21 consumes only the result: which coordinates fall, which stay protected, and what repair would make the quality claim true.
The mgdaColdReaderRecoverability layer asks whether a reader without the DRR, campaign notes, or evaluator memory can recover the object, kind or ordinary status, relation or claim position, admissible use, next exact assertion when one is needed, and next concrete defining, constraining, or checking contribution. If a repair replaces a specific phrase with object, item, value, relation, record, condition, basis, material, or unqualified specialization and the reader cannot recover what specializes what, which relation is live, or which assertion or concrete pattern contribution is required, this layer is not clean.
When the recovery check finds a hidden candidate ontic or publication-form confusion, the E.21 result records only the quality effect and affected coordinates. Candidate detection, ontic placement, slot-relation design, and publication-boundary repair remain with E.24.CD, E.24.PUB, or the concrete pattern whose content defines or constrains the affected object.
The kindRestorationCheck is required when a changed FPF-governed expression can alter the meaning-bearing object, kind, relation, current ontic slot, relation position, use relation, claim kind, admissible use, or scope. It records those live values before and after the proposed repair, then names the concrete contribution when another pattern defines, constrains, or tests the affected kind, relation, claim, or position (A.6.0, A.6.5, A.6.P, C.29, A.15, E.24.CD, E.24.PUB, E.10.ARCH, or another relevant pattern). Every value that can drift receives an explicit preserved, split, intentionally changed by accepted decision, or blocker disposition. When no such risk is present, F.19's ordinary local revalidation is sufficient and the profile omits this field.
The underlying slot, ontic, publication-form, and mathematical-lens rules remain with their subject patterns. A lexical replacement is not a repair when it only removes a trigger word, substitutes one umbrella for another, narrows a graph or method into a work sequence, widens a work occurrence into a method, turns a publication form or evidence source into the object itself, or otherwise changes kind, current ontic slot, relation position, use relation, or claim kind without an accepted decision.
If the kind, current ontic slot, relation position, use relation, or claim kind cannot be recovered, the profile is at least lowersCoordinates; if the proposed repair would change one of them and no accepted DRR or concrete defining, constraining, or checking pattern content justifies that change, the result is repairBeforeUse or holdForArchitectureDecision.
When the profile is not clean, lower every affected coordinate named by the profile. Do not confine a precision-restoration issue to EntityOfConcernPrimacyAndSemioBiasResistance, and do not raise the result through related-pattern-boundary praise, projection evidence, or "correct but true" guards when those materials compete with the pattern's own EntityOfConcern, first useful move, practitioner action, practical delta, or next useful action.
RequiredPatternQualityCoordinates
For every conforming E.21 result, an evaluator applies the evaluation specification to every coordinate, and the result episteme states every coordinate value and rationale.
Constraint, harm, safety, security, compliance, deontic, self-application, recursion, and high-assurance questions do not add a second coordinate family. Evaluate them through the applicable coordinate for that content: related-pattern authority, traceability, formal-claim admissibility, falsifiability, affordability, corpus ecology, evolution, or refresh.
Coupled-flow unity and separation for pattern quality. Use this account when the declared quality use needs the relation between development, use, evaluation, and repair flows. Dated E.21 assessment work evaluates one exact PatternOfConcernRef inside a development, refresh, or admission flow. Another flow may make the same pattern a pattern of concern for a different use relation, for example a practitioner selecting and using it, a reviewer applying it to another text, or subsequent assessment work reopening it.
One TransformationFlowStructure may join pattern development, pattern use, use-found evaluation, and repair or refresh flows through transfer, feedback, return, edition-change, or projection relations only when every included valuation or internal portion resolves to that same TFS and structural boundary. Independently identified TFS values and relations across their boundaries use E.18.NET.
Keep three positions distinct in each sentence: the pattern as concern of the current flow, the intended reader addressed by the pattern, and the pattern's own primary EntityOfConcern inside its Problem, Solution, or guidance.
E.21 and E.19 are specifications; dated assessment and review work are the checking operations; handoffs, ledgers, README, ToC, E.11, retrieval outputs, and landing evidence are distinct records, publications, or evidence loci in the development/evaluation flow. Those objects may support edits to the pattern, but they are not automatically user-facing content for the reader addressed by it.
DesignRunTag stays on the subject-context, claim, work, trace, publication-form relation, or source relation inside the transformation-flow structure; recover currentness, obsolescence, development, and use from their own relations. In pattern development, use quality-loop evidence to guide separately performed repairs and keep that evidence in the evaluation record.
Frequent value-3, value-4, and value-5 calibration points
These rows calibrate common disagreements. They do not replace the coordinate definitions above.
For EntityOfConcernPrimacyAndSemioBiasResistance, do not compensate a bad PrecisionRestorationProfile with NeighborAuthorityAndBoundedUseFit or CorpusEntryProjectionAndEcologyFit. Ask which governed object, claim, relation, and reader use the sentence serves. Material about developing, reviewing, projecting, landing, evaluating, or proving this pattern's quality belongs in the evaluation, projection, release, or publication locus that carries that work. Related-pattern statements can be true and still damage the pattern when they precede its own EntityOfConcern and application guidance.
If the opening Problem frame or Solution starts with precision-restoration material before the subject and move, this coordinate is at most 2; if the reader must traverse that material across sections to find the action, it is at most 3.
Put compact concrete contributions in Relations or a late boundary row. Add local guard prose only when it passes F.19:4's full independent-ground, plausible-reader, contribution, and smallest-clear-correction test and the subject pattern does not already settle the needed distinction. Also lower PatternApplicationGuidance, WorkingSituationAndUseBoundaryRecognizability, PracticalUseDeltaAndHarmPrevention, and UseAffordabilityAndApparatusProportionality when the profile shows that auxiliary material displaces first use.
If the declared use is Stable, landing-input, release-input, external-review-ready, or another corpus-facing use, assessment work must inspect the applicable corpus-entry and projection evidence and the result's EvaluationEvidenceBasis must name it. A host-only assessment can produce values about the pattern body, but it cannot silently turn missing README, ToC, E.11, card, retrieval, monolith, or projection evidence into a high CorpusEntryProjectionAndEcologyFit value.
Status and stop condition
Default floor is 4 wellExpressedForDeclaredUse on every coordinate for ordinary practitioner use, authoring-input use, landing-input use, Stable, external-review-ready, release-input, canonization-input, stop-improving claims, and ordinary improvement-loop use.
Every result presented as an E.21 result contains every coordinate and its rationale, including a diagnostic or exploratory E.21 result. A bounded diagnostic may borrow selected E.21 questions; it reports only their findings and makes neither an E.21-result nor an admissible-use claim.
If the request asks for corpus-facing, landing-input, Stable, release, or external-review use, the evaluator measures that required use and returns repairBeforeUse, holdForArchitectureDecision, or refreshNeeded when the floor is missed.
An all-5 result is a local exceptional result under the declared scope and qualification window. Development may continue. E.23 can reopen improvement when use, source, comparison set, front, affordability, or payoff changes.
Consume Pattern-Edition Use-Value Evidence Noncompensatorily
When an E.19:4.3.3 replay is current, use that one stable-candidate replay as evidence for the E.21 assessment. During assessment, keep materially affected predecessor and candidate-only uses distinguishable whenever their action, result, boundary, necessity, or consequence can differ; do not copy clean per-use dispositions into the E.21 result.
The evaluator applies every existing coordinate required by the declared scope once. The result names the replay loci in EvaluationEvidenceBasis and carries only distinctions, failures, or improvements that change a coordinate rationale or PatternQualityStatus; it does not replace the coordinate set with one use-value score, average replay results, or infer a coordinate value from an E.19 outcome.
Apply these consequences:
The cap is 2, not 3, because 3 sufficientlyExpressedForDeclaredUse already means usable for the declared scope while the required action or semantic member here is unusable.
Unrelated strengths, source count, formal cleanliness, or corpus projection cannot compensate for the failed required use. Conversely, preserved, improved, transferred, intentionally retired, or adequate candidate-only evidence can support only the existing coordinates whose claims it tests; it cannot raise unrelated coordinates or determine status by label.
Compact result form
An E.21 result uses this result-bearing form:
Include one PrecisionRestorationProfile under E.21:4.3a: overallEffect, checkedLoci, and affectedCoordinates, plus issue-bearing detail only when it changes the quality result.
Coordinate values.
When SoTABindingAndCurrentness is 4 or 5, the result also includes one completed instance of the canonical E.8:11 comparison contract in the rationale or immediately after the coordinate table. Record that comparison using the E.8:11 fields below:
Source identity, publication status, currentness, and maintenance evidence may support ClaimJustificationTraceabilityCurrentnessAndReplayability and the qualification window. They cannot fill bestKnownLine, raise this coordinate, or replace the comparison payload.
The header, compact PrecisionRestorationProfile, complete coordinate table with ShortRationale, required evidence basis, and stop and reopen conditions constitute the E.21 result. Incomplete material supports further assessment. The result asserts local PatternQualityStatus.
Separate E.19 review work and result, plus the authority-bearing release or admission work or decision named by value, govern gate-specific carry-through, projection, monolith, packaging, authority, and receiving-use boundaries.
Finding and proposal rows
When E.22, E.23, returned-finding absorption, or exceptionalImprovementEvaluation asks for improvements, cover every below-floor coordinate with a finding and add proposal rows only for substantive non-dominated improvement opportunities inside the declared scope.
Record one finding for one independently repairable defect; its Coordinate or status affected field names all affected coordinates, while each coordinate keeps its own value and rationale. A receiving E.22 typed proposal retains its one-coordinate interface; coordinate-specific proposals may share one correction description and closure test.
Do not treat every value below 5 as a defect. For above-floor coordinates, the evaluator searches by value when exceptional improvement is requested, but the proposal must name a content improvement such as stronger positive action guidance, a worked slice, case or countercase, source-currentness carry-through, mature-content discharge, relation cleanup, deletion of displaced apparatus, split of overloaded content, or another content gain.
When ending improvement at the current values, give one aggregate no-proposal or stop disposition showing why further substantive change is dominated, unavailable, or outside scope. Cite the checked loci and relevant coordinate rationales; keep independently different reasons recoverable within that disposition.
Archetypal Grounding - worked slices
Complete compact evaluation
Exact example edition EX.1@source-pin-1. The quotation gives the entire edition evaluated here. Campaign notes and unstated appendices are outside it.
EX.1 - Pin a reused rule to its source edition.
Use this when a team relies on a rule from a source that can change.
First move: beside the decision, record the source title, exact edition or date, the exact rule used, and what that rule changes in the decision.
If the edition or rule cannot be recovered, stop that reuse and retrieve it.
Not this pattern when the source is background reading and no claim or decision relies on it.
The pin lets a reader recover which source rule changed the decision.
Example: a team records
Cooling Guide, edition 3, rule 7beside the chosen inspection interval and notes that rule 7 sets the maximum interval; a mention of the guide in a reading list is outside this use.Reopen the decision whenever the source publishes any new edition.
The final sentence is deliberately defective: an unrelated editorial revision would trigger the same reopen as a change to rule 7. The evaluation below uses that text with the defect in place.
Configuration and evidence basis.
PatternOfConcernRef: the complete quotedEX.1@source-pin-1edition.ClaimScope: first-use assessments of the quoted edition from its supplied text under the conditions of this rehearsal. The declared floor is3 sufficientlyExpressedForDeclaredUsefor this rehearsal.WorkingReaderScope: a new evaluator who has E.21 and the quoted text but no campaign history.IntendedUse: learn whether this edition is coherent enough for the diagnostic rehearsal and identify the first repair; if EX.1 is later proposed for publication or ordinary authoring use, that receiving use requires its own admission decision.QualificationWindow: until the quoted edition, E.21 scale, or named comparison evidence changes.EvaluationEvidenceBasis: all seven sentences of the quoted edition; its filledCooling Guidecase; its background-reading non-use boundary; the absent material-change test in the last sentence; E.2.DA's pinned source-use discipline and G.11's bounded currentness contribution as mature comparators; no README, ToC, retrieval, external SoTA, observed-use, or corpus-projection evidence.- Ordinary path only: the evaluator reads and judges the text; this diagnostic use needs no additional reliance-bearing identity or receiving decision.
The profile, complete table, status, stop and reopen, plus the quoted pattern's grounded background-reading boundary, constitute this example's non-arithmetic PatternQualityQBundle. The single value of 2 is the one below-floor defect, not an arithmetic penalty or a reason to lower unrelated qualities.
This is the ordinary path. The evaluator needed no dated-Work account or operation-application record to produce a complete result.
Precise names without a first move. A pattern has precise Tech names and current source rows but no first user-facing action. WorkingSituation..., PatternApplicationGuidance and PracticalUseDelta... fall. Current source rows cannot establish ordinary usability.
Short architecture pattern. A compact pattern has a triage form but no worked slice and no mature-pattern comparison. It can be useful as local expert reference material, but MaturePatternParity... and CaseCountercase... stay below exceptional until selected mature content is present.
Precision-restoration profile in a non-semio pattern. A pattern tries to introduce a non-semio EntityOfConcern through a catalog of other claim kinds or objects outside its own subject. Its membership rule—outside this pattern's subject—does not explain which distinctions the reader needs. If copied boundary doctrine leads the Problem frame or Solution, EntityOfConcernPrimacyAndSemioBiasResistance falls to 2 or 3 even when every individual boundary is true.
Lead with this pattern's own subject, first useful move, practitioner action, practical delta, and positive guidance. Add one local explanation, stop, or non-use boundary only when it passes F.19:4's full independent-ground, plausible-reader, contribution, and smallest-clear-correction test. Replace other copied doctrine with the relevant pattern ID and its concrete contribution. If the doctrine is distributed across sections, repair that distribution rather than only its sentences.
Reference apparatus before Solution content. A pattern's first Solution paragraph assigns other patterns or related-pattern mappings before it unfolds the ontology, method, norm, worked action, or other positive solution for the pattern of concern's own EntityOfConcern. Even if the related pattern id is correct, PatternApplicationGuidance, EntityOfConcernPrimacyAndSemioBiasResistance, PracticalUseDeltaAndHarmPrevention, and sometimes NeighborAuthorityAndBoundedUseFit fall.
Move discoverability to README, ToC, E.11, or retrieval loci; put compact pattern references and their concrete contributions in Relations or a late boundary row; put architecture-placement rationale in a DRR or architecture document; and make the Solution answer “what do I do with this pattern's EoC?” first.
Overformalized precision. A pattern uses correct FPF kinds, slots, references, and cross-pattern pointers so densely that the working reader cannot recover the first useful move, practical delta, or generalizing insight without auditing the text themselves. Lower UseAffordabilityAndApparatusProportionality, WorkingSituationAndUseBoundaryRecognizability, and sometimes PatternApplicationGuidance.
Repair by keeping the ontology named by value only where it carries a current FPF-governed claim, moving restoration evidence to the evaluation result or DRR, and adding a short worked slice or plain recognition sentence that preserves the same kind without extra apparatus.
QualityEvidenceLeakage in the pattern. The pattern says that corpus projection, README, ToC, E.11 alignment, retrieval or cold-reader evidence, monolith parity, external-review readiness, landing evidence, PatternQualityStatus, all-4 or all-5 result framing, or another quality-result locus is what the user should do with the pattern's EntityOfConcern, or records developer, reviewer, or executor correspondence as if it were pattern content.
The defect is not limited to Problem frame, Solution, examples, or checklist; notes, appendices, Relations, Rationale, SoTA-Echoing, tables, and conformance rows are also parts of the pattern in hosts and the monolith.
That evidence may be required for E.21, E.19, landing, or retrieval loci, but it is not automatically a user action in the pattern of concern. Lower EntityOfConcernPrimacyAndSemioBiasResistance, PatternApplicationGuidance, UseAffordabilityAndApparatusProportionality, and CorpusEntryProjectionAndEcologyFit when this evidence enters the pattern.
Repair by moving the evidence to the E.21 result, E.19 run record, README, ToC, E.11, card, retrieval, projection, or release or landing evidence locus, and keeping in the pattern only the user-facing move or boundary that follows from that evidence.
Quality table without rationale. A result gives values but no adjacent-value rationale. Recover or perform the judgement and supply ShortRationale under §4.3. Keep the evaluation in draft until the dependent value is established. Retain independently supported coordinate values. Lower the affected value only for an established weakness under its declared anchor; lowering an unsupported number does not supply the missing judgement.
Goodharted improvement. A rewrite improves source refs and proof sketches but becomes hard to use, or treats every non-5 coordinate as a defect to be fixed with more apparatus. Re-evaluate affordability, repair locality, proxy-for-value, and corpus ecology before stopping. When exceptional improvement is requested, look for a substantive improvement to the pattern. Additional proof alone does not provide it. The aggregate no-proposal disposition in E.21:4.7 needs loci showing that further content change is dominated, unavailable, or outside scope.
Bias-Annotation
E.21 resists Goodhart-style quality substitution: a high value is not produced by length, source count, approval state, checklist closure, or elegant phrasing when the required coordinate evidence is absent. It also blocks semio-bias by checking whether the evaluated pattern leads with its own EntityOfConcern and user-facing action rather than with description, publication, source, review, or repair apparatus.
Conformance checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
E.21 keeps the declared measurement structure simple: one checked-object class, one ordinal scale, one required coordinate set, one non-arithmetic PatternQualityQBundle result payload, one local status set, and one stop-condition form. All coordinates remain active. The evaluator applies them all, and the result episteme states what value the exact checked pattern edition and named evidence basis support under the declared use.
The mature-pattern parity coordinate tests whether formally clean wording also carries the worked slices, source carry-through, lowering conditions, and transfer coverage selected from mature FPF patterns for the declared use. Carry those selected ingredients in the body or the neighboring pattern that governs the claim; length alone establishes none of them.
SoTA-Echoing
The comparison follows E.8:11. Its practice question is how one complete, use-scoped pattern evaluation can expose semantic and practical defects, preserve distinct quality dimensions, and stop without turning a checklist or visible score into the value being sought. The selected answer is an FPF-local synthesis of four best-known branches.
No cited source validates E.21's coordinate set or demonstrates inter-evaluator agreement; the comparison below states the exact transfers and limits instead of converting publication status, prevalence, freshness, or academic praise into rank. An official source would be admissible here if its answer won the same substantive comparison, not because it was official.
Screening narrows where to look. The complete, use-scoped evaluation constitutes the E.21 result. Stronger validation claims require actual evidence suited to those claims. Currentness evidence keeps the comparison replayable and serves no stronger evidential role here. More current citations cannot compensate for a missing serious alternative, defect, or pattern mutation.
Relations
E.21:End
Improvement-Oriented Quality Evaluation Question Framing
Status: Stable.
Problem frame
Use E.22 when a request such as “review this” leaves it unclear what the recipient needs to learn about an object version. A floor check, a search for further improvement and a check of applied findings can require different answers about the same object.
First useful move: state what the recipient will use the answer for, identify the object version and choose the judgement that is still missing. Then name the criterion, the grounds the evaluator must examine and the answer to return. Section 4.0 develops this short construction; section 5 gives a filled request.
When those bindings are already settled and one direct evaluation is enough, run that evaluation. Rewriting its question as a separate frame adds no answer. Use E.23 when repeated improvement across passes is needed.
What can go wrong. A reviewer can answer a different question, count closed findings without judging the changed object, or propose improvements that damage a quality the recipient needs.
Practical gain. Requester and evaluator agree on the question before doing the evaluation. A floor question can stay small. A request for alternatives can name the additional trade-offs and proposals it needs.
Framing yields a question about the selected object version. The evaluator must obtain the quality result. The recipient must make any decision that uses it.
Problem
Quality evaluations fail when the evaluator has to infer the question. The same object can be checked for floor adequacy, evaluated for possible improvement toward exceptional expression, compared across trade-offs, examined for open questions, or evaluated after finding absorption. Those purposes produce different findings.
When a question is left unframed, the evaluator must guess which object is under improvement, which evaluation supplies its values and what kind of result that evaluation may return.
Forces
Solution
E.22 gives requester and evaluator a compact way to declare the quality question that will guide improvement. The evaluator answers the question. A decision or work product beyond the authority of that evaluation must be obtained separately.
Form a usable first request
Start from the result the recipient needs. The recipient may need a document to explain an operation, the operation to obtain a result, and the result to inform a further decision. Select the object and judgement needed now. A candidate's narrower description cannot establish that the larger receiving need has been satisfied.
- Choose the object and its extent from the intended use. Say who needs the answer and what work or decision it will inform. Which part, whole or connection could make the difference under evaluation? Follow that relation until you can identify the version to examine and the conditions a judgement about it must distinguish. An instruction, the Method it describes and the result of performing it can call for different evaluations. If the requested use is unclear, recover it with the recipient or the applicable problem-formulation method.
- Choose the question and criterion. Say whether the recipient needs floor adequacy, further improvement, a comparison of trade-offs or the effect of applied findings. Use the characteristic space and predicate or comparator supplied by an adequate evaluation. State the floor or improvement aim and the qualities a proposed repair must preserve.
- Connect the criterion to attainable grounds. Identify the content, observations or reasoning needed to answer it. Tracing a source contribution into an instruction can establish a coherent construction. Trying the instruction on material can reveal a missing action. An observed application can support a performance claim under the conditions observed. State which answer is required, what the intended reader or performer can supply, and what the candidate must supply. If indispensable grounds are absent, ask for the exact unresolved condition rather than an unsupported pass.
- Specify the useful return. Name the evaluation's result form and the conclusion it can support. A floor answer identifies sufficiency or the first needed repair; an improvement answer also describes the proposed change, expected gain, protected trade-offs and closure test. State evaluator conditions only when they change the question or admissibility of the answer.
When several contributions are needed, retain what their available evaluations establish and ask about the necessary connection still unanswered. A good explanation, a supported source claim and a correct constituent operation can all remain valid while their combined use is unestablished. Evaluate that connection for its own object and use. Combining unlike scores cannot establish quality for the whole.
Read the request in the words the evaluator will receive. If the requester supplies the criterion, a model or the relation between two contributions, what can the assigned work establish about using it? If the judgement concerns the practitioner's ability to choose or construct it, identify where the practitioner made that choice under the conditions being evaluated. An observation, the design itself or an earlier record can already establish it. Ask for the reasoning when the judgement requires it and the available work leaves it open. Where a narrower assessment supplies a premise, keep the wider evaluation question with the person responsible for its conclusion.
Write the request in ordinary language with these bindings recoverable, as in section 5. The local names and schemas below specify their exact meanings when the question is represented as a QualityEvaluationQuestionFrame and QualityEvaluationUseDeclaration. Reuse the selected evaluation's adequate descriptions and defaults. Add an intended evaluator's identity or a dated-Work account only when the receiving question requires it.
Stop framing when the question and expected answer are sufficiently determined for the intended evaluation. If an existing request already supplies them, perform the direct evaluation and inspect whether its grounds support the consuming conclusion. A separate frame is useful when it resolves a missing or disputed binding. Restating settled bindings adds no answer.
Local names and kind settlement
E.22 frames the question. The evaluator judges the object. evaluationPatternLocator identifies the FPF pattern description containing the evaluation predicate or constraint; an optional semanticEvaluationMethodRef names the separately identified U.Method used for that evaluation.
A characteristic-space specification, Q-Bundle description, rubric description, review-profile description, evidence-basis description, and result-form description constrain or describe that evaluation. The subject assertion and any semantic Method retain their independent bases.
For example, E.21, E.9.DA, or E.2.DA may supply the predicate for evaluating one FPF object, while A.19.ECS and C.25 supply supporting quality-model descriptions. E.19 supplies admission or refresh review questions and findings. Apply E.19 when that review is the requested result. When its findings support a quality evaluation, evaluationPatternLocator names the pattern that defines that evaluation, and the quality result cites the relied-on review findings as evidence.
A below-floor value, finding, improvement aim, or need for evaluation is not by itself an actual Problem. If the consuming use relies on an actual Problem, cite one current C.22.PFR ProblematicForRelation occurrence with its direct participants and temporal identity. The frame, evaluation, result and evidence may support a claim about that occurrence. None creates or splits it.
The framing episteme, evaluation method, descriptions used by that method, any question-changing evaluator condition, dated evaluation Work, actual operation application, evidence use, and result have distinct functions. QualityEvaluationUseDeclaration collects the applicable evaluation bindings. A plan, declaration or named candidate does not establish a current performer or occurrence.
The local support names ending in @Context serve only compatibility and retrieval. The suffix supplies no context entity, scope, participant, relation, or identity component. Every episteme follows C.2.1 identity, every set is identified by its stated extensional rule, and every neighboring Work, decision, evidence, viewpoint, grounding, or result relation remains under its direct governor.
In each filled episteme form below, claimGraph carries its complete interpreted claims and semantic relations by value, in the C.2.1:4.1.2 range NonemptyClaimContent. referenceScheme carries the effective rule collection by value in EpistemeInterpretationScheme; those rules must interpret the form's designations, predicates and qualifications. The remaining fields state the claims of that episteme. Range membership alone establishes neither their truth nor a completed evaluation.
Every field above with a *Ref suffix stores the stated A.6.5 RefKind; resolving it yields the referent kind named after referencing. The use declaration and expected evidence basis carry the same object version, governing evaluation pattern, selected characteristic space, criterion binding, ClaimScope, and qualification window.
Construct the expected basis from those values, the expected evidence positions and relation kinds, and the missingness rule. Then construct the declaration with a reference to the completed basis. The expected basis has no back-reference to the declaration. This keeps the construction acyclic.
At least one of selectedEvaluationPredicate and selectedComparatorSpecRef is present. Both may be present. A label such as review, quality, or current context supplies neither. A.19 defines the predicate by value. For comparator admission, use A.19.CPM or the rule of the consumer that directly admits the comparator. Identify any actual comparison application separately. Selecting the predicate or comparator does not by itself establish the evaluation scope, evidence, time, Work, or result.
evaluatorConditionRef states only a condition that changes the evaluation question or admissibility of its result. intendedEvaluatorSystemRef is present only when the declared question depends on that exact intended System. Neither field establishes assignment or performance.
The actual evaluator System, every obtaining assignment, and dated evaluation Work belong to the separately identified evaluation application or result account. Keep any local evaluator system-role classification separate and route unresolved role wording through E.10.ROLE.
evaluationPatternLocator locates the pattern that defines or constrains the evaluation; it is not the Method, performer, Work, or result. Claim Method or MethodDescription identity only after A.3.1 and A.3.2 admit it. Characteristic-space, Q-Bundle, rubric, profile, evidence-basis, and result-form references identify separate descriptions.
None of these declaration fields is dated evaluation Work or an evaluation result. A pre-evaluation frame does not assert that the future assessment has occurred. It may reference independently obtaining Work as the checked object, consumer or prior basis, retaining that Work’s admission and temporal conditions. An ordinary result that asserts no actual Work needs no Work identifier.
If a compact projection does assert dated evaluation Work, recover every exact actual performer through A.13 and follow A.15.1 for independent Work admission. The performer, Method, time, containing System, Work identity, and the result relation remain recoverable. Add assignment and F.6 refs only when the projection or receiving use expressly represents precise assignment-bound attribution. Missing or failed F.6 alone leaves independently admitted Work intact. Recheck any admission premise invalidated by the same finding and its dependent conclusions under A.15.1.
Keep any durable result episteme, evidence use, provenance, currentness, viewpoint, grounding, and Work-to-result or decision-use relation under their own patterns. A frame, declaration, description, assignment, dashboard, or carrier establishes none of them.
Two carriers may publish the same edition of either episteme. A QualityEvaluationUseDeclaration changes edition when its object version, claim graph, reference scheme, question-changing evaluator condition or intended-evaluator identity, evaluation pattern, semantic Method, selected characteristic space, predicate and comparator, ClaimScope, qualification window, quality-model descriptions, expected evidence-basis edition, or result-form description changes. Replacing one qualified actual evaluator with another does not change the declaration unless the declared condition or claim changes.
An ExpectedEvaluationEvidenceBasis@Context changes edition when its object version, claim graph, reference scheme, evaluation pattern, selected space, predicate and comparator, ClaimScope, expected evidence positions or relation kinds, missingness rule, or qualification window changes. Carrier, context label, viewpoint, grounding record, or support serialization alone changes neither episteme.
The two local set names above require only their declared element ranges and ordinary extensional equality. Permuting a list or repeating an entry leaves its represented set unchanged. A changed question-frame edition changes the admissibility or use question, not the set identity when its members remain the same. A changed proposal episteme contributes a different member under C.2.1, even if its short label is retained. An episteme may describe or publish a set. Its record and carrier have their own identities. These set-valued uses require no additional public set kind.
Quality evaluation purposes
A request can combine purposes. The result distinguishes the answer to each one. A floor result does not answer exceptional improvement. Absorption count does not establish a changed evaluation result. A proposal is not a selected work item.
Question frame
Use the construction in section 4.0 to settle the requested judgement before expressing these bindings. The schema preserves that question for its declared use.
An improvement aim is not a command to make every coordinate exceptional. Only the named evaluation can assign a 5, after the changed object earns it.
The frame may ask for substantive non-dominated proposals that could move named coordinates toward exceptional expression, while admitting no proposal or stay at current value when every plausible change would add apparatus, proof prose, boundary catalogues, or process evidence while damaging protected qualities. Support that no-proposal result with the review locations checked and references to its evidence basis.
The frame's exact object version, characteristic space, predicate/comparator binding, ClaimScope, and qualification window equal those of its use declaration and expected evidence basis. These bindings let another evaluator replay the question. The space, predicate, comparator, scope, method and consuming object retain their own identities. A changed binding creates a changed frame edition and requires a newly evaluated result.
resultConsumingUseRef is not a generic use placeholder. Before the intended work or decision occurs, the reference may resolve to one A.15.2 U.WorkPlan that names the intended Work, or to the decision question or decision-governing object under its directly governing pattern. It may resolve to U.Work only when that dated Work already obtains under A.15.1. The frame neither creates the consuming Work or decision nor authorizes it.
A minimal floor frame names the object version, one QualityEvaluationUseDeclaration, the exact selected characteristic space, applicable predicate and/or comparator, ClaimScope, result-consuming work or decision, purpose floorEvaluation, and the declared floor.
The declaration may cite defaults supplied by the governing evaluation pattern for its quality-model descriptions, evidence basis, result form, and qualification window, but defaults do not replace the exact selected space, criterion, scope, or consumer.
If the question depends on another edition, source state, comparison set, time window, or declared use, state that window explicitly. For one FPF pattern version under E.21, compactness never permits omitted coordinates, missing ShortRationale, absent PrecisionRestorationProfile, scope narrowing, or a blocker-only substitute result.
Connect the question, grounds and supported conclusion. State what the evaluation must establish about the object for the declared consuming use. In the expected evidence basis, identify the judgement that connects the selected criterion to the relevant content. For example, ask whether a Method's conditions permit the proposed action, whether a computation yields the stated result, or whether a source supports the claim made from it. A direct presence or identity check suffices when the question asks about that property.
When receiving the result, read its rationale and observations for that connection. Determine whether they support the requested conclusion under the declared criterion, scope and qualification. Retain supported findings. Identify which judgement or inference is missing before claiming more. Obtain that answer through the applicable evaluation, reusing current supported parts and keeping the answer in its existing result. Apply the same connection to an already-scoped direct evaluation. It needs no new QualityEvaluationQuestionFrame.
These instructions use the frame's existing claimGraph, the declaration's expectedEvaluationEvidenceBasisRef and the declared consuming use. The expected evidence basis describes what must be answered; the obtained grounds belong in the evaluation result.
For landing-input, corpus-facing, Stable, release, external-review or another stated use, evaluate the object under the scope for which the floor was requested. If another use is to be evaluated, open a new QualityEvaluationQuestionFrame. An easier scope cannot make the current request pass.
The frame and declaration state the evaluation's intended use. Naming an intended evaluator or planned condition establishes neither a current assignment nor Work. When dated evaluation Work is asserted, recover the exact evaluator through A.13 and use A.15.1 to admit the Work independently. Keep evidence use, typed result binding or direct result relation, and optional result episteme separate. Add F.6 only when the evaluation account expressly consumes precise assignment-bound attribution. An expected result-form description is not the result, and the consuming work or decision does not become current merely because the frame names it.
Finding and proposal rows
An actionable finding identifies where an issue was observed, which exact entity would change, the affected evaluation characteristic or coordinate, the current evaluation result for that characteristic or coordinate when known, the proposed correction, and the closure test. A proposal adds a typed expected evaluation effect, protected trade-offs, and any outside claim. For each outside claim, include the subject-pattern locator needed for an independent check.
Exactly one of proposedNextOperationDescriptionRef and proposedNextMethodRef is present. The question frame, proposal row, and follow-up hypothesis preserve the same object-version EntityOfConcern and ClaimScope unless a proposal explicitly opens a new frame for a different version or scope. QualityEvaluationQuestionFrame changes edition when the object version, use declaration, selected space, predicate/comparator binding, ClaimScope, consuming work or decision, purpose, floor or aim, trade-off set, qualification window, non-use boundary, claim graph, or reference scheme changes.
A proposal row changes edition when its frame, ClaimScope, correction target, affected evaluation coordinate, current result reference, proposed correction, expected effect, trade-offs, outside-claim nodes, closure test, claim graph, or reference scheme changes.
A follow-up hypothesis changes edition when its frame, ClaimScope, finding description, proposed operation or method, expected effect, test condition, claim graph, or reference scheme changes. Changing only a context label, carrier, viewpoint, grounding record or serialization leaves all three epistemes unchanged.
ProposalEvaluationEffectValue is the closed local value set repairFloor | raiseTowardExceptional | preventProtectedQualityLoss | classifyOutsideEvaluation | preserveCurrentValue. It identifies the broad type of substantive effect the proposal is expected to have on the evaluation. E.23 gives a separate coordinate-qualified prediction in ExpectedEvaluationResultChange@Context. Neither prediction asserts that the result has changed.
ProposalKindRestorationCheckDispositionValue is triggered | notTriggered | ordinaryProse | alreadySatisfied | blocker. The triggered and blocker states include kindRestorationCheckRef; the other values leave it absent.
Current affected-evaluation result ref and kind are both present or both absent. When present, the exact result resolves through the direct evaluation pattern's typed result relation or A.6.1 application binding, and any durable result episteme remains separately governed. The proposal row neither produces nor reidentifies that result. The exact kind recovers whether the named evaluation returned a scale value, status, or another admitted result for that characteristic or coordinate.
Outside value ref and kind are paired. When the outside value is a relation, recover its direct predicate, participants and applicability from the subject pattern. Add outsideRelationSignatureRef only when the proposal uses that reusable declaration. The relation occurrence and declaration remain distinct under A.6.REL.
CandidateImprovementOutsideClaimReference@Context is a bounded local ClaimGraph node form, not a U-kind, episteme, relation, or relation-reference episteme. It is constructed inside one proposal row without a back-reference to that row. Its node identity is determined by the containing proposal edition and ClaimGraph position.
reviewLocationDescriptionRef describes where the issue was observed in the reviewed object. correctionTargetRef identifies the exact entity that would change. Keep the observed location distinct from the correction target even when both concern the same object. The row is a faithful typed proposal form of QualityReviewFindingRow and one possible member of a CandidateImprovementProposalPortfolio@Context set. It describes a proposal. Selection, planning, performed Work, actual Transformation, result binding and proof of improvement each need their own grounds.
For wording, naming, and precision-restoration proposals, proposedCorrectionDescriptionRef states the correction and its intended content effect. Apply F.19 for the ordinary repair and local revalidation. When the changed FPF-governed expression can alter meaning, KindRestorationCheck states the live object, kind, relation, slot or use position, claim kind, admissible use, and scope before and after the change. If the proposed repair cannot preserve those values and no accepted decision justifies changing them, the proposal row remains blocking.
Absorption impact values
In the absorption result, state how the object’s quality changed or why it remained the same, using the evaluation of that object. Counting accepted rows does not establish that result.
OEE and NQD proposal portfolios
When the object is a candidate, archive or front member, selected set, parity report, refresh report, or declared transformation result, use E.22 to frame the quality question and return proposal rows. Use C.17 for candidate characteristics, C.18 for archive and front relations, C.19 for pool policy, G.5 for selected-set result declaration, G.9 for parity, and G.11 for currentness and refresh. When the question concerns availability to an audience, use E.17 for the source-backed publication face and return to source. Use E.24.PUB for the publication occurrence, form, carrier, audience, bounded use and availability.
Worked slices
Floor evaluation: from “review the studio method” to a usable request. A studio practitioner needs to decide whether to use C.39 to develop the recording-preparation method in C.39:5.3. “Review the studio method” leaves three possible objects: the pattern's explanation, the conditional preparation it describes, and an actually prepared recording. Here the practitioner needs the first judgement: whether the complete pattern provides usable development guidance. Let M1 name the immutable complete C.39 edition supplied with this illustrative request.
Evaluate the attached C.39 edition M1 for my decision whether to use its guidance to develop the studio preparation in C.39:5.3. The intended user is a practitioner familiar with FPF and studio editing who can obtain the recording engineer's interval qualification. The pattern must explain how to construct and connect the needed operations, identify missing prerequisites and reconsider affected uses after a change.
Use the E.21 nineteen-coordinate characteristic space and its floor predicate: every required coordinate at least 4 for this declared use. Evaluate the complete pattern. Examine its instructions, conditions, worked reasoning and necessary direct suppliers; use the studio case to trace what the reader can actually obtain. Preserve the distinction between the engineer's premises and the pattern's contribution. A worked conditional calculation supports a claim about the explained construction; it does not establish performed editing, practitioner learning or a better movement judgement.
Return the complete E.21 result: coordinate values with adjacent-value rationales, the precision-restoration profile, evidence basis and supported status. For a failed coordinate, identify the smallest sufficient repair and how to check it. Protect the useful source contribution and independently valid viewing use. This request concerns M1, the stated reader and studio use with the case's supplied premises; it does not ask for a general release decision or exceptional-improvement proposals.
Once the requester and evaluator agree on this question, the evaluation can begin. Its purpose is floorEvaluation; the complete E.21 result remains to be obtained.
Suppose the evaluator observes that M1 explains the justified interval [40,45] ms, qualified editing and the 42.5 ms correction with a 2.5 ms bound. With that qualified interval, a constant correction of 42.5 ms meets the 3 ms analysis and 8 ms viewing requirements. To judge M1 as guidance for developing the preparation, trace where a practitioner starting with those requirements can decide what to ask the engineer, choose a correction and establish its consequences for both uses. The worked arithmetic supplies one part of that judgement. Evaluate each remaining E.21 coordinate on its own grounds. If the interval qualification is unavailable in the receiving studio, obtain that missing basis. Three marker readings alone do not justify the edit.
Tightening the analysis requirement to 2 ms changes the preparation's applicability: no constant correction can meet it for the given 5 ms offset span. The pattern's instruction to return that dependent use for development remains useful, and the unchanged 8 ms viewing use remains qualified. A request to assess an actually edited recording would need that recording version and evidence about it. Where the receiving question is already the limited calculation under the supplied interval, evaluate that calculation directly. A full E.21 pattern evaluation and another frame add no answer to that calculation.
A method recovered from a source. A recipient needs a usable way to examine a proposal and decide how to continue. A draft faithfully summarizes a source's discussion of criticism and correctly names proposal, evidence and action, but never explains how the critique changes the next choice. A text-quality result can support the summary's fidelity; it cannot establish the missing explanation of how to obtain the result. The frame asks whether the candidate gives the intended reader that connection under the stated preparation and available grounds. C.39:5.4 develops this use.
An adequate answer may instead return a proposal worth testing, with the exact condition that prevents immediate enactment and a justified attainable next inquiry. Whether that is sufficient depends on the recipient's requested result: it can identify the next inquiry and leave a request for completed enactment open. The evaluation must not reject every conditional answer as defective or silently lower the receiving requirement.
Deleting the whole source proposal over one unsupported inference can remove its useful propose–criticize–revise construction. Ask the evaluation to inspect both the repaired connection and the contribution it must preserve.
When the ground for one later application changes, retain the source-fidelity judgement and any unaffected use, and reopen the dependent application claim. If a proposed change to the evaluation itself was developed on these cases, use A.19.ECS:4.5 to compare it with adequate execution of the incumbent evaluation and examine an independently grounded new case before claiming broader improvement.
Exceptional improvement. A pattern already passes the floor. The frame asks for substantive non-dominated improvements for named coordinates while protecting usability and related-pattern fit. The result gives proposal rows for content improvements such as missing worked cases, source-currentness carry-through, mature-comparator discharge, deletion of displaced apparatus, or relation cleanup, and gives checked no-candidate dispositions for coordinates where no non-dominated content move remains. A 5 on every coordinate is not the aim.
Absorption. External review returns many suggestions. The frame asks for absorptionEvaluation. The result says which changes improved coordinates, which were already satisfied, which introduced trade-offs, and which belong outside the evaluation.
Proposal portfolio. A candidate improvement campaign needs alternatives before editing. The frame asks for candidateImprovementProposalEvaluation. The result returns bounded proposal rows. For example, a proposal to reduce latency and a proposal to reduce energy consumption remain two alternatives with their own trade-offs. Listing the first twice or swapping their display order leaves the proposal set unchanged. A separate protection set can keep both latency and energy coordinates visible. Selection or generation uses its own governing pattern.
Physical-system proposal. A vibration evaluation of PumpAssembly@Prototype-3 selects the vibration CharacteristicSpace, RMS-vibration predicate and any comparator, one evaluation ClaimScope over the declared operating-point slices, and the design decision that will consume the result.
This case asserts dated test-bench evaluation Work. First recover the exact evaluator through A.13. A.15.1 independently identifies the Work, Method, time and containing System. If this test-bench account also needs the exact assignment under which the evaluator acted, add F.6 through the same obtaining A.13 assignment. Missing or failed F.6 alone leaves independently admitted Work and its independently grounded result route intact.
An exact application of the declared vibration-evaluation operation returns the finding of excessive RMS vibration at one operating point under its result declaration. The result binding connects that application to the finding. The frame, subject-pattern reference, assignment, expected evidence basis, and result-form description remain separate.
The proposal's reviewLocationDescriptionRef points to that evaluation row. Its correctionTargetRef points to ImpellerBladeGeometryDescription@v3, the design episteme that would change. The measurement row is the finding location. The affected coordinate is RMS vibration. The coarse proposal effect is repairFloor because the proposal addresses excessive vibration. kindRestorationCheckDisposition=notTriggered; the protected trade-offs are efficiency and manufacturability.
If the proposal is selected for a repeated loop, E.23 adds a scale-qualified ExpectedEvaluationResultChange@Context. Manufacturing a new impeller remains dated Work under A.15.1 rather than an E.22 result.
Bias annotation
E.22 favors naming the intended quality question and what would answer it. Without that information, reviewers often give a plausible answer to the wrong question.
The quality values, repeated improvement, publication of a selected set, decision about work and certification of a project claim each require the operation that produces them.
Conformance checklist
Common anti-patterns and repairs
Consequences
Rationale
A generic request leaves the evaluator to choose what counts as quality. To show which question a quality result answers, bind its object version, selected characteristic space, predicate and any comparator, one evaluation ClaimScope, consuming work or decision, evaluation pattern, any separately identified semantic Method, purpose, expected evidence basis, expected result form, and boundary.
When needed, the question frame also states a question-changing evaluator condition or intended-evaluator identity. The pattern, Method, assignment, descriptions, dated evaluation Work, evidence use, result, decision and project authority each retain their own grounds.
SoTA-Echoing
Choosing the lightest sufficient question construction
Practice question. How should a requester obtain an evaluation question that will support the recipient's actual next use, without adding another description when direct evaluation already suffices? Retain an adequate direct question. If it is inadequate, recover its missing purpose, object, criterion, grounds or use. Express exact E.22 bindings when they are needed to preserve that question across evaluators and consuming uses.
Two serious alternatives already provide purpose-first construction. Basili et al., Linking Software Development and Business Strategy Through Measurement (2010), especially “Background” and “GQM+Strategies”, links goals, derived measurement questions and interpretation to context and assumptions. This historical method contributes an organizational alignment model when the question requires that larger linkage.
Kidder et al., CDC Program Evaluation Framework, 2024, Steps 3–5, develops evaluation purpose, intended users and use, questions, design, credible evidence and supported conclusions. It expressly accommodates available resources and revision when needed evidence is infeasible. Their software-organization and program-evaluation settings supply no FPF characteristic values.
Comparison on the same receiving question. Keep the complete M1 edition, reader, studio use, available premises and E.21 criteria from section 5 fixed. The following is a worked adaptation of the methods to that question, not a measurement of practitioner time or error rates.
Choice and accepted trade-off. Adopt the purpose-to-question and use-to-evidence direction shared by the alternatives; adapt it to the FPF object, criterion and result distinctions. Sections 1 and 4.0 put that construction before schemas, section 4.3 retains the direct-evaluation exit and section 5 supplies the filled request and its changed-use boundary.
The local addition is an explicit account of those bindings, useful when exchanging a question otherwise risks evaluating another object or carrying a result beyond its scope. Take on the cost of describing and maintaining those bindings only when that precision is needed. When an adequate direct request or the rival's question already preserves the same distinctions, reuse it. Neither a newer explanation nor the presence of a frame establishes better detection, lower cost or a reason to replace a sufficient evaluation.
The comparison selects an FPF adaptation for this question. It leaves the relative merits of GQM, CDC and FPF as whole frameworks open. Reopen when a competing construction preserves the needed question at lower total burden, when an intended reader cannot recover the first request, when a changed source or use defeats these transfers, or when exact binding introduces more maintenance than its receiving use warrants.
Conditions that change the evaluation question
Use the following contributions when the question calls for actionable feedback, protection of trade-offs or evidence about the reliability of an automated judge. Each constrains the question under its own condition.
Relations
E.22:End
Quality Improvement Loop Method
Type: Method-description pattern Status: Stable Normativity: Normative unless marked informative
Problem frame
Use E.23 when an object version will be improved through repeated passes under the evaluation declared for that object. The object can be a pattern, DRR, FPF corpus object, engineering quality object, naming candidate, OEE and NQD candidate, archive or front member, selected set, parity report, refresh report, or declared transformation result, if the evaluation supplies values and stop meanings for that object kind.
First useful move. Start with the difference a recipient needs from the improvement. Locate where that difference could arise, choose the object version to change and decide how to recognize an improvement in the receiving work. Name the evaluation to use again, the protected qualities, cost and risk, and the conditions for stopping.
For improvement of an operating work process whose useful next condition is clearer than the way to reach it, §4.4.1 develops the Improvement Kata branch: study current work, establish a near target condition, and use observations and bounded experiments to discover an effective change. Preparing that basis can finish with a useful next observation. An actual improvement pass needs the object, evaluation, protected qualities and continuation grounds above. Use the coached-practice application in HCD.9:4.10 when developing the learner's or coach's reasoning is also intended.
Not this pattern when one direct quality evaluation is enough. Run that evaluation; use E.22 first only when its question needs clarification. Use A.19.ECS first if the needed evaluation characteristic space does not exist.
When the intended improvement concerns how further improvements are obtained, name that way and the means needed to apply it as the object. Use C.40:4.7 and C.40:4.11 to construct alternatives and compare their subsequent development work; §5.1 shows the connection for people, organizations and AI. Changing an instruction, using the changed way in the next improvement, and obtaining better further development are different results. Keep an adequate improving method when changing it offers no worthwhile gain.
When the evaluation could report improvement without the claimed change, §4.1.1 constructs that absence through the same evaluation. An already adequate comparison needs no additional control merely because another improvement pass occurs.
Use E.23.CAE first when the apparent object to improve remains ambiguous because a holder's capability may be outside its claimed envelope, unavailable through the current configuration, not selected as applicable, inaccessible or inactive, context-dependently unexpressed, unadapted, unenacted, or actually changed. That separate probe returns observations, a qualified disposition, surviving rivals, and candidate routes; it does not choose the object under improvement.
Use E.23.CDI instead when a separate applicable steering or choice result has made capability development current for one admitted holder System and named Work family, and the change must be checked in representative Work. That is a separate capability-development Method; E.23 points to its description without copying its actions. Use a population assessment when the question is a distribution across member capabilities, C.36 when generation, transmission, recognition, selection, retention, or loss across a cultural population is current, and C.32.MWA first only when the target-practice architecture itself must be recovered or compared.
What goes wrong if missed: teams close discharge rows instead of improving quality, retry blindly, optimize visible values while damaging protected qualities, stop merely because every value is 5, or treat a review recommendation as a decision, work, evidence, a selected-set result declaration, actual publication, parity, or refresh.
What this buys in practice: each pass has a declared object version, an intended evaluation-result change, a rerunnable evaluation, protected trade-offs, and a stop or switch condition. The team can then spend its effort on improving the object’s quality instead of accumulating review records, and stop when no non-dominated change is worth its cost.
Primary EntityOfConcern in plain terms: the repeated quality-improvement method for one object version under one declared evaluation.
Problem
FPF often improves artifacts by repeated review, repair, and re-evaluation. The loop is useful only when the changed object is evaluated again under comparable conditions, or a change of evaluation is justified for its receiving use through §4.1a. Without that discipline, repeated passes become checklist closure, agentic retry, source citation, or process state.
The loop also avoids the maturity-ladder trap. Meeting the floor or receiving all 5s supplies quality evidence within the current scope. Stopping additionally requires the practical judgement in §4.3; neither result proves that the object cannot improve.
The loop also fails when an ordinal value becomes a work target. 5 is an assigned result after measurement, not an instruction to add apparatus until a 5 can be defended.
Below-floor values return a repair proposal or intended-work claim; they do not establish that Work occurred.
Above-floor improvement becomes a selected proposal when chosen under the applicable selection rule, but the target is a substantive content improvement: stronger positive action guidance, worked slice, case and countercase coverage, source-currentness carry-through, mature-content discharge, relation cleanup, deletion of displaced apparatus, split of overloaded content, or another named content gain. Stay at 4 or no proposal is admissible only after a by-value search finds no non-dominated content improvement worth its cost under protected qualities. Selection does not perform the Work or bring about the Transformation. Each occurrence needs its own grounds.
A below-floor value, finding, improvement aim, or repeated-evaluation need is not by itself an actual Problem. If one improvement use relies on an actual Problem, cite one current C.22.PFR ProblematicForRelation occurrence with its actual-condition and criterion-applicability participants and its maximal continuous adverse-episode identity. Evaluation Work, result epistemes, evidence, and loop records may support a claim about that occurrence; they neither create nor split it.
Forces
Solution
E.23 guides repeated improvement of one object version under a settled evaluation question. Reuse an adequate ordinary request; use the E.22 QualityEvaluationQuestionFrame and QualityEvaluationUseDeclaration when a receiving use needs those declarations. The evaluation pattern defines the evaluation; restoration and subject patterns supply the relevant rules or guidance. A pattern body or locator does not by itself establish a semantic U.Method or qualify as its description. Claim Method or MethodDescription identity only after A.3.1 and A.3.2 admit it.
When an evaluation or improvement pass is claimed as actual A.15.1 U.Work, identify every actual performer through A.13 and independently identify the occurrence, time, Method, and containing System through A.15.1. Add A.2.1 assignment-occurrence identity and F.6 only when the loop record or receiving use expressly represents precise assignment-bound attribution through the same obtaining A.13 assignment; missing or failed attribution alone leaves independent Work admission intact.
If the same finding invalidates a premise consumed by A.13 or A.15.1, apply A.15.1's premise-recheck rule to that premise and its dependent conclusions, preserving independently grounded history and relations. A proposed pass, intended performer, planned condition, or A.15.2 WorkPlan remains modal until its predicates obtain.
Keep a returned value, durable result episteme, changed object and actual Transformation separate. Connect a returned value through its A.6.1 binding or the evaluation’s direct result relation. Connect Work to a result or change only through a declared direct relation or supported local claim that obtains.
Return missing-governor when the required predicate or governing condition is unavailable. If the governor exists, apply its criterion to the case: retain a supported positive, negative or inapplicable result, or name the missing facts and how they can be obtained.
The repeated organization changes the object, re-evaluates the changed version through the same declared method and quality model, checks trade-offs and cost, and identifies the admissible continuations: stop, continue, switch methods, open a new frame, wait for sufficient information, branch, or return to a subject pattern.
When the receiving use needs its structure, admit that organization under A.22.CGUS through §4.2a; use E.18 only when an independently selected transformation-flow structure is the EntityOfConcern. Neither the method, record, visible cycle, nor selected continuation is an enduring Work occurrence or context container.
Ordinary loop method
For one quality-improvement loop:
-
Choose the object from the intended improvement. If the same component could serve the user well in one arrangement and poorly in another, examine the way the components work together. A changed instruction can improve what a reader is taught to do. A claim about what a team actually delivers also depends on its working conditions and performance. Select the extent and evaluation that can distinguish the changes at issue. For a request confined to the instruction, its improved guidance can be the full result. Reuse the current
QualityEvaluationQuestionFrameandQualityEvaluationUseDeclarationwhen they fit. Keep the evaluator, evaluation Method or pattern, characteristic space, evidence basis, result form, ClaimScope and qualification window distinct. -
State the content change sought, protected trade-offs, cost and risk account, and local stop condition. Do not use
5, all-5, or5-defensibleas the target; say what should become better in practice. -
Resolve any remaining ambiguity in purpose, scope or result-consuming work or decision through E.22. Keep an adequate question; constitute a separate E.22 frame only when its declaration is needed.
-
Run the declared evaluation. When E.21 is the selected evaluation, retain its complete result: every coordinate,
ShortRationale,PrecisionRestorationProfile, evidence basis, coordinate payload, and status. A loop note, blocker summary, or "no blockers" statement is not a substitute. If dated evaluation Work is asserted, identify it and its result binding or direct result relation; keep any durable result episteme separate. -
Record each returned finding or proposal separately. A grouped memory summary does not close skipped items, and a proposal remains a proposed next action rather than performed Work.
-
Choose the change that could improve the commissioned result. If this pass calls for implementation, carry it out before evaluating the changed object. When different contributions must work together, change the relevant connection or constituent and keep the conditions of their joint use available for the next evaluation. When the account asserts dated improvement Work, identify that Work and connect it to a returned value or changed object only through an obtaining A.6.1 binding or declared Work-to-result or Work-to-change relation. If that relation is unavailable, keep proposal, Work, changed object, and Transformation separate and return the missing relation.
Repair below-floor findings first. Above the floor, prefer a substantive gain—such as clearer action, a missing case or countercase, current source support, restored predecessor content, cleaner relations, or a split of overloaded material. Do not add guards, catalogues, or quality proof merely to defend a higher score. Close with no change only after the evidence shows that no feasible non-dominated improvement remains worth its cost under the protected trade-offs.
For a precision-restoration defect, apply
F.19and open a restoration or subject pattern for an unresolved FPF-specific meaning. Claim a Method or MethodDescription only when A.3.1 and A.3.2 admit it. Keep locator, Method, description, performer, assignment, Work, result, and responsibility separate. Run one boundedKindRestorationCheckwhen the changed expression can alter FPF-governed meaning; otherwise F.19's local revalidation completes the ordinary repair. -
Re-evaluate the changed object as a separate pass through the same declared evaluation with comparable evidence and conditions. If the evaluation itself needs to change, use §4.1a before comparing values across that change. Keep the later Work, application or direct result relation, evidence, returned value, and result episteme distinct from the first pass.
-
Record what improved, what stayed at the floor, what was unchanged by value, what became worse, and which findings moved outside this evaluation. Compare the earlier and later results at the scale of the intended gain. Use a result about one component for the whole when its contribution and the conditions governing the remaining parts support that inference. When result epistemes or dated Work are asserted, preserve their separately governed identities.
-
Decide
stop,continue,switchMethodFamily,openNewFrame, orholdUntilInformationBasisSufficient. When later replay depends on alternatives and guards, use the conditional structure block below. A decision or selected continuation neither authorizes nor performs the next Work. -
Leave an account that lets the next reader recover the object versions, evaluation, proposals, performed passes, result or change bases, evidence, trade-offs, cost and risk, continuation, stop and return boundaries, and the reason for the decision. Use the structured
QualityImprovementLoopRecordonly when a named replay, handoff, audit, or machine-facing use needs that form; otherwise a short result with the same recoverable facts is enough.
Stop here when this route answers the current use. Open the names, record schemas, and unfolding-structure block below only when a named receiving use depends on that added assurance detail.
When evaluation can imitate improvement
Use this branch when sampling, repeated assessment, selection or the calculation of a reported result could produce the apparent gain while the claimed change is absent. The practical question is not just how variable the score is, but what this procedure would report without that change. Keep an existing comparison when it already answers that question adequately.
Name the claim and the reported quantity. Better average performance, newly successful cases, fewer losses and faster subsequent learning are different quantities. State the object, cases, conditions and change that the comparison concerns. A control for one quantity needs a separate basis to support a claim about another: a stable average does not prevent cases from moving in both directions between success and failure.
Construct the absence of the claimed change. Use an applicable model, a matched control, or both. For an unchanged stochastic solver, the model can describe repeated attempts under unchanged conditions. For a grading procedure, reassess the same archived answers without revealing their earlier grades. Say what remains fixed and what can vary. If a changed task population, assistance or operating condition can explain the difference, include it in the comparison or narrow the claim. Use C.16.MR to construct the measurement relation, C.16.IR to determine what it can resolve, and C.16.RM when the arrangement needs repair; the relevant mathematical or empirical method supplies the particular calculation or study.
Carry that absence through the whole procedure. Apply the same inclusion rule, thresholds, selection of a best attempt or version, recording, calculation and stopping rule used for the reported gain. If only initially unsuccessful cases are counted, reproduce that conditioning too. Preserve shared cases and other dependence: all ordered pairs of a few repeated runs are not that many independent experiments. An idealized model is sufficient only where its assumptions support the intended use; an empirical control can expose effects omitted from it, such as altered batching or assessor behavior.
Use the resulting behavior to decide the stated question. Obtain the relevant distribution or control estimate and its uncertainty, then choose a comparison rule for the intended claim and tolerated error. A mean control value is not itself a threshold, and subtracting it is not a universal correction. Check a rule tuned on control results on material not used for that tuning. Carry the uncertainty of the control into the final comparison. If the arrangement cannot distinguish a worthwhile change from these effects, obtain more informative evidence, improve the arrangement or leave the claim unresolved. Failure to detect a difference does not establish equivalence or absence of improvement. A difference beyond the control still needs any additional evidence required by a causal or wider capability claim.
Worked calculation. Suppose an unchanged solver has probability 1/2 of success on each of 20 cases, with independent attempts and independent cases. In two evaluations, failure followed by success has probability (1/2) × (1/2) = 1/4; success followed by failure has the same probability. The expected counts are five new successes and five losses, while expected net change is zero.
The new-success count has distribution Binomial(20, 1/4): the probability of eight or more is about 0.1018, and of nine or more about 0.04093. For a single comparison whose rule was fixed in advance, nine would therefore meet a chosen 5% upper-tail rule under this model; eight would not.
Choosing the best of several comparisons changes that calculation. Neither threshold establishes a cause, a general capability or improvement on a different task population. These are calculated properties of the constructed example, not intervention data.
Human and organizational limits. Regrading unchanged answers tests variation in assessment, not whether their author learned. Reprocessing unchanged service records tests the reporting procedure, not whether the service improved. A person can retain practice, fatigue or changed expectations, and an organization retains its history even if a document is restored. Construct a comparison appropriate to the claimed change in their work and support; do not manufacture an unchanged person or organization by copying an instruction. If only the reporting artifact can be held fixed, preserve that useful bounded result and leave the stronger claim open.
When the evaluation needs to change
Use this branch when a missed defect, false objection, harmful proposed repair, changed use or excessive evaluation cost gives reason to reconsider the evaluation. An ordinary object repair can keep using the existing adequate evaluation.
First identify what changed. A new object version, a new requirement, a different task population, a different measuring procedure and a different evaluator implementation can require different responses. Fixing an implementation error can preserve the intended evaluation while invalidating the results affected by that error. A threshold that varies with load under a fixed rule still uses that rule; it does not by itself change the evaluation.
Keep the evaluation meanings and comparison conditions stable within one comparison. To change them between cycles, use A.19.ECS:4.5: compare the old and proposed evaluations on independently grounded defect and admissible cases, inspect harmful repairs and full cost, and use a new independent case when known cases were used for tuning. Decide whether to adopt the change for a stated use, retain it on trial, revise it, reject it or keep the current evaluation. A stricter result or higher evaluator agreement is not the adoption criterion.
After adoption, name which earlier conclusions depend on the changed rule. Re-evaluate affected preserved cases where cross-evaluation comparison is needed; otherwise retain earlier results with their original scope. Declare a bridge when it is justified, and leave unlike values incomparable when it is not. New task populations and requirements create new questions without retroactively falsifying supported old answers.
Recover how the evidence was obtained. The old criterion may have decided which variants to generate, which cases to examine, which branches to continue and which observations to retain, as well as how to score an answer. Changing a score does not undo these earlier choices. Start from the receiving decision under the adopted evaluation and trace the evidence it uses.
For each affected result, recover the object or response, case and conditions, observation or raw material, rule used for its assessment, and any old-criterion dependence in its generation, selection or retention. Use existing provenance when sufficient; add only the distinctions needed to settle the present use.
Distinguish three kinds of repair. When the same intended evaluation had an implementation error and the saved observations contain the required information, correct the affected calculation and its dependents.
When a changed evaluation can be applied to a preserved object, obtain the new judgement and retain the old one under its original question; evaluating an old artifact now is not rerunning its former performance.
When the new question needs outcomes never observed, a different population or consequential conditions absent from the record, rescoring cannot provide them. Obtain suitable new evidence, use a justified inference under the relevant subject Method, or restrict the conclusion. A large archive can still lack precisely the needed observation.
Rebuild the basis that directs the next choice. Follow the affected results into the summaries and estimates used to choose further work: success proportions, branch or method rankings, accumulated utility, fitted predictors, uncertainty estimates and resource priorities. Suspend their affected use until their basis has been repaired. Do not mix old- and new-criterion values in one current estimate just because their field names agree. Recompute the needed dependents from still-applicable evidence and newly obtained judgements, keeping population, denominator, conditions and uncertainty meaningful. A narrow correction requires only the dependent part; independent observations and unrelated conclusions remain usable.
Keep raw artifacts, observations and earlier qualified results recoverable. Withdrawing an old contribution from a current utility estimate is not deleting the historical event or declaring it false. Where a record contains both an unaffected observation and a criterion-dependent judgement, separate their uses instead of discarding both. Conversely, retaining an artifact does not qualify an old ranking built from it. If dependency cannot be recovered, mark that basis unresolved for the affected decision rather than assuming independence.
The repaired result should let the practitioner make the next warranted choice: continue with a valid comparison, test a promising alternative under the new conditions, retain a limited old answer, or return an explicit unresolved question. Include the work of obtaining new evidence and rebuilding estimates in the decision. No automatic instruction to reevaluate the whole archive follows. C.40:4.8.1 supplies a bounded history replay when its recorded responses and conditions suffice; changing the evaluator may require repairing both those responses and the selection mechanism that chose them.
Worked selection change. An organization received ten project proposals. A low-upfront-cost criterion admitted six to pilots; four were never piloted. Complete pilot records permit a later, adopted reliability evaluation of the six: three satisfy it. Those records support the descriptive result “three of the six selected pilots meet the new criterion.” They do not support a reliability rate for all ten proposals, and the four absent outcomes are not four failures. The old cost criterion affected the population observed, not only the score assigned afterward.
The organization keeps all proposals and pilot observations, preserves the earlier cost judgements under their old use, and replaces the affected reliability judgements for the six. It withdraws an old all-project promise or resource priority if that claim depended on treating the selected pilots as representative. It then rebuilds the comparison needed for the next allocation.
If the four untried designs are serious alternatives, an appropriate new pilot or justified subject model supplies their missing consequences; if they are irrelevant to the present decision, it can retain a narrower answer without testing them all. If the new requirement concerns a longer service period than any pilot recorded, even the six lack the required observation. A revised score cannot extend observation time.
Human self-development. A person formerly chose familiar exercises because the self-assessment rewarded fluent completion. A new question asks whether the acquired way works on unfamiliar problems. Saved answers can reveal previously missed reasoning defects, but awarding them new marks supplies no performance on the excluded problem classes. Preserve what those answers show, retire a transfer claim based only on the familiar exercises, and obtain suitable new cases. Update the practice-selection rule that used the old success estimate. Neither erasing the person's history nor averaging old fluency marks with new transfer results answers that question; the subsequent practice and its actual use remain separate from changing the evaluation.
Worked slice. A review evaluation rewards concise explanations but misses an omitted dependency that makes the instruction unusable. Adding the missing dependency improves the object under its existing usefulness requirement. Changing the evaluator so it notices that defect is a different proposal. Compare that evaluator with its predecessor on the defective text and a concise, fully usable alternative; reject a repair that makes every explanation longer without need. A new independently grounded case supports adoption beyond those development cases. Previously supported readability judgements remain scoped to their text and reading conditions; only conclusions depending on the missed dependency need reopening.
Conditional names and kind settlement
Source and practitioner phrases such as "loop engineering", "agent loop", "harness loop", "prompt loop", and "workflow hardening loop" are entry phrases. Lower them into ObjectUnderImprovementRef, QualityEvaluationQuestionFrame, QualityEvaluationUseDeclaration, ImprovementAim, MethodFamilySelection, CostAndRiskAccount, and QualityImprovementLoopRecord, or else name the subject pattern for the live claim and leave E.23 closed.
Quick lowering map:
The next table names the local values used after this routing choice.
The claimGraph and referenceScheme positions in the episteme forms below carry, by value, the complete interpreted content and effective rule collection of that episteme. Their C.2.1:4.1.2 ranges are NonemptyClaimContent and EpistemeInterpretationScheme. The rules must interpret that form's version designations, predicates and qualifications, including the evaluation scale used by a prediction. Interpreted negative claims or predictions whose truth is unresolved can satisfy the content range; their evaluation and warranted use remain separate questions.
The two named forms organize claims within the claim content of a result or loop episteme; a table row or serialization may publish them but does not become the claim.
A SourceComposedResultClaim uses exactly one of resultClaimNodeRef and resultClaimEpistemeRef. resultClaimNodeRef is the C.2.1:4.2.5 ClaimAddress pair: an exact episteme-edition reference and an intrinsic claim identity declared by that edition's ClaimGraph. Resolve the edition and then require exactly one claim with that identity. An unresolved edition, absent or non-unique identity, or display-only node token leaves the address unresolved.
An edition change requires an explicit claim correspondence for any address migration. When no stable intrinsic claim identity exists, use resultClaimEpistemeRef if the receiving use can take the whole episteme's content; otherwise constitute the required result claim as its own C.2.1 episteme and reference that whole episteme. A row number or rendering label does not supply the missing claim identity.
Checked evidence values stay paired with their kinds; evidence relations form a separate pair. Every actual Work reference points to an independently identified occurrence governed by the central §4 Work rule. A compact rendering may use only the omission allowed there.
For an improvement pass, the changed-version reference appears only after that version exists. The workResultOrChange... fields keep the declared predicate, its pattern locator, and the obtaining basis distinct. That basis must be an A.6.1 result binding, a direct Work-to-result or Work-to-change relation, or a local relation claim that names its participants, conditions, and obtaining facts.
If only the predicate or locator is known, keep the proposal, Work, changed object and Transformation separate. Recover the predicate’s applicable test and the facts it needs. An unavailable definition or condition is missing-governor; when occurrence evidence is unavailable, leave the relation unestablished and identify the missing basis. A supported negative or inapplicable result retains its own meaning. An A.15.PROD route points to its applicable local claim rather than to the pattern as a generic claim.
The two record epistemes follow C.2.1 identity: claim content, exact EntityOfConcern, and effective reference scheme determine each episteme edition. The listed loop fields contribute to claim content; editionId designates an already distinguished edition but does not constitute it. Empirical grounding, viewpoint membership, claim scope, model-use structure, applicability, qualification, evidence currentness, and source currentness remain separate relations or values defined elsewhere. A change in one of them changes a record episteme only when its claim content, EntityOfConcern, or reference scheme is revised; carrier and support serialization alone change neither episteme.
These records do not create quality values, project evidence, release state, selected-set result declaration, actual publication, parity, refresh, Work, Transformation, or proof of quality.
The @Context suffixes on support species such as LoopEvaluationEvidenceBasis@Context, CandidateImprovementProposalRow@Context, TradeoffProtectionSet@Context, and ImprovementLoopBoundaryCondition@Context are compatibility and retrieval spellings only. No suffix or context label supplies a container, participant, ClaimScope, applicability, or identity discriminator. Use the suffixless interface names QualityEvaluationQuestionFrame, QualityEvaluationUseDeclaration, and QualityImprovementLoopRecord.
Conditional Improvement Unfolding Structure Block
Use this block when a named review or replay use relies on the improvement loop's constraint-governed unfolding structure rather than only its method record. The block identifies the proposal epistemes, predicted evaluation-result changes, independently identified pass Work and results, guarded alternatives, decision value, information-basis hold, stop, and returns to neighboring work. A generic position in a structure does not identify which proposal, result or continuation occupies it.
ImprovementLoopUnfoldingStructure is a local A.22.CGUS U.Structure specialization whose improvement-loop membership predicate is defined here. Its constituents are the independently identified values named above; its selected obtaining relations and guard claims keep their exact predicates, occurrence-identity rules, and defining ClaimGraphs. A position row, adjacency, or selected continuation creates none of them.
When that selected structure additionally satisfies the transformation-flow membership and boundary conditions, E.18/E.18.3 recognizes the same U.Structure; do not manufacture a generic CGUS plus a second transformation-flow structure from reciprocal references. The organization is neither a root U-kind, enduring Work, context container, evidence, nor quality proof.
E.23 governs the coordinate-qualified prediction episteme:
ExpectedEvaluationChangeExpressionKindValue is expectedValue | expectedRange | expectedDirection. Exactly one of value, range, or direction is present according to that kind. An expected value includes its exact kind and is admitted by coordinateScaleRef; an expected range belongs to that scale. EvaluationScaleDirectionValue is increaseOnScale | decreaseOnScale | preserveWithinRange | enterDeclaredRange | leaveDeclaredRange. Free direction prose does not close this episteme.
The episteme predicts a later re-evaluation result. Its listed prediction fields contribute to claim content; a new claim content, EntityOfConcern, or effective reference scheme yields another C.2.1 episteme edition. A changed grounding, viewpoint, applicability, qualification, source-currentness, carrier, or rendering relation does not by itself change the prediction episteme; revise its claims when that change alters the prediction. It is not an operation, move, transition, work occurrence, or proof of improvement.
ImprovementLoopDecisionValue is stop | continue | switchMethodFamily | openNewFrame | holdUntilInformationBasisSufficient. The hold value has non-empty unfilledInformationBasisPositionDescriptionRefs[] and an informationBasisSufficiencyConditionRef; other values leave both absent. Each description says which information-basis position is unfilled without referencing a nonexistent entity. The sufficiency condition says what information would make continuation admissible. A decision value or selected-continuation claim neither authorizes nor performs the next action.
ImprovementLoopBoundaryCondition@Context carries boundaryConditionKind = stop | subjectAssertionReconsideration | informationBasisSufficiency, a condition description, the affected object-version ref and exact kind, the unresolved assertion ref, and an optional non-semantic candidateSubjectPatternLocator. Source currentness, selected-set result declaration, actual publication, Work, evidence, and assurance remain distinct subject assertions under their exact predicates. A reconsideration boundary ends or redirects this E.23 use; it makes no later Work, decision, or relation obtain.
A visible cycle such as "draft -> evaluate -> repair -> re-evaluate" may be useful before execution. While any constituent, obtaining relation, guard, expected result change, protected trade-off, selected continuation, decision value, stop, or return needed for the wider improvement CGUS remains unresolved, keep that presentation as a ProvisionalUnfoldingDemonstrationDescription@Context about the object version and proposed continuation set. It may guide slot discovery, but it is not yet a structure or a slice.
Admit the wider ImprovementLoopUnfoldingStructure first. Only then may a separate DemonstrativeUnfoldingSlice@Context select one traversal through that admitted structure and name it as EntityOfConcern. Neither episteme is a QualityImprovementLoopRecord, performed Work, actual Transformation, or proof of improvement.
How the conditional detail supports the loop
The names and schemas in 4.2 and the unfolding-structure block in 4.2a support the ordinary steps above; they do not define a second loop. Use them only when a receiving use must inspect exact record identity, evidence positions, independently admitted Work and result relations, guarded alternatives, or replayable structure. Otherwise keep the ordinary account and do not manufacture a record or structure merely to complete the schema.
For a structured use, the complete result of the selected evaluation belongs to step 4 (including the complete E.21 result when E.21 is selected), proposal and Work separation to steps 5–7, before/after comparison to step 8, guarded continuation to step 9, and the replayable record to step 10. The central Work rule, direct result or change relation, and precision-restoration checks remain the same in both forms.
Stop, continue, and reopen
Stop when the current object version meets the declared floor or improvement aim and no feasible non-dominated proposal remains worth its cost under the current use, comparison set, source state, and protected trade-offs. If the remaining proposal mainly makes a value easier to argue while adding apparatus or worsening use, affordability, locality, source preservation, or ecology, reject that proposal; continue searching for a substantive content improvement if the improvement aim is still open, and stop only with a by-value no-proposal disposition.
Continue only when at least one predicted evaluation-result change states a scale-qualified change worth its cost and risk. Use ExpectedEvaluationResultChange@Context when the conditional account in §4.2a is needed.
Switch method when the current method family is not changing the evaluated result, is too costly, or no longer fits the evaluation. Use holdUntilInformationBasisSufficient only with non-empty unfilled-position descriptions and the sufficiency condition that would make continuation admissible.
A needed observation can prepare that continuation without itself changing the object under improvement. Identify the unresolved question, how the observation could change the next decision, and what information would be enough. For example, §4.4.1 can return to current-process observation before selecting an intervention. Keep that preparatory result distinct from an improvement pass and from evidence that quality improved.
An all-5, all-exceptional, current-front-reaching, or current-front-improving result is neither necessary nor sufficient to stop. Use the practical gain, protected trade-offs and attainable cost in the preceding rule. A stopped loop can reopen when a new use, Q component, source anchor, SoTA front, comparison set, affordability boundary or worthwhile proposal changes that judgement. If the object still fails its required use and no repair is feasible, choose replacement, withdrawal or another admissible continuation; stopping work does not make that object adequate.
Treat the five decision values as current continuation dispositions, not as Work states. When the admitted A.22 structure is used, a branch is usable only when its guarded continuation cites the exact current guard or constraint claim and the already-obtaining relation occurrences that make that alternative admissible. A stop or subject-assertion reconsideration is a boundary until an exact stronger predicate and current facts establish another relation. Naming A.15, E.22, G.11, G.5, or another subject pattern as a locator neither performs Work nor creates an object described there.
Method-family selection
The selected family is justified by characteristic-space fit, the declared scale-qualified predicted evaluation-result changes, cost and risk, and protected trade-offs. Familiarity, automation, or current popularity is not enough.
For PDSAorPDCAFamily, use the selected formulation. Deming's PDSA emphasizes study of a prediction; ASQ's PDCA also includes analysis and learning in Check. Distinguish maintaining work under an adequate standard, testing a change to it and organizing sustained use of the change. The relevant method supplies the operations needed by the improvement question.
For POOGIFamily, distinguish selecting the system constraint, responding to a current buffer condition, changing buffer settings and organizing a transformation. The Cox–Boyd account connects Buffer Management, Five Focusing Steps and Change Question Sequence as distinct contributions. Their use does not impose a fixed number of control levels. Current buffer regulation remains a neighbouring operational method, not a compulsory step of every quality-improvement pass.
Where recurring departures or changed conditions require more than current regulation, connect the selected inquiry to the change it can justify and to later performance that can test the change. A local improvement may require an agreement across organizations; tactical configuration may require reconsidering strategic assumptions; a proposed remedy may require inquiry into feedback and delayed effects. Library.Reference:6.5.3.2 explains these distinct extensions and their source conditions. Select the contribution needed by the improvement question, preserving the evaluation basis and the rights of affected participants.
For OODAFamily, recover orientation, implicit guidance, explicit hypotheses and their tests through action. Common understanding and intent can support coordinated initiative under changing conditions. Preserve Boyd's own distinction between command/control and appreciation/leadership when relating these contributions to the general control account in C.43.
If choosing the family requires clarification of the controlled subject, a changed goal, development of the performer or the organization needed for performance, use Library.Reference:6.5 (source document not included in this publication: FPF-LIBRARY-REFERENCE.md#libraryreference65---compact-cycles-in-working-control-arrangements). It connects these family conditions to the subject methods and control arrangements that supply the missing contribution. An already adequate cycle can be used directly.
Improve a work process through target conditions and experiments
Use this branch when a team must improve an actual process but cannot responsibly prescribe the whole path to the desired state. The working question is: What should the process do next, what currently prevents that, and which observation or experiment can justify the next move? Rother's Improvement Kata provides a way to answer it repeatedly. A useful local result can be a bounded target condition and an informative next step; it need not be a complete solution to the longer-term challenge.
Choose an observable process and an attainable period of work. When people are learning the way, a small process with frequent, visible repetitions makes the connection between action and consequence easier to examine. Later applications can address harder or slower work. In a long-cycle process, an offline trial or simulation can provide earlier information, while the conditions that only real operation can establish remain to be checked there. Use an already adequate way directly when the receiving task needs no inquiry or practice.
Establish direction and study the current process. State the challenge that makes this improvement worthwhile, the desired outcome and the qualities that must be protected. A direction may come from a larger organizational target, but the team still needs to understand its own process. A long-term direction is not a predetermined list of intermediate solutions.
Observe where the work happens. Follow the material, information or customer request through the relevant steps; identify how work arrives, how an output is completed and where waiting, rework or variation occurs. Compare recent outcome observations with the required performance. Then examine the operating pattern that could explain that difference. For repeated output, observations at a consistent completion point distinguish the interval between outputs from the time spent on an individual task. Retain unusual observations and the circumstances that may explain them. An average can conceal the interruption or variant that should change the next action.
Use the analysis that the process requires. Where demand and available time govern output, derive the needed rate from those values. Where equipment is involved, distinguish unattended running, handling that delays the next cycle, and work that can occur in parallel. Where staffing capacity is the question, examine repeatable work content and variation before treating a simplified capacity estimate as available capacity. For office or project work, trace actual cases, their information dependencies and delays. These distinctions select a useful next question; they do not impose one manufacturing worksheet on every process. Initial observations can be sufficient to choose a step even while other facts remain unknown. Refresh them after intervening changes.
Construct the next target condition. Combine a near achieve-by date, the outcome sought within that scope, and the operating attributes expected to produce it. For example, a target might concern how complete inputs become available, how often an output is completed, or how work passes between two operations. Keep safety, correctness and other protected qualities explicit. The target describes a desired operating state; buying software or instructing operators describes a possible means of reaching it.
Derive this target from the current observations and the direction. Choose enough stretch to require useful learning, with a horizon and scope in which the team can act and see consequences. A wider ambition may first require a smaller target. Keep the operating attributes already adequate and leave genuinely unknown details unresolved rather than filling a form with invented knowledge. When a missing fact prevents even a meaningful target, obtain that fact first. The next target after this one will depend on what is learned; it need not be fixed now.
Select an obstacle and the boundary of present knowledge. Ask what observed condition prevents this target. A current queue, an inconsistent input or an unexplained variation can be an obstacle. “We have not installed our preferred solution” has not yet identified one. Describe what that solution would change and check whether the corresponding problem is present. An acknowledged lack of knowledge can itself determine the first step.
Keep a provisional list when several possible impediments would otherwise be lost, and select one relevant, manageable obstacle for the next investigation. The list is not a commitment to remove every item. Later observations can show that an item was mistaken, has disappeared or no longer matters to the target. A beginner may learn from a tractable obstacle; selecting the largest one is not a universal prerequisite.
Choose the next step from what is not yet known. Three useful choices have different results:
- Observe the process without changing it when the current condition or the supposed obstacle is insufficiently known. State what the observation should distinguish, where to look and what facts to bring back. An expectation about what will be seen helps expose assumptions; it need not be a fabricated numerical effect prediction.
- Try a bounded exploratory change when a small, permissible perturbation can reveal how the process responds. Predict the response as far as the present knowledge permits and identify what remains uncertain. Protect the people, product and continuing work if the prediction fails.
- Test a particular explanation when the proposed causal relationship is clear enough to discriminate. State the predicted effect, the observations and comparison that could support or challenge it, and the conditions needed for that inference.
Match the comparison to the claim. To learn how a component affects performance, arrange variation that distinguishes its contribution from other changes. When components may interact, compare the relevant combinations: the effect of one change may depend on whether another is present. A qualified factorial design can support those distinctions.
When the decision concerns adopting a complete new way of working, compare that way with the available alternative under comparable receiving conditions; the warranted conclusion concerns their overall difference. A comparison of whole ways does not by itself identify their components' separate effects. The subject's experimental method determines allocation, replication, measurement and analysis. Carry changes in inputs, assistance or operating conditions into that design or restrict the resulting claim.
Choose the size and timing of the next contribution. Select the least burdensome sufficient action or coordinated set of actions for the decision, including observation, analysis or preparation where needed. One obstacle may require several factors or several experiments to be considered together. When the experimental arrangement remains unsettled, compare feasible alternatives, such as a single trial, a planned batch or an adaptive sequence, by the information they can supply, their full preparation and execution burden, delay and risk.
With long feedback, a batch can examine useful combinations in one waiting period; an adaptive sequence can justify additional waiting when an earlier result can usefully change a later experiment. A planned sequence may also need preparation whose value appears only in the later comparison. C.40:4.7 supplies the related choice of how to search and compare candidate ways; the relevant domain method supplies the experiment itself.
Before acting, agree the intended action or set, the expected result or learning, how it will be observed, the limits on harm and the point at which useful information should be available. A short trial is useful only if the process has had time and opportunity to produce the relevant observation. Involve the people who perform and understand the work. A coach can help the learner design this contribution while leaving the learner's needed judgement observable.
Compare and decide what follows. After the step, distinguish the observed facts from the explanation drawn from them. Compare them with the expectation stated before acting. The result may support the explanation, contradict it, partly fit it or leave the question unresolved. Use the difference to update the current condition and choose what now needs to be learned or changed. A successful prediction can also provide useful knowledge. Rother uses experimentation broadly for this learning sequence; an observation or exploratory trial does not acquire the warrant of a controlled causal test by that name.
Retain a useful change only within the conditions its evidence supports. Before sustained use, check the protected qualities, the affected process as a whole and the operating conditions under which the change must continue to work. A successful trial and a stable working arrangement are different results. Return to §4.3 for the justified improvement continuation or stop; observation undertaken to fill a missing basis retains its preparatory status. Keep an adequate current arrangement when another change has no worthwhile contribution.
When the target is achieved or its agreed date arrives, compare the result with the original target and review what the process and the participants have learned. A missed target remains a missed target for that interval. Decide the next target from the resulting current condition and direction. If demand, safety, scope or the larger purpose has materially changed, reconsider the target and its grounds; do not preserve an obsolete target merely to keep a routine going or rewrite the old result as success. HCD.9:4.10 supplies the separate review of the learner's reasoning and the coach's help.
Worked example — a contract-entry bottleneck. This constructed example adapts the reasoning difficulty in the Practice Guide's sales case; its times and subsequent trials are illustrative, not measurements reported by Rother.
A team wants an ordinary contract completed within 20 minutes. Older observations suggest 31–39 minutes overall, including 10–14 minutes entering data. The learner proposes automating entry. The coach asks when those observations were made and what now consumes the entry time. Because the available observations predate a change to the input form, the next useful step is to observe current contracts, separate typing from lookup and correction, and bring back the sequence and times. The team does not yet have a justified automation proposal.
Current observations of the selected ordinary variant show 11–13 minutes of entry, including three to five minutes repeatedly searching for approved item codes. The team agrees a near target: by Friday, entry for this variant should take at most eight minutes, with verified inputs available when entry starts, no invalid codes and no added delay elsewhere. The 20-minute overall challenge remains open. The desired input condition does not prescribe whether its means should be software, a changed handover or another arrangement.
The learner now predicts that a view of the existing approved code source, available at entry, will remove repeated lookup without adding a separate preparation task. A safe offline comparison uses matched saved contracts, keeps their content and correctness checks comparable, and alternates which procedure is tried first. This helps distinguish the changed lookup procedure from simple familiarity with a particular contract. Both code correctness and time are observed. In the constructed result, entry takes seven to eight minutes and the codes remain correct for the tested variant. That is a reason to examine a bounded supervised application and the whole-process consequences. It is not proof that every contract can now be completed within 20 minutes. Representative live work and maintained access still determine whether the change can become the ordinary arrangement.
The same contract work can pose a different experimental question. Suppose the proposed route changes both the input handover and the code view. If the team needs to understand whether the changes work separately or together, it needs observations that distinguish the four combinations, with the assignment of contracts, repeated observations and analysis chosen to support that inference. If it only needs to choose between the complete new route and the current route, compare those two arrangements for correctness, completion time and total work. That second comparison can support the route choice without identifying which component caused the difference.
A condition that changes the next move. A special contract needs a customer choice before an item code can be approved. A faster lookup cannot supply that missing decision. If this contract type is now required, the team first establishes when and by whom the choice can be obtained, and revisits the input handover or target scope. It preserves the demonstrated benefit for the ordinary variant without claiming the broader target. If the customer response time makes Friday's broader target infeasible, that fact changes the target decision; it is not a reason to keep running identical lookup experiments.
The process and coaching sequence is developed from Mike Rother's The Toyota Kata Practice Guide, the 2018 electronic edition, Chapters 5–9 and 11–12. It supplies current-condition analysis, target-condition construction and learning at the boundary of knowledge. The experimental distinctions use the NIST/SEMATECH handbook's interaction, factorial-design and randomized-comparison explanations. Treating a complete way as the compared alternative is the application made here. Huan, Jagalur and Marzouk, revised 2026, §§5 and 6.3 distinguish batch and adaptive experimental design and its modeling and computational demands; model-based experimental-design optimisation requires a suitable model and enough expected benefit to justify its computation. The contract examples are constructed applications. Rother supplies a practice prescription and teaching cases; the experimental-design sources justify comparisons under their stated model conditions.
Operation-family selection
An operation family is selected only when the loop account names:
- one scale-qualified predicted evaluation-result change;
- failure mode addressed;
- cost or risk reason;
- protected trade-offs;
- stop or removal condition.
Typical operation families are specification articulation, task decomposition, context refresh with carry-forward evidence, failure-context retry, verification against specification, memory or distillation, external critic or co-regulation, proposal portfolio use, search breadth or variants, bounded object-change budget, held-out evaluation, rejected-change memory, optimizer-memory separation, source-anchor contribution assignment, agent-tool-interface hardening, and task-family adaptation signature. They remain selectable only for the loop that justifies them.
Cost and BLP discipline
Use C.19.1 when the decision relies on a scale advantage or an adopted preference for general methods. Establish the advantage for the intended tasks and resource settings, or apply the preference under its stated conditions. Compare material resources, tools and instruments, adaptation attempts, skilled attention, rework or delay, risk exposure, and avoided loss on their admitted scales. Keep the components separate. Establish which options satisfy the binding conditions for the intended result. Exclude an option on grounds of dominance only if an eligible rival dominates it under the governing comparison. Use the declared project policy to choose or hold among the remaining alternatives.
Net-cost arithmetic is permitted only after every term has been converted to one declared unit through an admissible conversion whose basis, uncertainty, and scope remain visible. Until then, avoided loss is a separate project estimate rather than a quantity subtracted from concrete burden. A justified avoided loss can still make an expensive loop preferable. For a simple object, a direct edit or adjustment, small repair, lower-cost performer, specialized cycle, or one-shot evaluation can remain the better option.
Choose harness improvement when a named defect in framing, tools, memory, verification or stopping causes avoidable retry and repairing it compares favourably with direct object repair or another available method. Apply C.11.DUA to preparation, recurring use, transition, independent judgement, maintenance and displaced useful work. Leave unavailable cost evidence unknown; a larger harness is not itself a gain.
Source-composed, OEE, and NQD improvement
Accepted SoTA is the working external front only when assigned by the object-under-improvement evaluation, accepted source-use decision, or declared comparison set. E.23 can govern a loop that reaches, maintains, or improves relative to that front; it does not self-assign SoTA.
When an evaluation-result change depends on source use, source currentness, or a dated external front, the loop record cites the exact accepted source contribution and bounded use, including a G.2 pack result when that route was used, or the relevant G.11 currentness result, with the edition or date needed for replay. E.23 carries that reference; it does not make the source-use or currentness decision.
When several source anchors are used, retain each accepted source-use decision and contribution. If the improvement depends on their combined use, construct the needed connection under the subject Method; for framework Methods, use C.39 and E.4.CM. Preserve independently usable contributions when no composite Method is needed. Relate the changed result claim to those decisions and contributions and re-evaluate the changed object version. Use SourceComposedResultClaim within a separately constituted result episteme when the receiving use needs that account.
For NQD and OEE, use E.23 to change one object version or candidate and re-evaluate it on declared Q coordinates. Use C.17 for novelty, diversity, descriptors, and distances, C.18 for archive and front insertion, C.19 for pool policy, G.5 for selected-set result declaration, G.9 for parity, and G.11 for currentness and refresh. When audience availability is current, use E.17 for a source-backed publication face and return to source and E.24.PUB for the publication occurrence, form, carrier, audience, bounded use, and availability.
Archetypal Grounding
Working rule. Name the object version and evaluation, make one bounded change, and re-evaluate the changed object before claiming improvement. Identify dated improvement Work only when that occurrence is asserted. Keep proposals, performed Work, the changed object or Transformation, the later evaluation, and its returned result distinct.
Agent harness improvement from a loop-engineering request. A user asks to improve a local DPF seed. The record names that seed version as the object under improvement, selects E.4.DPF.DA or E.21 for evaluation, and states the aim: make the seed usable for local first entry without public-Core claims. The loop may change only that seed or another declared evaluation or harness slice. Prompts, adversarial examples, and harness checks enter only when the record states the expected evaluation change and their removal or stop condition. Selection makes them proposals. Each actual harness run or seed edit is separate Work governed by the central §4 rule; returned values, durable results, changed versions, and Transformations stay separate and use their declared bindings or relations.
Use the smallest sufficient source route in E.4.DPF:4 for the seed's needed source contribution, using G.2 only for a broad, refreshable CG-Frame pack. When the seed lacks the needed way or its usable explanation, develop that contribution through C.39 and E.4.CM before allocating pattern text. Carry the resulting contribution and limits back to the seed. Use G.11 for refresh, G.9 for parity, C.18 or C.19 for retained variants, and G.5 for selected-set results. Use E.17 and E.24.PUB for publication, and E.4.PFAD or E.4.PFR for their respective claims. A change outside the declared slice opens that neighboring work; it does not enlarge one E.23 loop without a new boundary.
Affordable floor evaluation. E.22 resolves the evaluation question if needed; the evaluator applies E.21 to every coordinate and returns the result through the declared application binding or result relation. If that evaluation is recorded as dated Work, the central §4 rule applies. If the result is admissible and no improvement aim was requested, E.23 stays closed. Any admission, refresh, landing, or release claim still uses its own E.19 or release gate; the E.21 result is quality evidence, not the gate.
Pattern exceptional improvement. A pattern already passes the declared floor. An additional worked case could extend its coverage, or a clearer account of dependent uses could make source refresh more replayable. Use E.22 when this further improvement aim needs framing. Selecting a proposal does not perform it. Add the useful worked case or refresh the source-bound rule, identify the repair pass as Work only if that dated occurrence is asserted, and re-evaluate the changed pattern through a later E.21 pass with its own result. Check what became worse. Stop at 4 when no worthwhile content improvement remains under the declared use; do not add apparatus merely to defend all-5.
Physical prototype improvement. The object is PumpAssembly@Prototype-3. Its evaluation declaration states any evaluator condition that changes result admissibility and keeps the vibration evaluation pattern and Method, Q-Bundle and characteristic-space descriptions, expected calibrated evidence, and result form distinct. The question frame binds those values to the engineering decision that will consume the result.
A test-bench measures Prototype-3 vibration. If this evaluation is asserted as dated Work, the central §4 rule applies. Its returned vibration value uses the evaluation's binding or result relation; any durable result claim is a separate episteme. E.22 then proposes an impeller-geometry change while protecting efficiency and manufacturability. The proposal is not performance.
Machining and assembly produce Prototype-4 through separately identified Work. Link them to the changed version or Transformation only through an obtaining A.6.1 binding, Work-to-result relation, Work-to-change relation, or applicable A.15.PROD local claim. A later vibration evaluation uses the same characteristic space and evidence basis before any measured-improvement claim obtains; each asserted Work occurrence follows the central §4 rule.
Three proposals remain three evaluated alternatives. Under the same evaluation use, quality model, and expected evidence basis, E.22 can return three CandidateImprovementProposalRow@Context values: change impeller geometry, change bearing-support stiffness, and add vibration isolation. The QualityImprovementLoopRecord cites all three separately, without treating any as performed. Each has its own ExpectedEvaluationResultChange@Context for the pump-assembly version and keeps protected trade-offs such as efficiency, mass, manufacturability, and service access separate. If comparable operating-point measurements are missing, the actual LoopEvaluationEvidenceBasis@Context names that gap and holdUntilInformationBasisSufficient states the comparability condition. A later pass may select only proposals still worth their cost and risk; actual improvement begins with separately identified Work and an obtaining result or change basis.
DRR improvement. A DRR needs drafting adequacy for authoring across several selected pattern hosts. Use the coordinates supplied by E.9.DA, return row-atomic proposals, repair the decision, and re-evaluate the changed DRR through a separate E.9.DA pass and result. Identify each repair or evaluation as dated Work only when that occurrence is asserted, using the central §4 rule. The improved object is still a decision record, not prewritten pattern prose.
NQD quality-side improvement. A generated candidate has declared Q components and a comparison set. An E.22 application returns proposal rows. Use E.23 to organize candidate changes and later re-evaluation of Q; apply the central §4 Work rule only to performed passes. Use the direct definitions and tests for archive or front insertion, selected-set result declaration, publication, parity, and refresh. None is a quality-loop decision.
Improve the way further improvements are obtained
The following constructed cases concern a way of improving work. They show how to obtain and use a changed way; they are not evidence that a particular training intervention succeeds.
A training organization develops its preparation of facilitators. Its participants can repeat a worked solution but keep using it after a decisive condition changes. The current preparation gives facilitators examples to explain. A proposed preparation instead teaches them to construct pairs of cases: retain the apparent task, change one condition that determines the appropriate action, and explain why the answer must change. The object being improved is this preparation method and its usable description. The facilitators, methodologist and tools are its actual means of application; stages in their reasoning are not thereby parts of a person under A.12.
Begin with one consequential failure. A solution is acceptable only when its entire error interval lies inside a tolerance. An exercise that compares only the point estimate teaches the wrong decision. The first facilitator repairs the exercise and the rule for constructing exercises: recover the condition that makes the action appropriate, retain its basis in the explanation, and build a contrast in which losing that condition changes the answer. C.40:4.7 supplies the comparison of obtaining ways, and C.40:4.11 supplies their further development from retained material. Preserve the previous preparation, its useful cases and the reasons for the change; a less successful variant may still contain a useful operation without becoming the selected preparation.
Try the old and proposed preparation on suitably comparable work under a declared allowance. Include case construction, tools, the methodologist's help, practice, assessment, correction and subsequent use in the cost. Judge the exercises' validity and what participants do on a fresh case, including earlier valid uses that must survive. If repeated grading itself changes which answers pass, apply §4.1.1 to the unchanged answers before attributing those transitions to learning. The comparison of human learning still needs its own basis. Use enough independent judgement for the decision; extra generated exercises do not create additional assessment capacity.
A better exercise is the first result. A different result occurs when the second facilitator uses the changed preparation to obtain the next improvement in that preparation. For example, the second facilitator applies its condition-and-contrast construction to a new failure, discovers that it omitted how to obtain the condition's basis, and changes the preparation so the next facilitator can do that work. Preserve the received construction, the new case, the reasoning behind the change and the resulting preparation. An instruction to make such a change is only a proposal. If the original methodologist still does all the construction, the result demonstrates a supported arrangement, not the successor's claimed independent operation. Assistance can remain part of the selected arrangement when that is the useful result.
A changed condition opens a different question. The next project depends on who has authority to make a decision, rather than on a numerical error bound. Substituting the word authority for a number is insufficient. Obtain the applicable basis for the decision right, determine the action permitted when that basis holds and what to do when it is absent, then construct cases that require the participant to use that distinction. Training does not create a missing right. C.40.CD develops this problem and the way of solving it together, retaining the original need to prepare useful facilitators.
Before calling this an expansion of possibilities, identify what the earlier way could obtain. If it already supported constructing and justifying cases about any relevant condition, the new example may be ordinary adaptation. If obtaining and using a non-numerical decision basis required a previously unavailable operation, demonstrate the work that the changed way now makes feasible. Use C.18:4.3a for that stronger comparison. A new project, a novel exercise or a higher score alone does not establish an expanded space. Better means may enable harder work that takes longer; acceleration is not required.
Make continuation possible for the next participant. Use C.36.RP:4.3–4.6 to decide whether the recipient needs to execute, adapt or develop the preparation. For development, give access to the construction, its reasons and conditions, relevant cases and the means or assistance for obtaining a missing contribution. Observe the recipient using these means on changed conditions and making the resulting construction usable in another preparation episode. Merely circulating the exercise set leaves this result untested. If a methodologist leaves, a surviving document may no longer be sufficient: recover the missing support or revise the arrangement before relying on the old success.
An independent participant outside the organization can continue this work without becoming its component. If the question now concerns transmission, retention or change across a cultural population, use C.36; count actual receiving and developing episodes for the declared question rather than treating popularity as the result. Changing those transmission conditions is itself a proposal to test.
A person improves a way of learning. An engineer has 90 minutes of personal practice and 15 minutes of mentor help each week. Familiar solutions are available, but changed constraints are missed. Instead of another rereading, the engineer retains a first unaided answer, compares it with the governing condition and a contrast, practices the missing operation, and tests a fresh case. To examine improvement of the learning method, follow its next use: can the engineer derive a useful learning question from a new work episode, select suitable practice and use the result for the following choice? Include help and the whole practice cycle in the allowance. One correct subject answer does not settle this question. If restored access to current data removes the difficulty without training, use the capability-access distinction in E.23.CAE and retain the adequate learning method. Developing a person's capability, when separately selected, uses E.23.CDI and the relevant human-development practice; the person cannot be reset to an earlier state for comparison.
An AI system provides the same distinction with different means. A fixed external author may repeatedly improve an agent's code without using the resulting agent as the author of the next improvement. To test that latter claim, place the candidate in the improving function and observe the next bounded development work under comparable resources and evaluation. A copied software version can preserve a starting basis; it does not justify an analogous rollback claim about a human or organization.
Return the observations to the selected question: what work improved, what changed in the way of improving, which changed means were reused, and what subsequent development became better or newly feasible at what cost. Keep each supported result within its conditions. Open a new E.23 frame when the object or evaluation question changes; use §4.1a when revising the evaluation. Stop additional method development when the available way is sufficient and a further change is not worth its burden.
Bias-Annotation
This pattern biases FPF toward adaptive improvement with explicit re-evaluation. The bias is useful because many objects improve only through feedback and revision.
The bias is bounded. One direct evaluation can close without a loop. Repetition is justified only by a scale-qualified predicted evaluation-result change and acceptable cost and risk.
Scope: limited. The pattern covers repeated improvement of one declared object version under one rerunnable evaluation. It is not a universal account of change, learning, capability development, cultural evolution, publication, release, or project authorization; use the subject pattern for those claims.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Rationale
The shared method is to select a proposed improvement, perform it, connect any returned value or changed object through its direct relation or A.6.1 binding, then re-evaluate through a separate pass and result. CC-E23-4 supplies the dated-Work account whenever either performed pass is asserted as Work. Check trade-offs and cost, then stop, continue, switch method, open a new frame, or hold.
When §4.2a is needed, A.22 carries the guarded alternatives, selected continuation, stop, and returns; when transformation-flow membership is independently current, E.18 and E.18.3 recognize that selected structure rather than a second loop object. Classical improvement cycles, agentic loops, fixed-performer optimization, MCDA, Goodhart, and OEE and NQD lines contribute useful operations and boundaries, but they do not replace this method or turn the cycle into enduring Work or context.
SoTA-Echoing
SoTA here means the current best-known problem-solving practice for the stated question, not the newest, most official, or most familiar source. Comparisons are qualified as of 2026-08-19 unless a row states its own qualification date. Lineage, bounded current research, internal governing dependencies, and rejected transfers are named as such.
Relations
E.23:End
Developing Capability for a Named Work Family
Tech-name:
WorkFamilyCapabilityDevelopmentMethodPlain-name: develop a System's capability for named work and check it in representative work Type: Method-description pattern for a separate capability-development Method; coordinated withE.23Status: Stable Normativity: Normative unless marked informative
Problem frame
Use this pattern when one named System must become more capable of performing a named Work family and transfer must later be checked in representative Work. The holder can be, for example, a person, team, organization or equipped arrangement. In every case, that whole must be independently admitted as the System whose capability is at stake.
First useful move. Distinguish the current capability account from the target.
Current: name the holder, the named Work family, operating envelope, measures and evidence for the baseline, the contribution that limits performance, and how long that account remains current.
Target: name the desired measures or success predicate, representative Work that will test them, intervention and any provider, and trade-offs that must remain protected. In both accounts, use only values that can change the development decision. If the holder or Work family cannot be named, stop before choosing training, tooling, or another intervention.
What goes wrong if missed. Attendance, an exercise score, a certificate, a published description, or successful provider Work can be mistaken for changed capability. A development programme can then optimize visible activity while the holder still cannot perform the target Work under its real conditions.
What this buys in practice. The project develops the capability that matters for named Work, directs effort at a real limitation, protects important conditions, and tests transfer where the capability will be used. It can stay small: use the compact entry.
Not this pattern when.
- Use
A.2.2when the question concerns only the holder's identity, qualified ability, measures, support or current reliance on that ability. If the present result can be obtained from an already capable provider and no change to the named holder is needed, arrange that provision using its qualified ability and the applicable work or commitment method. - Use
E.23.CAEfirst when previous performance or failed transfer leaves it unclear whether the live issue is envelope, configuration, applicability selection, access or activation, context-dependent expression, adaptation, enactment, or actual capability change. Its disposition is a premise, not selection of development Work. - Run the direct evaluation when that is enough; use E.22 only when its question needs framing.
- Use base
E.23for repeated improvement of an arbitrary object version. - Use
C.32.MWAfirst only when the target-practice architecture must itself be recovered or compared. - Use a population assessment for a distribution or statistic over member capabilities, and
C.36for cultural generation, transmission, recognition, selection, retention, or loss.
Problem
Capability development is often organized around available courses, tools, exercises, providers, or credentials. Those can contribute to an intervention, but none tells the project which System must become capable, which Work family matters, which contribution is limiting, or whether the change transfers.
The opposite mistake is to demand one universal decomposition or balanced scorecard. Human, technical, organizational, artistic, and hybrid holders can require different capability characteristics and different development Methods. The reusable part is the problem-solving move, not one curriculum, provider architecture, or scale.
Forces
Solution
Name the holder and target Work
Keep the holder System, its capability, the target Work family, development intervention, provider Systems, development Work, transfer Work, and evidence separate. A familiar group name does not by itself identify the capability holder.
Decide whose capability to develop from the work that will be entrusted. Follow the result to the contributors and resources needed to obtain it. When several contributions are required, could their timing or handover change the outcome even if each one succeeded? Include the relevant coordination in the account of the holder. For a claim about an individual using ordinary support, identify the contribution that individual must make and the help available at the time.
Apply the Method
-
Start from an evidence-backed account of the domain Methods needed for the named Work and how their contributions relate.
-
Name the operating envelope and the qualification window or currentness condition that can change whether the Work succeeds.
-
State the current capability baseline and desired result before choosing an intervention: the decision-bearing current measures and evidence, the desired measures or success predicate, and the trade-offs that must remain protected. The desired result is a target for later evaluation, not an established current ability or an actual change.
-
Find the contribution that is missing, limiting, or poorly coordinated. When the issue is which organization of a particular Method the holder can use, apply A.2.2:2.1 at the grain that exposes the missing relation; use :4.2.1 below to choose how it will become usable.
-
Choose an intervention directed at that limitation, and name any provider System on which the intervention relies. The intervention may use, for example, practice, coaching, adaptation, tooling, changed support, or a changed Method; the receiving domain supplies the choice.
-
Carry the protected conditions into the chosen intervention. Keep the selected intervention or plan separate from any development Work that is later performed.
-
Declare a transfer check that tests the stated target in representative Work under the relevant operating conditions. When that check is performed, keep the transfer Work, its evidence and result, and the resulting capability statement or currentness assessment separate. Exercise completion, attendance or provider delivery alone does not establish transfer. Reuse evidence from an exercise or another check when it tests the declared target in representative Work under the receiving conditions and satisfies its evidence requirements.
Build the transfer task from the choices the holder will have to make. When the next job supplies a procedure, representative work can ask the holder to carry it out. When the holder must devise a procedure, start with the problem and the resources available on that job. Identify the choices the holder made in constructing it. Where the outcome depends on several parts working together, establish what joins them under the receiving conditions. A worked solution or a heading that supplies the decisive relation narrows what the performance establishes. An earlier work product can already show the required choice under the relevant conditions.
-
Reopen the Method account, holder boundary, baseline, target, intervention, protected conditions, or currentness claim when the transfer evidence shows that the earlier account was wrong, incomplete, or no longer current.
Choose how Method structure will become usable
Use this branch when deciding how a holder will acquire or obtain a contribution of a Method, and a correct explanation or supported performance leaves that choice unresolved. Begin with the A.2.2 Method-relative profile and the actual receiving demand. For a sufficient ordinary intervention and transfer check, the preceding eight moves are enough.
Locate the missing connection. Work backward from the required result and through an attempt. Establish what selects the right action, what the holder performs, and what a teacher, tool, provider or reference supplies. Separate a missing description from an unsettled way of doing. Obtain the missing explanation from the source or author; if the Method itself needs development, assign that problem before treating it as a learner deficit.
The holder can discover an operation through its own inquiry and practice, recover it from a retained description, or learn from another agent's demonstration and feedback. These contributions can occur in the same attempt. A live demonstration is the demonstrator's Work. A recording or manual is a source the holder may use. Identify what the holder actually tried and what the source or helper supplied. A methodologist may reconstruct and critique the way from several performances, an author may describe it, and a performer may enact it. Identify the contribution at issue even when one agent does several of these jobs.
Compare attainable arrangements. Name the desired holder contribution before choosing among these alternatives; use the receiving domain's learning or provision Methods.
Connect help to later work. State what the next attempt receives and must produce, which help remains, and how its output reaches the receiving work. Keep legitimate tools or a useful division of work when independence is not part of the target. If removing help would disrupt a live service, use a protected rehearsal or retain qualified provision while developing the missing contribution.
If a teacher's exercise conflicts with a manager's production demand, settle time, workload and acceptance with the role holders who can authorize the needed changes; more instruction alone will not obtain those conditions.
Use feedback for the claim it can support. An adaptive arrangement can use a progress signal to choose another exercise or data batch. The resulting capability claim needs evidence of the selected Method relation in receiving work. When the signal improves but that work fails, retain the narrower claim, inspect whether the examples and feedback distinguish the missing relation, and revise that basis or the intervention.
When learning one contribution damages another required contribution, compare the whole receiving result and revise the arrangement against the protected conditions. A control Method may already supply its own approximation; develop a separate numerical estimator only when its answer changes the decision.
Return a bounded next action. Preserve the supported contribution and state the selected learning, provision or arrangement change, its expected effect on the gap, and the receiving check. Immediate success with help, later independent use, retention and transfer require the observations appropriate to those claims. Compare before and after under :4.3 before claiming acquisition or attributing change. The domain supplies the mechanism: a human learning model, a computational training algorithm and organizational coordination do not become interchangeable because they share this construction.
Return the first useful result
The first useful result names the admitted holder System and target Work family. Describe the current capability baseline, operating envelope, measures and evidence that matter to the decision, qualification window or currentness condition, and the contribution that limits performance.
State the desired measures or success predicate, selected intervention and any provider dependency, protected conditions, representative transfer check and reopen condition. If member distributions or cultural propagation are also at issue, return their separate assessment or C.36 result. A population can be the capability holder only when that whole is independently admitted as the System.
The holder, its qualified A.2.2 ability claim and the basis for relying on the current baseline must be recoverable. The practitioner-facing result can remain the short account above plus the evidence used for the baseline and transfer result. Expose record identifiers, a separate E.22 evaluation frame, or detailed provider and service relations only when a receiving use needs them.
Keep the selected intervention or plan, performed development Work, performed transfer Work, transfer evidence and result, pre- and post-intervention capability statements, a comparison claiming capability change, and any actual Transformation claim separate. A plan remains prospective.
For performed Work, identify every actual performer through A.13 and let A.15.1 independently admit the occurrence; add F.6 only when the record or receiving use expressly consumes precise assignment-bound attribution.
A capability-change comparison needs commensurable measures, envelopes, windows, and current support; it does not by itself say that the development Work caused a world-side change.
A causal Transformation claim additionally needs an independently identified A.3.4 Transformation and a named obtaining Work-to-change predicate or a supported local claim under A.6.RCD. Completion of development or transfer Work alone establishes none of these later claims. If the direct relation is missing, retain the Work, evidence, capability statements, and comparison and state what prevents establishing that relation.
Archetypal Grounding
Working rule. Begin with a current, bounded capability account and a desired result; choose an intervention only after the limiting contribution is known; then test that same target in representative Work. Completion of an intervention or exercise alone is activity evidence; a transfer claim needs the representative performance and support specified above.
The compact range table below is a recognition aid: it shows where the Method can be filled differently.
Incident-response team. An incident-response team performs drills successfully but loses coordination during real handovers. The team is independently admitted as the holder System for cross-shift handover Work. Its current baseline covers the present roster, dispatch platform, and staffed incidents: in six representative incidents, three handovers omitted the current owner or next action and median coordination recovery took eleven minutes; that evidence remains current only for the present roster and platform through the next quarterly qualification point.
The target is five consecutive comparable incidents with owner, current state, and next action handed over within three minutes, without worsening response time or safety. The project changes rehearsal and handover support.
The next five comparable staffed incidents each meet the three-minute handover condition. Median response time and recorded safety outcomes meet the protected criteria for that series. This completes the specified five-incident target. To qualify the team for later incidents, establish which roster, platform, duties and operating conditions the series represents, and for how long the conclusion remains current. Attribution of change to the intervention requires its own Work-to-change grounds.
Robotic inspection cell. One calibrated robotic vision cell is the holder System for inspecting machined impellers under the declared lighting, temperature, part-finish, and software-configuration envelope. On 200 representative parts, the current configuration detects 89 percent of the seeded reportable cracks, raises 9 percent false alerts, and takes 40 seconds per part; the account remains current through the named calibration window.
The target is at least 97 percent detection, at most 5 percent false alerts, and at most 45 seconds per part while traceability and safety interlocks remain unchanged. The intervention changes optical calibration and the inspection Method with support from the sensor provider.
The follow-up covers 300 identified parts over three shifts. Independent reference labels give 160 parts with one reportable crack each and 140 crack-free parts. The cell detects 157 cracks (98.125%) and wrongly flags six distinct parts among the 140 crack-free parts (about 4.29%). A 43-second mean is compatible with half the inspections taking 41 seconds and half 45, or with half taking 40 and half 46. Only the first timing record meets the 45-second limit on every part. With the first record, complete part coverage and qualified traceability and interlock operation, the engineer can accept this finite series under the stated criteria.
Did the cell improve? Suppose that in the original 200 parts there were 78 easy and 22 difficult cracks. Detecting all easy cracks and half the difficult ones gives 89 of 100. In the later series, 154 easy and six difficult cracks give 157 of 160 at exactly the same sensitivity within each grade. Compare the cell for a common part mix and operating envelope before claiming a change in capability. To attribute a change to optical calibration, the Method or provider Work, distinguish their effects from other changes that could explain the result.
A time guarantee for later parts can have a different basis. Suppose the entire interval from entry to result delivery consists of two serial stages taking at most 18 and 25 seconds, plus at most two seconds for everything else, including transfer and waiting. If these bounds hold for every admitted part, each inspection finishes within 45 seconds. Check these bounds for the installed configuration. A shared controller with an unbounded wait would leave the guarantee open. Detection and false alerts retain their respective qualification conditions.
Acquisition choice — a rhythmic controller. This constructed design case extends the A.2.2:6.5 contrast. The controller receives every pulse, including silent pulses, plus an onset indication, and must emit the hand assigned by pulse-wise alternation at each onset. With onsets 0,3,6, one correct first cycle leaves attack-wise and pulse-wise rules indistinguishable. A second cycle or a phase change supplies the missing distinction.
If the requirement is a controller for this fully specified rule and installing an implementation is allowed, choose direct implementation and check its outputs at the required phase and cycle transitions. This obtains a capable configuration without a learning experiment. Its evidence supports the configured controller, not a human performer's coordination.
Now change the requirement: the System must learn changing rule families from examples. Installing the one known rule no longer supplies that capability. A fixed sequence of examples that distinguish the relevant rules is a candidate when the distinction set is known. Adaptive selection becomes a candidate when successive observations can change which examples are useful.
For adaptive selection, the choice Method must define the observations and how they change later choices. Select it only when the additional effort is justified against the fixed sequence under the same resource boundary.
Suppose a trial learner's training signal improves while it still produces the attack-wise answer at the second cycle. The next action is to check whether the supplied examples and feedback ever distinguish the silent-pulse transition, then add or change that contrast as needed. Keep separate receiving cases for the phase/cycle behavior the claim promises. A successful training update does not settle those cases; a passing case supports only its stated scope. If the holder must retain another rule as well, check that contribution rather than trading it away unnoticed. This comparison specifies a proposed learning use and a response to a stipulated failure; it reports no completed training result.
Bias-Annotation
Scope: limited. This pattern offers one cross-domain capability-development Method for a named holder and Work family. It does not supply a universal curriculum, capability scale, intervention catalogue, provider architecture, population model, or cultural-evolution account. The receiving domain or DPF supplies the Methods, measures, evidence, intervention, and representative Work.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
The same Method can guide human, organizational, artistic, engineering, operating, technical, or hybrid capability development without copying one domain's curriculum or measures into FPF. Receiving DPFs supply the domain Methods, characteristics, interventions, providers, evidence, protected conditions, representative Work, and transfer limits.
The added cost is that a project must identify the holder and target Work before buying or designing an intervention, and it must test transfer rather than closing on activity. That cost prevents training, tooling, publication, or provider delivery from becoming a proxy for capability.
Rationale
The reusable cross-domain contribution is a problem-solving Method: start from named Work, find the capability limitation, change it through an appropriate intervention and provider arrangement, protect what must not be sacrificed, and test transfer. The domain content varies; the action and stop boundary remain recognizable. Keeping the holder, provider, development Work, target Work, and cultural continuation separate prevents a useful common Method from becoming a universal curriculum or a second capability ontology.
SoTA-Echoing
The comparison below asks which current practice changes the capability-development action at comparable effort. Standards and competency catalogues are treated as current practice references, not as ontology authority or proof that a holder is capable. Domain-bounded reviews remain qualified to their studied populations and interventions.
Currentness and reopen. Recheck only the affected row when an official edition changes, a newer systematic review overturns a used transfer or intervention conclusion, the receiving domain supplies a better method at comparable effort, or this pattern changes the baseline, target, transfer, or direct-relation rule carried by that row.
Relations
E.23.CDI:End
Capability Access and Expression Differential Probe
Tech-name:
CapabilityAccessAndExpressionDifferentialProbeMethodPlain-name: test whether a capability is unavailable, unrecognized, unexpressed, unadapted, unenacted, or changed Type: Method-description pattern for an observation-first differential probe; coordinated withE.23andE.23.CDIStatus: Stable Normativity: Normative unless marked informative
Problem frame
Use this when one identified holder previously produced a result, passed a qualified reference test, or has a current capability claim for a named Work family, but performance now fails, transfers poorly, or changes with conditions. Use it only when the next question depends on distinguishing at least two of these possibilities:
- the demand lies outside the claimed capability envelope;
- a performer, tool, record, interface, authority, state, or other support is unavailable;
- an available response is not selected as applicable;
- a response cannot be accessed or activated;
- context or interference changes expression;
- an available response cannot be adapted to the changed demand;
- the response cannot be enacted in the current performer arrangement; or
- the holder's capability has actually changed.
First useful move. State the case in ordinary language:
This holder has a qualified capability claim or prior result for this Work family under these conditions. The present demand is not met. Before choosing more development, compare available observations under a qualified reference condition and under the present demand. Reuse them when the holder, relevant conditions, update state and evidence requirements match the receiving question. Obtain only the missing safe contrast, without further development, and identify what it distinguishes. Report which failure came first when the observations establish an order. If that reference cannot be obtained, return the missing basis or an unresolved disposition.
First useful result. Return the controlled observations, one or more qualified differential dispositions, the strongest surviving rival (if any), the limits of the result, and candidate routes to the patterns or domain Methods that could receive it. The disposition can inform a later choice. It does not choose or authorize the next Work or establish that the Work was performed. The probe result alone does not establish hidden memory or a causal mechanism. Either claim needs further grounds specific to this holder.
What changes in practice. One failed performance alone does not show that the capability or memory disappeared. Ask whether the claimed demand, configuration, cue or routing, applicability selection, response access, adaptation, and enactment can be separated by a safe contrast. Development or redesign begins only after a separate steering or choice result uses that evidence.
What this buys. The probe can recover a still-available response, avoid unnecessary redevelopment, locate an observed failure in a contribution or in its coordination with others, and return a smaller next question to the applicable pattern or domain Method. It works across unlike holders because the common part is the contrast and disposition, not one theory of memory, organization, learning, or model internals.
Not this pattern when.
- Use
A.2.2when only the holder, Work family, envelope, measures, evidence, or currentness of a qualified holder-ability claim must be stated. - Use
A.15.8when one Work or WorkPlan configuration and its recovery relation already bound the whole question. - Use
E.23.CDIwhen capability development has already been selected and the live question is the intervention and representative transfer check. - Use
A.15.7when ongoing Work merely needs one next action; useC.11only when a current chooser andOptionSetalready exist and comparison can change the choice. - Use the direct domain Method when a human memory mechanism, organization routine, continual-learning algorithm, robotic controller, medical condition, safety rule, threshold, or intervention is the live subject.
- Infer capability only within the scope and conditions warranted by the occurrence and its supporting evidence. Use replay, simulation, staged testing or another protected evidence route when a live probe would be unsafe.
Problem
Apparent capability loss compresses different failures into one sentence: “they knew it yesterday,” “the organization forgot the routine,” “the model catastrophically forgot,” or “the robot cannot do it anymore.” Each sentence can trigger expensive or harmful Work before anyone checks whether the old response still appears under a qualified condition.
The opposite error is to explain every recovery with one theory.
Human sensorimotor memories can be expressed according to contextual inference; conceptual knowledge can remain inert until relational retrieval. Organizational performance can depend on roles, rules, records, artefacts, and authority; an AI function can remain represented while its activation is biased; a robot can retain a policy while sensing, state estimation, actuation, or configuration prevents enactment. These structures are not one memory system.
The reusable problem-solving move is narrower: bind one capability claim, control further change long enough to compare conditions, recover a reference response where possible, separate observable positions, retain rivals, and return an evidence-bounded disposition. The method must admit both genuine capability change and an unresolved result.
Forces
Solution
Bind the claim before probing
Name only the values that can change the probe or its later use:
- the holder System;
- the named Work family or current demand;
- the capability envelope, measures, evidence, and qualification window currently relied upon;
- the prior qualified result or reference condition;
- the present failure or unstable expression;
- the relevant performer and support configuration;
- any development, rehearsal, procedure change, parameter update, model update, calibration, or other change that must be held fixed where safe;
- protected conditions and probe limits; and
- the receiving question that a differential observation could change.
If the holder or Work family is unclear, return to A.2.2. If the current demand is outside the claimed envelope, return outsideClaimedEnvelope and stop the loss diagnosis. If a known configuration failure fully explains the case, use A.15.8 and stop here.
Use this compact record only when another use needs it:
This optional form records the probe result; it introduces no new FPF kind.
Run the differential probe
-
Check the demand against the claim. Confirm that the current task belongs to the Work family and capability envelope being relied upon. Compare measures and qualification windows before calling a difference loss or transfer failure.
-
Recover the relevant configuration. Name actual or intended performers, roles, tools, records, interfaces, authority, environmental conditions, state, and other supports only where their direct rules apply. Use
A.15.8when the question concerns one Work or WorkPlan configuration.Select the scale of the observation from the demand that failed. If it concerns one performer's action, a record of that action may suffice. If it concerns a joint result, ask whether all the observed actions could succeed separately while their timing, handover or use defeats the result. Compare the arrangements that differ in the relevant relation. Keep the conditions in which each observation was made.
-
Hold development and updating fixed. During the contrast, avoid new teaching, rehearsal, procedure rewrite, fine-tuning, parameter update, recalibration, or other development where safe and feasible. If the probe necessarily changes the holder, mark the affected distinction unresolved or narrow the claim.
-
Return to a qualified reference condition. When the needed contrast is not already supported by matching observations, recreate a previously successful or separately qualified condition without further development. Prefer an
A → B → Aor equivalent return when order, recency, or transition history can matter. Reappearance rejects simple global loss under those tested conditions; it does not establish general transfer. -
Vary a decision-bearing condition. For a remaining unresolved contrast, change the smallest condition that can distinguish live rivals: an overt cue, cue reliability, preceding decision state or uncertainty, role, record, authority, tool, task or context identifier, state feedback, blocked versus interleaved order, transition frequency, or another domain-relevant condition. A label or visible setting is not assumed to be the effective context.
-
Observe applicability separately. Ask whether the relevant Method, response, routine, policy, or prior case is selected as applicable before judging execution. For a person this may involve recognition or choice; for an organization it may involve routing, role, record, or authorization; for AI or robotics it may involve task/context identification, policy routing, or function activation. These are different mechanisms occupying one observational position.
In particular, observe what was chosen before a cue. If the cue supplies the missing operation, a later success demonstrates its use with that help. If it merely makes an already available relation recognizable, the contrast can support a difficulty in recognizing applicability under the original conditions. A prior observation that distinguishes these explanations may suffice. When both explanations still fit and would lead to different interventions, vary what the cue supplies.
-
Separate availability, adaptation, enactment, and result. Where the case permits, observe whether the response can be accessed or activated, whether it can be transformed for the changed demand, whether the actual performer arrangement can enact it, and whether the required result obtains. A later failure does not by itself establish an earlier one.
-
Compare live rivals. Retain every explanation still compatible with the observations. Ordinary cue effects, interference, envelope mismatch, configuration loss, applicability failure, access or activation failure, adaptation failure, enactment failure, and actual holder change are not synonyms.
-
Return a disposition and candidate routes. State what the contrast distinguishes and under which conditions. Identify the earliest established failure when the evidence gives an order. When failure depends on how several contributions fit together, describe the relation the observations support. Give the evidence, the strongest surviving rival (if any), any materially unsupported use at issue, and patterns or domain Methods that could receive the result. If no rival remains among those tested, state that bounded result. Do not choose a route on the strength of the disposition alone.
These steps organize observations, not a universal cognitive pipeline. A holder may implement them through simultaneous, recurrent, distributed, or structurally different processes.
Qualify the disposition
One case may support several dispositions, ordered where the observations support that order. Keep each observation and its limits visible. No disposition is a ChoiceResult, authorization, selected next Work, or performed Work.
Keep recognition and assurance separate
Recognition. A quick return probe is warranted when the practitioner hears “it worked before,” “the model forgot,” “the team knows the routine,” “the dancer can do it only in class,” or another apparent-loss phrase and at least two live explanations would lead to different next Work.
Assurance. Stronger reliance needs proportionate evidence:
- a qualified reference basis rather than a nostalgic recollection or one cherry-picked success;
- commensurable measures, envelopes, configurations, and windows;
- protection against probe-induced learning, fatigue, priming, adaptation, or update;
- enough order, cue-reliability, and transition variation to address the live rival;
- replay, simulation, staged testing, or specialist assurance when live probing would be unsafe; and
- holder-specific evidence before claiming a memory mechanism, causal explanation, or actual capability change.
A cheap reversible probe can support a narrow disposition. A high-stakes capability-loss, safety, medical, employment, deployment, or public-performance decision may require a direct domain evaluation and assurance account before anyone relies on the result.
Route without choosing
Keep the first result compact; one useful shape is:
Claim tested: [holder, Work family, envelope, window].
Controlled contrast: [reference, changed condition, and what was held fixed].
Observation: [what changed under which conditions, including any order or interaction the observations establish].
Disposition: [qualified differential result].
Surviving rival and limit: [what remains plausible and what is not established].
Candidate routes: [patterns or domain Methods that could receive the result].
During ongoing Work, A.15.7 can use the disposition as current information while recovering a next action. Use C.11 only when a current chooser and OptionSet already exist and comparison or another probe can change the choice. E.23.CAE supplies neither pattern's result. Use E.23.CDI only after a separate applicable steering or choice result selects capability development.
Archetypal Grounding
Human knowledge that remains inert
Situation. A practitioner can explain a structural decision principle when it is named but does not use it in a differently worded workplace case. The failure may concern retrieval, recognition of applicability, adaptation, enactment, or an envelope limit; “they forgot” decides none of these.
Probe. First confirm that the workplace case lies inside the claimed capability envelope. Without reteaching, present a qualified reference case that the practitioner previously solved, then the workplace case, then the reference again. Ask separately whether the principle is relevant before asking for a solution.
If a structural comparison makes the principle recognizable and the practitioner can then adapt it, the observations show reference availability and successful prompted use. They support applicability-selection failure under the uncued condition only if the comparison made an already available relation recognizable without supplying missing guidance. If new guidance remains a possible explanation, retain that rival and leave the stronger distinction unresolved. Neither observation proves one universal retrieval mechanism or complete workplace capability.
Human Capability Development or a direct domain Method can receive the result only after a separate next-action decision.
The case keeps the first probe small while retaining the distinction between prompted use and the stronger applicability diagnosis.
Organizational handover
Situation. An incident-handover arrangement previously obtained an accepted result, then fails after a shift, tool, role, record, or authority change. Calling this “organizational forgetting” hides the relation that changed.
Probe. Name the admitted organizational holder or performer arrangement and the handover Work. Compare the earlier roster, record, dispatch tool, authority, and interface configuration with the current one. Run a protected A → B → A replay or simulation: established configuration, changed configuration, then restored configuration. Observe which routine or Method is selected, whether it can be adapted to the new shift, whether authorized performers enact it, and whether the handover result obtains.
If the result returns when record access and authority are restored, configurationOrSupportUnavailable is supported. Restoring both together leaves open whether restoring either alone would also suffice. A request to identify which change the shift needs calls for evidence that distinguishes those alternatives. The combined result already answers whether the earlier configuration can recover the handover under the replay conditions. Routine theory, organization design, staffing, governance, and collective learning remain OCE or organizational questions; no human-like organization memory has been established.
AI or robotic apparent forgetting
Situation. A continually updated model or robot previously expressed a task function or policy and later appears to forget it. A benchmark drop can reflect parameter change, task/context routing, function activation, interface state, sensing, actuation, support configuration, or a demand outside the tested envelope.
Probe. Hold parameters fixed during the probe where feasible. Compare no task/context cue with a qualified task cue or routing intervention; vary cue reliability, state feedback, blocked versus interleaved order, and relevant transition statistics; then return to the earlier condition without further update. For a robot, keep sensing, controller state, actuation, calibration, tool, and environment conditions explicit.
If the earlier function or policy reappears under a qualified activation condition, availability under that condition is supported and simple global loss is weakened. Function vectors, context inference, routing, controller state, continual-learning algorithms, and parameter overwrite remain model-specific explanations.
If no response reappears and direct model or controller evidence supports change, capabilityClaimRevisionWarranted may be returned without claiming a universal memory mechanism.
Bias-Annotation
Scope: limited. This pattern supplies a conservative cross-holder probe and disposition. It supplies no universal memory substrate, latent context, capability pipeline, organization-learning theory, continual-learning algorithm, intervention catalogue, or choice rule.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
The pattern reduces premature retraining, redesign, procedure rewrite, and parameter updating by making still-available responses observable before a capability claim is revised. It also returns smaller questions: configuration recovery, applicability selection, access or activation, adaptation, enactment, development, claim revision, or the uncertainty the available evidence leaves unresolved.
The cost is a qualified reference basis, controlled contrast, explicit claim boundary, and possible protected replay or simulation. Context dimensions can be coupled, reference evidence can be stale, and the probe can change the holder. A narrow unresolved result is therefore a successful outcome when the available evidence cannot support a stronger distinction.
The practitioner asks which observable contrast could change the conclusion before moving from failed expression to capability loss or development Work.
Reopen condition
Revisit this pattern when a current FPF neighbor supplies the whole differential with less burden; actual human, organizational, and AI or robotic uses cannot share the observation-only action; a direct source correction removes a load-bearing contrast; a new direct consumer requires a different disposition; or repeated uses show that one observation position is an independent Method with its own result and boundary.
Rationale
A capability claim is bounded by holder, Work family, envelope, measures, evidence, and currentness. One failed occurrence does not rewrite that claim automatically. A controlled return to a qualified condition can cheaply distinguish global change from availability under at least one condition, while separate applicability, access, adaptation, and enactment observations prevent a later-stage failure from being projected backward.
The method stays transdisciplinary only by refusing a common hidden mechanism. The minimum result these different cases share is an account of the observations, the disposition, any surviving rival, the limits, and candidate routes.
SoTA-Echoing
Practice question. When prior performance, failed transfer, or unstable expression leaves several live explanations, what is the smallest current defensible move that distinguishes an envelope or configuration failure, applicability or access failure, context-dependent expression, adaptation or enactment failure, and actual capability change without yet choosing development or repair Work?
Selected best-known line and serious alternatives. Adopt an observation-first differential: bind the capability claim, hold development or updating fixed where safe, return to a qualified reference condition, vary the smallest decision-bearing condition, separate the observable failure positions, and retain genuine-change and unresolved exits. Two alternatives are to infer capability loss and begin development immediately, or to treat a latent-context or COIN-style account as the common explanation.
At comparable first-decision effort, the selected line can be as small as one safe reference return and one discriminating contrast. It supplies a discriminating observation that the immediate-loss default omits, while retaining genuine-change and unresolved outcomes. Its affordability depends on the qualified contrast and the alternative work it can avoid. The extra contrast needs a qualified basis, may contaminate or endanger the case, and may return unresolvedDifferential.
What the comparison changes. Immediate loss/development projects one failed expression backward into capability change; a universal latent-context account projects one useful human explanation across non-isomorphic holders. This comparison explains the separate observations in E.23.CAE:4.2 steps 4–8 and the evidence required for each disposition in 4.3. It also informs the three holder cases in 5, the assurance stop in 4.4, the anti-patterns in 8, and the source-sensitive reopen condition in 9. It leaves mechanisms and interventions with their direct owners.
Reopen this comparison when a direct-source correction removes a load-bearing contrast; a stronger current line supplies an equally safe, cheaper, or more discriminating first move; actual human, organizational, and AI or robotic uses cannot share the observation-only action without importing one holder's mechanism; or a direct consumer requires a different disposition or assurance boundary.
Relations
E.23.CAE:End
U.Ontic and Ontic Introduction Discipline
Type: Part E FPF authoring discipline pattern Status: Stable Normativity: Normative unless a section is marked informative
Use This When
Use this pattern when FPF work appears to need a durable ontic: a connected action-facing ontology unit whose stable identity and admissible uses depend on keeping several direct relation kinds, their participant identities and meanings, any required kind conditions, reusable declarations, and neighboring subject patterns coherent.
Start with a question the ontology must let someone answer. Identify the distinction on which their explanation, decision or way of working depends. Then characterize the candidate and test the available rule content, identity, relations, dependent uses and possible duplication. The ontology disposition can be a new durable ontic, a bounded local episteme connecting existing claims, direct use of existing assertions and their ClaimGraph sources, or an unresolved result.
A current source-use record states quote-only, reduced use, or a selected stronger source use with its exact provenance; omit it when source use is not current. Source use can accompany any resolved ontology disposition as a separate provenance decision. Use source-only as a stop only when no exact payload assertion has been selected.
A durable ontic is a reusable ontology unit whose exact defining or constraining ClaimGraph states its identity rule and minimal relation set for dependent FPF use.
A bounded local episteme is a claim-bearing U.Episteme that coordinates already identified entities, exact relations, and subject assertions for one named use. Direct rule-content use relies on those existing assertions and ClaimGraph sources without adding another ontology unit. An unresolved stop retains the inquiry without selecting one of those three payload dispositions.
Typical examples include:
- a bounded local episteme starts being cited as though it were a new ontology unit;
- a source expression or project-side expression keeps pointing to several FPF values at once;
- a draft ToC row names a calculus or object family, but no current defining or constraining
ClaimGraphstates its meaning; - one pattern description begins to repeat local slot-relation doctrine that other uses also need;
- a proposed subject needs one stable identity, constitution, or recognition rule plus the smallest set of governed relations that dependent use must keep coherent.
Primary EntityOfConcern. The pattern defines or constrains U.Ontic, the durable action-facing ontology unit. Each ontic-introduction decision episteme has one independently identified EntityOfConcern before judgment: a candidate entity, proposal episteme, or source-construct entity that carries the inquiry before any disposition is known. That object remains the EntityOfConcern for direct, bounded, durable, and unresolved results.
The selected direct-use object, bounded local episteme, durable ontic, or unresolved reason is the branch payload recorded in the result, not a replacement for the decision's subject. Source-use status does not change the subject.
If a revised result changes the ClaimGraph, C.2.1 identifies another decision episteme; any edition continuity is stated separately rather than hidden by swapping the EntityOfConcern. An unresolved phrase or topic list is not itself an EntityOfConcern unless the exact source expression or source construct has been independently identified.
Primary working reader. The first reader is an FPF pattern author or reviewer deciding whether several nearby pattern descriptions concern one ontic, several already identified values, or only a compressed source expression. The downstream reader is the practitioner who needs the resulting subject assertions and practical guidance to decide what can be done, claimed, relied on, repaired, compared, or stopped.
If a separate U.MethodDescription claim matters, apply A.3.1 and A.3.2 to identify its Method and show that the episteme substantively describes how that Method is done; the E.24 locator alone establishes neither.
Working concern and viewpoint. From the FPF-authoring viewpoint, preserve the subject's exact relations, assertions, and defining or constraining ClaimGraph sources without duplicating kinds or promoting a claim-bearing episteme for one named use into durable ontology.
First useful move. Recover the receiving question from the work that prompted it. Which subjects have to be distinguished or related for someone to answer it? For the introduction decision, identify one candidate entity, proposal episteme or source construct before judging it. State its identity rule and keep it as that decision episteme’s EntityOfConcern. Record source provenance when the decision relies on a source.
Then run Checks 1–4: reuse existing exact predicates and ClaimGraph sources, test exact identity, recover the needed direct relations or constitution, and test dependent reuse without duplicate ontology. Fill the typed disposition result from those tests alone. Its direct, bounded, durable, or unresolved payload never replaces the decision subject. If no exact candidate or source construct can be identified, keep inquiry material and do not fabricate a decision episteme.
What goes wrong if missed. FPF grows shadow ontology. The same project concern becomes a method in one place, a mechanism in another, a record in a third, and a local checklist in a fourth. Later uses then repair visible symptoms instead of settling the underlying kind, slot, and subject-pattern question.
What this buys. A durable ontic has a stated identity and named direct relation kinds, participant meanings, obtaining conditions, and occurrence-identity rules. RelationSignature and SlotSpec declarations are added only where dependent uses need reusable participant typing.
When those rules already close the receiving use, apply them directly. Add a bounded local episteme only when that use needs several already governed claims read together; its ClaimGraph cites their subject assertions and rule-content locators.
Not this pattern when.
- If one existing defining or constraining
ClaimGraphalready states the claim kind, write the exact subject assertion and use the pattern id only as its locator. - If the issue is only wording, use
F.19; openE.10orE.10.ARCHonly for FPF meaning that remains unresolved. - If the issue is only a new or revised mechanism meaning, use
E.20. - If the issue is only durable naming, use
F.18. - If the issue is only a pattern publication-form or section-order matter, use
E.8.
Problem Frame
Some FPF objects are small enough to define through one direct relation ClaimGraph. Others become candidates for a durable ontic when several direct relations and rule-content loci need persistent coordination across dependent use. U.Episteme is the central example: correct reuse depends on keeping its identity, components, direct relations, dependent same-individual episteme kinds, descriptions, and publication-side relations coherent without treating a card field, RelationSignature, or C.29 representation as the episteme itself.
The same failure recurs elsewhere. A project label such as algorithm, process, model, architecture, service, quality, time, rhythm, change, or source can point to several FPF objects. Choosing a better word does not recover those objects. Introducing one umbrella kind fuses entities and relations that already have defining or constraining ClaimGraph sources. The decision method described here tests whether a durable ontology unit is needed and which direct relations make it useful.
Problem
Without this discipline:
- Local epistemes become pseudo-ontics. A repeated claim-bearing episteme or reusable publication form starts to be cited as a new ontology unit even though its claims or layout only refer to existing governed values.
- Draft ToC rows become false authorities. A planned ToC row is cited as if it already supplied current governing text.
- Pattern placement is mistaken for ontology. A numbering or placement label becomes the proposed ontic even though no primary governed subject kind, exact identity or constitution rule, minimal governed relation set, or subject pattern is named.
- Reusable SlotSpecs are copied without a direct relation. Several patterns list similar SlotSpecs, but no direct pattern states the relation kind, participant meanings, obtaining condition, or occurrence identity.
- Existing typed values are duplicated. A new head repeats
U.Method,U.Mechanism,U.WorkPlan,U.Work, evidence, gate, source, or result relations under a new name.
Forces
Solution
The ClaimGraph at E.24 defines U.Ontic as the FPF kind for a connected action-facing ontology unit.
Before dependent uses rely on that unit, the accepted ontic-introduction decision states its primary subject kind, exact identity, constitution, or recognition rule, the smallest exact relation set needed by dependent use, any identity-bearing direct relation selected by an exact identity assertion, any reusable RelationSignature declarations, rule-content locators, named dependent-use reliance, and non-use boundary.
Connected is an admission condition. The decision names the smallest set of independently defined relations that makes the subject usable across the named dependent uses and states why each relation belongs. When an exact identity assertion selects one identity-bearing direct relation, say so; otherwise do not invent a head relation.
Action-facing means that the connection among rules matters to what a practitioner can understand or do. To choose the relation set, work back from that use. Which wholes and constituents enter the question for each relevant kind of holon? What must the model distinguish at those scales? Consider what the individual rules already establish. If they allow situations with different answers, identify the dependency or condition that accounts for the difference. Show how the proposed rules together, with the facts of a particular use, warrant the needed inference. One existing rule can supply it in full. A question about organizing several practices may require only one new distinction, or may be solvable entirely with existing ones.
A kind's independently justified membership rule also supplies no additional ontology unit by itself. When direct rules already close the named uses, record that no additional durable ontic is needed and select direct subject-pattern use, or bounded coordination when several claims must be read together. An empty relation set is not positive evidence of connected coordination, and adding a nearby relation merely to fill the set does not establish a contribution.
Keep two layers explicit:
- Instance layer. For each included direct relation kind, state how its actual participants are identified under their subject rules, how they contribute to the obtaining predicate, and which kind conditions that predicate requires. The defining
ClaimGraphsupplies those conditions and the occurrence-identity rule. An obtaining occurrence relates the actual participants satisfying that predicate. Kinds or declaration epistemes can themselves participate when the direct rule admits them; citing them in the decision establishes no such occurrence. - Ontology and declaration layer. The ontic-introduction decision episteme states which subject kind, identity rule, relation kinds, declaration epistemes, rule-content locators, and dependent-use reliance claims belong in this ontology unit. These are typed claims in the decision episteme. A corresponding world-side declaration-dependency, inclusion, or reliance relation obtains only when its participants satisfy an independently defined predicate. Coordinating the objects in the ontology unit does not establish that occurrence.
The ontology unit is connected when every included relation kind has its instance-layer participants, exact predicate, and defining ClaimGraph, every included declaration is tied to the relation use it declares, and every dependent use names the exact identity rule, direct relation rule, or declaration it relies on. The decision marks any identity-bearing edge explicitly. This account of identities, relations, and dependencies establishes ontology-level coordination; it fabricates no relation occurrence among kinds, declarations, pattern descriptions, or the ontic.
Precise coordination does not require a new public kind for every participant. It does require a rule that recovers which object participates, what counts as the same or a different participant, and how it contributes to the relation.
Apply an existing kind criterion wherever the predicate depends on that classification. When a receiver additionally needs typed membership or substitution, A.6.5 requires independently established ValueKinds and complete SlotSpecs; an exact-value example cannot establish the whole range.
A missing identity or obtaining basis blocks the affected relation claim. A missing required range blocks its typed use while leaving separately warranted exact-value uses available. An ontic supplies neither missing basis by being declared.
Named dependent-use reliance states each dependent use, its pattern-description locator when useful, and the identified ontic identity, direct relation rule, or RelationSignature declaration on which it relies. A pattern name without that reliance basis is insufficient.
Reidentify one U.Ontic by its primary subject kind, the exact identity, constitution, or recognition rule supplied by its defining ClaimGraph, and the minimal relation set selected for dependent use. Include an identity-bearing direct relation only when an exact identity assertion selects one.
Only a change to the subject kind, identity rule, or relation set used by a named dependent use can reopen ontic identity; a change in how the subject is described, published, viewed, represented, or named does not. Use the typed object map in E.24:4.3a for those neighboring objects.
Compare the included relation kinds by their participant meanings, obtaining predicates and occurrence-identity rules. Unchanged printed names do not preserve a relation whose governing meaning changed. Publishing an existing rule in another expression differs from changing that rule.
Keep the subject under decision separate from every means of stating, presenting, or inspecting it. Open a neighboring row in E.24:4.3a only when that object's identity or direct relation changes the current choice or receiving use. A decision or description remains a C.2.1 episteme; availability, viewpoint conformance, and mathematical correspondence do not alter the subject's identity.
Keep direct verbs with their exact subjects and predicates: a designator designates, a reference resolves, an episteme contains claim content, a publication occurrence makes one edition available, a particular expression conveys its required claims for that use, a reusable form constrains that expression, and a carrier bears the expression. The typed map supplies the exact assertion, defining or constraining ClaimGraph, pattern locator, and stop for each current object; visible co-occurrence on a card supplies none of them.
When a durable ontic is selected, its branch of the ontic-introduction decision states at least:
- the primary governed subject kind and the named receiving use—such as comparison, preservation, teaching, publication, reference, work, or decision use—for which coherent identity and relation rules matter;
- the exact identity, constitution, or recognition rule supplied by the defining
ClaimGraph; - the smallest set of independently defined direct relations needed by named dependent use, with the practical use each relation enables;
- one identity-bearing direct relation only when an exact identity assertion selects it and its defining
ClaimGraphstates participants, predicate, and occurrence identity; - any
RelationSignatureepistemes used to declare reusable SlotSpecs for relation-participant meanings reused; - the current FPF patterns that define or constrain the subject kind, identity rule, and selected direct relations;
- the pattern that defines or constrains the durable ontic;
- the named dependent-pattern reliance: each dependent pattern and the identified ontic identity, direct relation rule, or
RelationSignaturedeclaration on which it relies without copying that rule or declaration.
A project entity does not fill an ontic. It keeps its own kind and may participate in the ontic's direct relation or in a neighboring direct relation. A SlotSpec belongs to a RelationSignature declaration. An assertion or description episteme may designate the world-side participants by value or reference and claim that the direct predicate obtains. The participant, SlotSpec, designation, assertion, and relation occurrence remain different objects.
FPF ontology is therefore not one flat class list and not a collection of filled records. A durable ontic is one connected ontology unit over a small group of direct kinds and relations, linked at the ontology layer by the typed claims in its decision episteme. At the instance layer, only actual participants enter obtaining direct-relation occurrences. The same project entity may participate in relations governed by several ontics without changing its kind or becoming part of a second ontology.
The cost is kernel growth and metamodel risk. Repetition, a reusable layout, or ontology-shaped wording does not make any object a U.Ontic. Admit one only when the decision supplies stable identity, the minimal relation set reused across dependent uses, existing-rule-content checks, and a non-use boundary.
Use E.24.UK for U-kind admission. Both patterns use the E24FamilySettlementDecision schema in E.24:4.0a:
- a durable ontic is a connected action-facing ontology unit;
- durable
U.*kindhood is admitted only through an acceptedUKindAdmissionResultunder that shared schema; - an ontic may coordinate already admitted kinds; a new kind may reuse an already accepted ontic settlement or be admitted without an additional durable ontic when direct rules suffice for its named uses;
- when the same case needs both a new ontic and a new public U-kind, one atomic co-decision returns a separate
OnticSettlementResultandUKindAdmissionResult; neither is evidence for the other inside that decision; - every non-ontic object keeps the kind, relation, exact subject assertion, and defining or constraining
ClaimGraphselected by the typed object map.
Use E.24.UK only when a candidate claims durable U-kind force. E.24 consumes its exact accepted result when that result changes the ontic settlement; naming or placement alone supplies neither output.
Constructive Foundation And Math-Lens Boundary
If a reader asks where an FPF ontic gets constructive grounding, follow its exact identity or grounding assertion and defining or constraining ClaimGraph. E.24 records a locator for that rule and only the relations needed by dependent use; it does not turn declarations, descriptions, publication objects, views, or representations into grounding participants. Their exact predicates, assertions, and rule-content locators remain in E.24:4.3a.
For structural identity, recover the subject’s direct identity rule and the construction or whole-forming facts it requires.
Use C.13 when the receiving use needs an inspectable construction account. When B.3.5 coverage is elected or a named requirement calls for that profile, apply the E.14/B.3.5 Working-Model requirements with the branch-specific validationMode and tv:groundedBy support, including the appropriate reconstructible Γ_m.sum, Γ_m.set or Γ_m.slice trace where that branch requires it.
Each constructor retains its own subject and conditions. A graph, tuple or other mathematical expression of that trace is a separate C.29 representation. The account, profile and representation do not replace the direct identity or construction facts.
For a non-structural ontic, use the exact identity, grounding, or recognition assertion and defining ClaimGraph located by its direct subject-pattern reference. Open E.24.UK only for U-kind admission, C.2.1 only for an episteme's identity, E.24.PUB only for current availability, and the other rows of E.24:4.3a only when their selection question is true.
A.14, B.2, and A.15.1 carry BORO- and CCO-compatible identity and occurrence discipline. They support the constructive foundation; they do not create a separate durable-kind ontology.
Before a dependent pattern relies on the ontic, classify each current object with E.24:4.3a. The selection question—not a shared label or visual container—decides whether the object is a world-side participant, relation occurrence, reusable declaration, claim-bearing episteme, publication object, view or representation, source expression, or durable ontology unit.
An encountered card illustrates the rule. Its claims, reusable layout, diagram elements, and carrier are separately governed only when their own identity and direct relation are established; the word card identifies none of them and does not make the collection an ontic.
When several current pattern descriptions already contain rule content for the same project concern, select an ontic only if one exact identity rule and minimal relation set must be reused across their dependent uses. Keep every otherwise current object in its E.24:4.3a row; shared topic or proximity cannot fuse their kinds. At the ontology layer, state reliance on exact relation rules without inventing an occurrence whose participants are the kind, pattern, or ontic.
Gather the evidence in this order. Select the disposition after the first four steps:
- Current case and stable decision subject. Start from what the recipient must understand, decide or do. Identify the working expression or source claim and the independently identifiable candidate entity, proposal episteme or source construct on which this decision bears. State its direct identity rule and keep it as the decision episteme's EntityOfConcern. Record source-use status and provenance when the decision relies on the source.
- Existing-governor reuse and non-duplication. Name the current direct patterns checked by value. State which current claim they already close, or the exact coordination they fail to supply. Reject a new umbrella when it would merely rename those governed objects or copy their rules.
- Identity, constitution, or recognition. State the exact rule supplied by the subject pattern and what would reidentify the subject across the receiving use. Do not replace several required facts with an invented universal relation.
- Identity, relations, and dependent use. Use
E.24:4.3ato classify only the objects that the dependent use consumes. Name each needed direct relation, its exact predicate and definition source, any relation that the subject’s identity rule selects as identity-bearing, the rule for identifying an occurrence of that relation, each declaration reused, and each dependent assertion's exact reliance basis. Omit every neighboring map row whose selection question is false. - Disposition, branch result, and boundary—fill last. Keep the decision EntityOfConcern from step 1. From steps 1–4, record exactly one branch payload: the closing assertions and direct patterns; the bounded episteme, its declared use, and stop; the selected ontology-unit individual and subject pattern; or the unresolved reason and missing evidence. Add an explanatory overread only when it passes F.19:4's full independent-ground, plausible-reader, contribution, and smallest-clear-correction test.
A relation-participant meaning belongs in one selected direct relation only when its predicate depends on an actual participant making that contribution. The defining ClaimGraph must state how the participant is identified independently of a declaration field and state any classification condition the predicate uses. When typed reuse is needed, a compatible RelationSignature additionally declares an independently governed participant kind as the SlotSpec's ValueKind. Another entity remains under its own direct relation when that relation already expresses the needed use. Reuse pressure can justify a RelationSignature; it cannot turn a neighboring relation, record field, or mathematical operand into a participant or SlotKind of another relation.
Optional-in-use status belongs to a declaration or description. It does not mean that a world-side relation occurrence has an unfilled participant. A missing designation leaves the assertion incomplete or the participant unknown to the current user. It does not show that the participant is absent, and it does not make the direct predicate obtain or cease.
Not every ontic needs every map row. Open one only when its selection question changes the named receiving use; otherwise omit it and keep the object under its subject pattern.
Keep annotation proportional. E.24 calls for recovery only where wording can change ontic identity, a direct relation, participant meaning, a reusable SlotSpec declaration, a description claim, admissible use, or the reliance basis of a dependent pattern. If readable domain prose already preserves those objects, do not replace it with declaration syntax merely to show that an ontic exists.
FPF patterns may define or constrain a kind or predicate, state an admission test, frame a judgement, or give practical guidance. That does not make every pattern episteme a U.MethodDescription or every subject a U.Method. An engineer-manager uses the applicable claims and guidance to decide what can be done, claimed, relied on, repaired, compared, or stopped.
If the current claim says that an E.24 episteme describes an ontic-introduction Method, apply A.3.1 and A.3.2 to identify that Method and show that the episteme substantively describes how it is done. The accepted ontic-introduction decision supplies the object discipline for those practical choices.
Precision restoration uses the same discipline without turning it into lexical style. First recover the source-side entities, direct relations, assertions, descriptions, and defining or constraining ClaimGraph sources compressed by the wording. Then repair toward a current FPF ontic only when one accepted ontic-introduction decision states how those objects are coordinated. If no such ontic exists, state the exact subject assertions, cite their pattern-description locators, keep only the needed claims in a bounded local episteme under C.2.1, or open an E.24 ontic-introduction decision.
When a source expression opens the ontic-introduction question, preserve its source-to-use path independently of the ontology disposition. Name the exact expression and its source episteme; name the source publication occurrence when availability through that occurrence matters; recover the entities, relations, and claims actually carried forward; and set the source-use status to quote-only, reduced use, or one selected stronger use with the smallest condition that licenses it.
Keep that trace beside a durable-ontic, bounded-episteme, or direct-use disposition whenever both are current. If no governed payload has been selected, mark the ontology disposition unresolved and retain source-only inquiry material rather than treating provenance as an ontology answer.
When a stronger-use condition occurs, reopen the source expression through C.2.P or the direct source-use pattern instead of treating the repaired noun as a substitute for the source relation.
When an E.10.ARCH wording-use restoration row opened the case, retain its four coordinates inside that source-to-use trace: semanticAreaBaseConcept is the source cue, semanticArea is the selected Part-F row or bounded row-set, semanticAreaSenseFamily prevents theme-level overgeneralization, and ontologicalNeighborhood is the applicability neighborhood used to recover the subject kind, relations, and subject patterns. These are coordinates of the wording repair under E.8 and E.10.ARCH. They are not components or identity criteria of U.Ontic; a subject discovered directly through engineering work does not need them.
The defining ClaimGraph located at E.24 states the admission conditions for U.Ontic, and E.24 gives practical guidance for the decision. The ontology unit, its identity rule, selected direct relations, declarations, claim-bearing epistemes, publication occurrences and forms, carriers, views, and representations remain distinct. Self-use does not establish a U.MethodDescription; apply A.3.1 and A.3.2 only when a separate Method and MethodDescription claim matters.
Shared E.24-Family Settlement and Atomic Co-decision
E.24 and E.24.UK use one settlement schema without changing its requirements. MinimalGovernedRelationSet means the smallest independently defined direct-relation rules needed by named dependent use. It does not require one universal head relation. IdentityBearingDirectRelationIfSelected is filled only when an exact identity assertion under its defining ClaimGraph selects such a relation; otherwise it is explicitly none.
When the E.24 ontology-disposition question is current, fill exactly one branch field inside OntologyDispositionResult; the decision EntityOfConcern remains fixed and the selected payload stays in that field. A changed result changes the decision's ClaimGraph and therefore identifies another decision episteme under C.2.1; state any edition continuity explicitly.
In ontic-only, cite the already accepted U-kind result consumed by the ontology-disposition decision and omit a new UKindAdmissionResult. A direct-use or bounded-coordination conclusion returns no new OnticSettlementResult.
In U-kind-only, either cite an accepted ontic settlement needed by the use, or explicitly conclude that no additional durable ontic is needed and record the sufficient direct-use or bounded-local-episteme disposition. The positive admission still requires every E.24.UK membership, extent, witness, need, non-duplication and branch-specific condition. Its result does not require creating a sibling ontic.
Use atomic ontic-plus-U-kind only when both new outputs are independently needed. Evaluate them from common candidate inputs; neither is prior evidence for the other, and they remain provisional together while either required branch is unresolved.
If one branch fails, preserve independently valid existing objects and return the exact gap. If the tests instead establish that no additional ontic is needed, select the explicit U-kind-only branch with its direct-use or bounded-coordination grounds; do not call an empty ontology unit connected or drop a required dependency.
The bootstrap co-decision is E24-CO-UONTIC-BOOT-01. Its EntityOfConcern is the exact source-construct entity defined by E.24:4 for the kind U.Ontic; it does not presuppose an admitted U.Ontic or a pre-existing ontic instance.
From that common input it returns two distinct accepted outputs: E24-OS-UONTIC-BOOT-01, which accepts this shared settlement schema as the direct rule for identifying future ontology-unit individuals, and E24UK-AR-UONTIC-BOOT-01, which admits the root kind U.Ontic.
The schema, pattern, decision episteme, and kind are not thereby instances of U.Ontic; each concrete ontology-unit individual still needs an ordinary OnticSettlementResult. No relation-about-relation or relation from the kind to itself is invented for the bootstrap.
E.24 is compatible with modular ontology and ontology-design-pattern practice: modular ontology libraries and ontology design patterns show why reusable small ontology structures matter, and recent process-modeling work reports loss of reuse when process patterns remain implicit. E.24 is narrower and more FPF-specific: it governs the decision whether FPF should introduce a durable action-facing ontic, rather than importing an external microtheory or treating every reusable repair table as ontology.
If the three resolved ontology dispositions need reusable comparison, use A.19.ECS to construct the evaluation CharacteristicSpace: retain the current subject assertions and relations, add one bounded local episteme for a declared use, or add a durable ontic with its own rule content.
The A.19.ECS locator establishes neither a Method nor a MethodDescription; apply A.3.1 and A.3.2 only if those identities matter. E.24 supplies the candidate dispositions and their ontic constraints, while characteristic selection and evaluation remain separate A.19.ECS assertions.
Source-use status remains an independent provenance choice, not a fourth candidate, and a comparison result does not establish ontic identity.
Within this split, the rule content located at E.24 states the distinction among the ontic, the claim-bearing decision episteme, reusable declarations, and publication-side objects, plus the ontic-introduction decision needed before dependent uses rely on a durable ontic. Publication-section rules, adequacy scales, wording-use restoration rules, and evaluation of the resulting FPF pattern-set structures remain separate exact assertions whose ClaimGraph sources are located through the neighboring patterns named above.
Use the current split this way:
- use
E.24forU.Onticidentity, the primary governed subject kind, exact identity or constitution rule, minimal governed relation set, subject patterns, named dependent-pattern reliance, and non-use boundary; - use
E.24.CDwhen the current problem is detecting and characterizing an apparent subject before deciding whether it should enter an E.24 ontic-introduction decision at all;E.24.CDsupplies detection and characterization only and selects no E.24 disposition.Local use frameis not an E.24 disposition: recover whether the payload needs direct subject-assertion use, a bounded local episteme under C.2.1, a durable ontic, or an unresolved stop; record any source-use status separately. - use
E.24.PUBwhen the current problem requires distinguishing the ontic, an ontic-description episteme, exact E/A/U/X/C publication, reusable form conformance or independently identified carrier; useE.17.0forU.Viewmembership, A.6.3 for source-to-receiving construction, A.6.3.RT/A.6.3.RT.OE for expression construction and C.29 for a mathematical lens; - use
A.19.ECSonly when the contested question is how to construct an evaluationCharacteristicSpacefor comparing the resulting FPF pattern-set structures after retaining the subject-pattern relations, adding one bounded local episteme whose claims cite them for a declared use, or adding a durable ontic and its subject pattern.
Questions about candidate detection, publication discipline, and contested evaluation remain separate exact subject assertions under their own defining or constraining ClaimGraph sources. E.24 governs the ontic-introduction decision.
Introduce or rely on a durable FPF ontic only after the ontic-introduction decision satisfies four checks.
Check 1: Existing Rule-Content Check
Name the current claim under decision and ask whether an existing exact defining or constraining ClaimGraph already states its rule content.
Use existing rule content first. If the case is method semantics, resolve the defining ClaimGraph located at A.3.1; if it is method description, use A.3.2; if it is mechanism meaning, use A.6.1 and E.20; if it is work planning or dated work, use A.15.2 or A.15.1.
For evidence, gate, source, assurance, decision, release, publication, or another case, name the exact current subject assertion and its defining or constraining ClaimGraph, with the pattern id only as locator, before selecting direct rule-content use. If no current rule content can be recovered by value, that disposition is unavailable; use the other E.24 dispositions rather than treating the topic word as authority.
Do not introduce a durable ontic only because several patterns are near each other or because one source word appears often.
For a candidate relation kind, recover the exact participants and test the current direct relations against their exact predicates, assertions, and defining ClaimGraph sources. If one direct relation closes the named dependent-use claim, use that settlement and stop.
If none closes it, A.6.RCD may derive the needed claim and return a local-claim, predicate-definition, derived-kind-candidate, or primitive-kind-candidate disposition. A local compound claim or reusable predicate-definition episteme is not a relation kind.
A derived-kind candidate proceeds only with a proposed direct subject settlement of its base dependencies, obtaining, applicability, and occurrence identity; a primitive candidate proceeds only with a candidate standalone defining ClaimGraph that supplies its own obtaining and occurrence identity. E.24 uses the direct settlement or A.6.RCD result and does not repeat the derivation method.
Check 2: Stable Identity Test
A candidate qualifies as a durable ontic only when it has stable identity beyond one local wording issue, source expression, or bounded local episteme used for first explanation.
Ask:
- What exact candidate entity, proposal episteme, or source-construct entity is independently identifiable before judgment, and what direct rule identifies it as the fixed EntityOfConcern of the decision episteme?
- Which exact result payload follows from each available disposition without replacing that fixed EntityOfConcern: closing assertions, a bounded episteme, a selected ontology-unit individual, or an unresolved reason?
- What changes the identity of that ontic?
- What does not change ontic identity, even if an ontic-description episteme, publication form, notation, view, or presentation carrier changes?
- Which direct world-side relations and grounding conditions are required for identity?
- Which dependent patterns may rely on that identity?
If durable identity is not established, use an existing assertion or a bounded local episteme under C.2.1 when it meets its threshold in §4.3. Otherwise return the unresolved question about the identified decision subject. If no decision subject can be identified, retain the inquiry material.
Test the invariant against the subject before filling a relation field:
Check 3: Direct Relation and Declaration Test
An ontic-introduction decision identifies each direct relation needed by the selected use before it introduces reusable SlotSpecs in a separate RelationSignature episteme. It singles out one identity-bearing relation only when the subject pattern does.
Choose the branch with three observable thresholds before opening the ontology object map:
- Direct use closes the case when one readable claim under current subject patterns gives the named receiving use what it needs. Point to that claim and stop; do not add a coordination episteme or ontic.
- A bounded local episteme is needed when the named use requires several independently governed claims about one identified subject to be read together. Reuse an adequate account or identify the bounded episteme under C.2.1. Check the dependencies that could change this choice. Each claim retains its direct governor.
- A durable ontic is needed when multiple current patterns require the same independently identified ontology unit and would otherwise duplicate or disagree about its identity, constitution or necessary relations. Check existing settlements first. Repeated citations or shared words alone give no such reason.
A crew choosing a pump instruction may need the equipment identity, the revision in force and the crew’s permission together. Those claims can settle the crew’s decision. If a known consumer may require their package as a reusable ontology unit, identify the dependence and resolve the durable question to the extent the evidence permits. An unknown unrelated use leaves the crew’s established result intact. When a missing fact could change the disposition, state that fact and the scope left open. Source provenance has its own status under §4.0 on every branch.
Use this short card to keep the effort of first use low. State the recognizable situation, the use that must close, and the exact subject; run the three thresholds; then fill ontologyDispositionResult last. Work and decision are examples of receiving use, alongside comparison, preservation, teaching, publication, and reference use.
Treat a filled card as the decision episteme only when its claim content, fixed pre-judgment decisionEntityOfConcern, and effective ReferenceScheme are recoverable under C.2.1. The branch payload is a result about that candidate, not the candidate's replacement. If the result later changes, identify the changed ClaimGraph as another decision episteme and state any edition continuity separately.
A working phrase, topic cluster, draft heading, or list is not that subject. If no candidate entity, proposal episteme, or source construct is independently recoverable, the card remains an inquiry prompt.
Every candidate receives one truthful branch result. Ordinary direct, bounded, and unresolved cases stop at this card: direct use needs only its current closing assertion, and bounded use adds only the C.2.1 coordination required by that use. Omit source, publication, view, representation, Work, U-kind, and other neighboring-object fields when no such claim is current; absence is enough and needs no blank or not current value.
Authoritative Typed Object Map
Open only the rows whose selection question is true for the chosen branch. Later sections refer back to this map.
Before opening the full OnticIntroductionDecision form, run two checks.
First, state the subject's identity, constitution, or recognition rule and the smallest relation set the named dependent use needs. For every included direct relation, write one readable sentence naming its participants and predicate; mark it identity-bearing only when its subject pattern does. Only then declare SlotKind, ValueKind, and refMode under A.6.5 for a relation whose typed reuse is current; when refMode is a RefKind, name that declared RefKind.
Second, treat bare role as an E.10.ROLE trigger and ask only whether the current ontic decision has confused a world-side participant, a local system-role kind and its A.2/C.3.2 classification, an exact A.2.1 assignment occurrence, or a declaration-local participant meaning in an A.6.5 reusable declaration. Keep the participant under its direct subject pattern and use F.6 only when Work attribution is current. E.24 does not reconstruct any assignment signature or occurrence rule, and bare role supplies no common head for these objects.
When an encountered card, table, schema, diagram, or record is current, apply the selection question in E.24:4.3a to each proposed use. Visible shape and field co-occurrence identify no episteme, publication object, representation, relation kind, or obtaining occurrence. Only an identified U.System performs description, rendering, or publication work.
Introducing an ontic organizes kinds, direct relation rules, declarations, and named dependent-pattern reliance in FPF. It does not create or individuate any project-side relation occurrence. For each such occurrence, apply the direct predicate and domain identity rule under A.6.REL. A designator may designate the already reidentified occurrence; a governed reference may resolve to it; an assertion or description episteme may carry a claim and designation about it. A publication occurrence instead makes one selected episteme edition available and neither designates nor creates the world-side occurrence.
Worked durable-branch replay:
Use this replay only after the first-use thresholds select a durable ontic. Its pre-judgment subject is EpistemeOnticProposal_v1, identified under C.2.1 by EpistemeOnticProposalClaims_v1 about source construct E24-Episteme-Ontic-Candidate-v1 under FPF-Ontic-Proposal-Scheme-2026; it exists before and is not identical to the selected EpistemeOntic.
The replay applies the object map to a pump-maintenance specification. C.2.1 selects an identity-bearing constitution relation for the Episteme ontic; the named project triple is one witness. Other ontics use their own identity rule and need not imitate this relation shape.
Record a branch result for each candidate, but ordinary direct, bounded, and unresolved cases stop at the one-screen card. Open the full form only when dependent patterns will rely on a proposed durable ontic, the current claim changes admissible use, or a receiving use needs a replayable reason why bounded C.2.1 coordination was insufficient. The durable branch meets the first condition: multiple current patterns must rely on it.
You can use the following form to publish one claim-bearing ontic-introduction decision episteme. When a guard above opens it, include only rows activated by the selected branch and receiving use. Omit every inactive neighboring-object or assurance row; the labels are prompts, not fields that must be filled, and they are not world-side participants, SlotSpecs, or components of the selected ontic.
An ontic-settlement result does not by itself admit a public U.* kind; that requires the shared decision's separate UKindAdmissionResult. Record a one-screen branch result for every candidate; complete the full form only when one of its three guards is true, and then only by the rows that guard, branch, and receiving use activate.
When typed reuse needs a declaration of one selected direct relation, its RelationSignature uses A.6.5 and the E.24 decision defines no second slot discipline; the direct relation retains its exact predicate and defining ClaimGraph. A SlotKind names one participant meaning only inside the selected RelationSignature, and its ValueKind constrains the admitted kind of the actual participant corresponding to that SlotSpec. Neither the SlotKind label nor its wording decides that kind; an exact participant-kind assertion does.
Check 4: Exact Rule-Content and Dependent-Use Test
State:
- the pattern governing the selected durable ontic;
- the exact defining
ClaimGraphfor each relation in the minimal set, and which relation is identity-bearing when an exact subject assertion selects one; - each dependent pattern and the identified ontic identity, direct relation rule, or
RelationSignaturedeclaration on which it relies; - each draft ToC row, planned pattern label, or absent subject-pattern section that remains non-governing.
Naming, publication placement, and evaluation remain neighboring authoring work under F.18, E.8, E.9.DA, and E.21. The ontic-introduction decision may point to those next moves, but none establishes ontic identity or replaces the subject pattern.
If the decision selects a durable ontic, write the pattern that defines or constrains it before dependent patterns rely on it. If the decision selects only a bounded local episteme, identify that episteme under C.2.1 and state its non-governing bounded use and claims by value. If no pattern governing the proposed durable ontic is written, do not cite that candidate as governing current FPF use.
Recover broad rule-content, provision, and support wording before admission
Do not admit a generic governance, provision, support, or rule-locus ontic merely because several patterns use those words.
First freeze the exact source occurrence and recover what it says by value: an exact subject assertion, defining or constraining ClaimGraph, direct relation, Work occurrence, Method or MethodDescription, promise or commitment content, source use, publication, evidence, assurance, authority, access, or ordinary-language claim.
The recovered C.2.1 assertion is the preservation object for that occurrence; a shared dispatch table, field name, or word family is not.
A broad candidate fails the durable-ontic test when its proposed members have no common identity or membership rule and no non-duplicative receiving use. In that case E.24 supplies no umbrella object. Keep each narrow recovered meaning under its existing pattern and use E.24.UK for the associated U-kind question with an occurrence-local reject. The rejection's RejectedCandidateRecoveryRef must resolve the exact preserved assertion. If the occurrence has not yet been recovered, leave its material wording unchanged rather than deleting content under a lexical rule.
This rule blocks generic U.Provision, U.Support, SupportRelation, governance relation kinds or occurrences, and rule-locus description kinds unless a later, independently accepted case supplies exact individuals, identity, membership, non-members, and a receiver that cannot use existing assertions and relations. It does not block genuine service-provision Work, operational support Work, evidence support, source use, publication support, human or institutional authority, or another exact relation whose own predicate obtains.
Bounded Local Episteme Decision
Use a bounded local episteme when one application family needs a readable coordination of entities and direct relations that are already governed elsewhere, but no new durable ontology unit is justified.
A bounded local episteme is a U.Episteme identified under C.2.1, not a new U-kind.
Select one independently identified EntityOfConcern before writing claims. It may be a world-side entity or individuated occurrence under an exact subject assertion, an admitted collection-as-whole or selected U.Structure, or an identified source, expression, or pattern-set architecture object. The selection test is the same: every claim must concern that one object.
When claims concern separately identified subjects, keep the corresponding epistemes distinct unless one admitted whole or selected structure gives them a common subject. A recipient can still reason from several epistemes together. Establish how their claims refer to the relevant participants and conditions, and what follows for the original question. A list of subjects settles a request for that list. An account of their interaction needs the relation on which the conclusion depends.
For that bounded use:
- name the application concern, exact EntityOfConcern, and direct pattern that identifies it;
- state why every carried claim concerns that one object;
- identify each other governed entity and direct relation designated by those claims;
- cite the pattern governing each direct relation rather than restating its participant or identity rules;
- state an ontic overread only when its correction passes the F.19:4 test in E.24:4.0;
- reopen the durable test if a dependent pattern requires the identity, constitution or relations of the package itself.
Positive example. Pump37MaintenanceCoordination_v1 has Pump #37 as its EntityOfConcern. Its ClaimGraph may designate the current maintenance plan, dated work, enacted method, and direct relations because every claim explains how this pump is maintained for the named scheduling decision. Pump #37's A.1 identity is independent of the coordinating episteme.
Blocked example. The expression workflow points variously to a method, work plan, dated work, and transformation-flow structure, but no one identified entity, admitted collection-as-whole, or selected structure yet unifies those claims. Do not make the word or the four-item list an EntityOfConcern. Keep the source inquiry material and split any already valid direct claims until one exact subject is recovered.
Precision restoration may use a bounded episteme when one receiving use needs several mapped claims read together. The episteme coordinates those claims for that use; every referenced object and relation still uses the subject pattern named in E.24:4.3a.
Archetypal Grounding
These examples illustrate different ontic-introduction decisions. Their order does not prescribe a progression. Each slice shows the exact governed payload, its ontology disposition, any independent source-use status, and the tempting overread that is blocked.
Episteme Ontology Unit as Durable Ontic
The ontology-unit individual EpistemeOntic coordinates the shared episteme constitution and the minimal relation rules used across current patterns. E.24.UK retains U.Episteme as the primary kind; C.2.1 identifies each episteme. The ontology unit is identified under :4 by that kind, the exact constitution rule and the included relation meanings. The PumpStation37 specification is one consuming example. Changing the example's publication or view leaves the ontology unit unchanged; changing a governing relation in its minimal set reopens the ontology unit's identity.
The coordination is needed because these uses must agree on which episteme they concern. Empirical grounding selects claims of that episteme; edition continuity compares distinct episteme identities; viewpoint conformance tests an identified episteme against an identified viewpoint; publication makes a selected edition available. C.2.1's constitution rule provides their common subject identity, while each direct pattern retains its own predicate. A content or interpretation change therefore reopens the dependent conclusions that use it, whereas a carrier change alone leaves that identity fixed. Without this shared rule, those patterns could silently evaluate, continue, or publish different knowledge objects under one name.
The coordination uses independently recoverable content and interpretation rules. C.2.1:4.1.2 supplies bounded ranges for their typed substitution; :4.2.1 and A.6.5 govern the receiving declaration. The ontology unit contributes the shared connection among these uses. It grants no missing participant identity, range membership or obtaining relation. A proposed general public U-kind still needs E.24.UK.
Multi-Pattern Subject Matter as an Ontic-Candidate Archetype
A project phrase such as "algorithm", "process", "solver", "workflow", "system", "quality", "time", "source", or "architecture" can point to one recognizable subject that is spread across several FPF values and patterns. A shared phrase does not establish one kind for all these subjects. The E.24 decision settles the status of the cross-pattern subject before patterns rely on it.
In this archetype, "process" and "workflow" begin as source expressions. Recover the one current object through E.24:4.3a, select the ontology disposition from that object's receiving use, and retain source-use status independently. The boundary fixture below supplies the concrete direct, bounded, durable-threshold, and unresolved cases; the label never becomes their common kind.
A source-driven use closes only after the exact expression remains linked to what was carried forward. For example, workflow may remain quote-only while one receiving question is answered directly by a claim about a recovered U.MethodDescription under A.3.2, and another by a claim about a selected TransformationFlowStructure under E.18. If one use needs both assertions read together, apply the bounded-episteme threshold. The source expression neither becomes their common kind nor disappears from the provenance of that use. If a later claim needs the source's stronger ordering, execution, or evidence meaning, reopen the named source episteme and state the exact stronger assertion under its defining or constraining ClaimGraph.
One Workflow Expression Across the Boundary
Use one fixture to see what changes the answer. The exact expression Line 7 pump-service workflow comes from source episteme Line7MaintenanceManual_v4; when availability matters, Line7ManualRelease_2026-04 is its source publication occurrence.
Keep its source-use status quote-only in every case below. The fixed EntityOfConcern of each decision card is Line7WorkflowInquiry_v1, an independently identified source-construct entity for this inquiry; it asserts neither a workflow ontic nor any branch payload.
The recovered objects are already governed: Pump37ServiceMethodDescription_v2 under A.3.2, Pump37WeeklyMaintenancePlan_2026Q3 under A.15.2, dated Work occurrence Pump37ServiceWork_2026-07-18 under A.15.1, and selected Pump37MaintenanceFlowStructure under E.18. The expression and provenance do not choose the ontology disposition; the receiving use and recovered evidence do.
- Direct-use result. A manual editor needs to check the one claim that
Pump37ServiceMethodDescription_v2describes the service method used in the manual. A.3.2 closes that readable claim. The decision's EntityOfConcern remainsLine7WorkflowInquiry_v1; itsDirectUseResultpoints to the exact assertion, the identified method-description episteme, and A.3.2. Do not add a coordination episteme or a workflow ontic. - Bounded-episteme result. A weekly scheduling review needs the method description, current work plan, dated Work occurrence, and selected flow structure read together because each claim explains how Pump #37 will be serviced that week. The decision's EntityOfConcern remains
Line7WorkflowInquiry_v1; itsBoundedEpistemeResultpoints toPump37WorkflowScheduling_v1, a bounded episteme under C.2.1 whose EntityOfConcern is Pump #37. A.1 identifies the pump. For this scheduling decision the equipment, plan and work claims each retain their own governor. The known dependent uses give no reason to settle their package as a durable unit. Leave every designated object under its exact predicate, subject assertion, and defining or constrainingClaimGraph. - Durable-ontic threshold—not met by the current fixture. The decision's EntityOfConcern remains
Line7WorkflowInquiry_v1. A positiveDurableOnticResultwould have to point to an independently identified ontology-unit individual such asMaintenanceWorkflowOntic_v1, its stable identity or constitution rule, and the exact reliance of multiple A.3.2, A.15.2, A.15.1, and E.18 consumers on that one unit. Those facts are absent, so this fixture has no durable result; when this stronger use is the active question, itsUnresolvedResultnames the missing identity and reliance evidence. The source expression and recurring four-object list do not identify a durable ontic. - Unresolved stop. The same manual may ask only to “align the workflow” while leaving open whether the concern is the method description, plan, dated Work, flow structure, Pump #37, or an admitted whole or selected structure. The decision's EntityOfConcern remains
Line7WorkflowInquiry_v1; itsUnresolvedResultrecords that no exact governed payload has been recovered and names the missing identity evidence. Keep the quote and provenance, and split any direct claims that are already valid; do not turn the phrase or list into a subject.
The fact that changes the result is one closing claim, several claims for one use, shared cross-pattern ontology reliance with stable identity, or no exact governed subject.
The E.24 move is:
- name the working expression and identify the pre-judgment candidate entity, proposal episteme, or source-construct entity under its direct rule; keep that fixed object as the decision episteme's EntityOfConcern. If none is identifiable, retain inquiry material; a missing governed payload instead permits an unresolved result about the identified decision subject;
- list the direct entities and relations that currently carry the subject; for every reused declaration, separately list its RelationSignature and SlotSpecs;
- run the existing-rule-content, exact-identity, relation-or-constitution, dependent-use, and non-duplication tests, then select one ontology disposition for the recovered payload—direct subject-assertion use, bounded local episteme under C.2.1, durable ontic, or unresolved stop—and record source-use status separately;
- if a durable ontic is selected, write or cite its exact defining or constraining
ClaimGraphbefore dependent uses rely on it.
The workflow fixture needs its method description, plan, Work, and structure because each changes the scheduling use. Select the objects needed by another case through E.24:4.3a; their co-occurrence alone supplies no durable-ontic result.
For another broad head such as system, relation, or architecture, recover its exact subject assertion and defining or constraining ClaimGraph first and apply the same thresholds; the head alone admits no ontic.
Dependent pattern descriptions may keep a thin cue: when one recognizable concern spans several direct entities and relations, name the exact relation assertion and cite the pattern locator for its defining or constraining ClaimGraph. That cue does not license treating a local set of references as a durable ontic before the E.24 decision, assigning one entity to two kinds without direct admission, or treating a SlotKind label as alternate ontology.
Draft ToC Row or Planned Pattern Label as False Authority
A draft ToC row or older source label may name a calculus, family, or object before current FPF has an exact defining or constraining ClaimGraph for it. Such a label can guide investigation, but it cannot establish current FPF use.
Example: older source wording may name a method calculus before current pattern text contains exact rule content for it. If no current defining or constraining ClaimGraph states it, the label is not current FPF rule content. Use the exact ClaimGraph sources located at A.3.1 for method semantics, A.3.2 for method description, A.15.2 for work planning, A.15.1 for dated work, and B.1.5 for method composition when ordering is current. A separate method calculus can constrain other subject assertions only after it has its own E.24-style ontic decision, stable identity, named direct relation kinds with obtaining and occurrence-identity rules, and dependent-use declaration.
The same test applies to any draft ToC row or planned pattern label. If no current defining or constraining ClaimGraph states the label's meaning, do not cite it as ontology. Either cite current exact subject assertions and their pattern-description locators, keep the label as investigation context, or open an E.24 ontic-introduction decision.
Broad Terms That Hide Several Governed Objects
A broad head such as system, architecture, or change is a working expression, not current ontology. Recover one exact subject assertion and its defining or constraining ClaimGraph, run the three branch thresholds, and classify only current neighboring objects through E.24:4.3a. If the subject or dependent use is still missing, retain the inquiry and name that missing value before ontic admission.
Bias-Annotation
Scope: the authoring decision about one candidate ontology unit. It selects durable U.Ontic, direct subject-assertion use, bounded local U.Episteme, or unresolved stop as the ontology disposition; when source wording is current, it separately records quote-only, reduced use, or selected stronger source use. The scope does not include the subject matter defined or constrained by the resulting ClaimGraph.
This pattern favors explicit identity, direct relation rules, reusable declarations where needed, and subject-pattern reuse. It resists five recurring distortions:
- shadow-kind bias: repeated use of one bounded local episteme is mistaken for evidence that a new durable ontic exists;
- placement bias: a pattern nest or draft ToC row is mistaken for the governed subject kind or governing text;
- name bias: a cleaner term hides unresolved kinds, slots, and relations;
- semio-bias: discussion of description epistemes, publication occurrences, forms, carriers, or review evidence displaces the ontic or subject matter being introduced;
- process-bias: development-state, publication-state, evaluation-state, or process evidence status is copied into ontic or subject-matter content.
The mitigation is the same in each case: recover the primary governed subject kind, exact identity or constitution rule, minimal governed relation set, any identity-bearing direct relation selected, any required RelationSignature, and subject-pattern reuse before naming, placement, dependent-pattern reliance, or publication form starts governing the decision.
Conformance Checklist
Activation rule: every candidate uses the cheap-card checks CC-E24-1, CC-E24-1b, CC-E24-1c, and CC-E24-2. CC-E24-1a activates only when a direct-relation or declaration-layer claim is recorded; CC-E24-3 through CC-E24-5 activate for a durable result; CC-E24-7 activates for bounded coordination. The remaining rows activate only when their named object or claim is current.
A direct result closes from its current assertion and direct pattern. An inactive row is omitted, not filled with blanks or not current.
Common Anti-Patterns and How to Avoid Them
Consequences
- FPF can introduce rich ontology units without treating every bounded local episteme as a new durable ontic.
- Draft ToC rows and planned pattern labels stop acting like current subject patterns.
- Dependent patterns can rely on the ontic-subject pattern and its named direct-relation patterns instead of reconstructing those rules locally.
- Selecting a durable ontic has an ongoing maintenance consequence: a change to the primary governed subject kind, identity rule, or any relation needed by named dependent use reopens the ontic-introduction decision and may affect those patterns. The bounded-local-episteme and direct-use dispositions avoid that cost when no durable coordination is needed; retaining a source expression for quotation or reduced use changes provenance handling, not that ontology cost.
Rationale
FPF needs a pattern for ontic introduction because many important ontology units require one exact identity rule and several direct relation patterns to remain coherent. One record-shaped episteme or universal head relation cannot stand in for every nearby object. Give the ontic stable identity, state the smallest independently governed relation set needed by dependent use, single out an identity-bearing relation only when the subject pattern does, and add RelationSignature declarations only where dependent uses need them.
U.Episteme is the main stress case. C.2.1 identifies one episteme through claim content, exact EntityOfConcern, and effective reference scheme, while separate direct relations govern grounding and edition continuity. A RelationSignature declares reusable participant typing only when another use needs it. If a card is current, classify its actual use through E.24:4.3a; neither its layout nor its publication makes the episteme's claims true.
A bare role cue is a second stress case because it can hide four different objects: a world-side participant, a local system-role kind and its classification judgment, an assignment occurrence, or a reusable declaration's local participant meaning. E.10.ROLE recovers the intended use; the participant stays under its direct subject pattern, A.2 and C.3.2 govern the local kind and classification, A.2.1 governs the exact assignment species and occurrence, A.6.5 governs the declaration, and F.6 governs any current Work attribution.
E.24 asks only whether the ontic decision has confused those objects or invented a common head. Their participants, predicates, obtaining and occurrence identity remain with the cited patterns.
Without E.24, FPF ontology development oscillates between two bad moves. One move invents a new umbrella name and leaves the mixed ontology intact. The other refuses the new name but still leaves several patterns carrying duplicated local slot doctrine. E.24 gives a bounded ontology decision: use an existing subject pattern, introduce a durable ontic, state only the needed claims in a bounded local episteme under C.2.1, or stop unresolved. A separate source-use status preserves or strengthens the source relation without replacing that ontology decision.
E.24's rule content and practical guidance concern the introduction decision. They do not define every ontic or become a registry of systems, epistemes, methods, mechanisms, architectures, sources, qualities, times, dynamics, or changes.
If an E.24 episteme qualifies as a U.MethodDescription, A.3.1 and A.3.2 must identify the Method it describes and show substantive guidance for doing it; that gives no other pattern episteme the same membership.
Each accepted subject still needs its own defining or constraining ClaimGraph; a bounded local episteme may contain claims for one declared use but does not define the ontology.
SoTA-Echoing
E.24 does not claim to replace ontology engineering, OWL-style formal ontology, or UFO-style foundational ontology. Its governing reason is the current FPF need for action-facing ontology compactness, plus a narrow SoTA echo:
The SKOS and OWL baseline bounds taxonomy-only use in E.24:4.1 and E.24:5.4; modular ontology patterns support the reusable ontic and subject-pattern move in E.24:4.3 and E.24:4.4; interoperability work supports the stable-identity and currentness tests; process-representation work disciplines the workflow case in E.24:5.2; and gUFO supplies a stress comparator for recovering the exact object behind role without importing or repeating its ontology.
These approaches offer different ways to represent knowledge. A taxonomy can answer a question about inclusion. Where a recipient needs to relate identities, participation or dependencies, compare the direct rules and the additional coordination a reusable unit would supply. Judge the cost and gain for the uses that will rely on it. An external pattern can contribute a distinction or a construction. E.24’s own admission test determines whether FPF needs a durable ontic for that work.
Use external sources when one ontic or subject matter itself depends on a source tradition. Put that source decision in the DRR and in the pattern description containing the exact rule content for that subject matter. Do not make E.24 contain a borrowed external theory of every durable ontic.
Currentness and Lowering Logic
Treat E.24 as current for ontic-introduction decisions while the subject patterns for relation-occurrence identity, reusable relation declarations, episteme identity, U-kind admission, wording-use restoration, and durable naming preserve the boundaries used here. Reopen one subject's ontic-introduction decision when one of these changes governs that subject:
- a new accepted FPF pattern changes direct relation identity, SlotSpec discipline,
EntityOfConcerndiscipline, U-kind admission, or durable-name discipline; - a bounded local episteme begins to be cited as if it governed a durable ontic;
- a planned pattern label acquires current subject pattern text and changes the ontic-introduction decision;
- dependent patterns start copying direct-relation rules or
RelationSignaturedeclarations instead of relying on their subject patterns; - external source work governs the introduction method itself rather than one selected ontic or subject matter.
Do not let an unresolved ontology disposition constrain dependent use. Use E.24:4.1 until the payload is selected for direct subject-assertion use, a bounded local episteme, or a durable ontic, or is explicitly stopped unresolved. Record source-use status independently: quote-only, reduced use, or stronger source use does not settle the payload's kind, identity, relation set, dependent-use reliance, or non-use boundary.
Relations
- Builds on:
A.6.RELfor direct relation occurrence identity, each direct relation pattern for relation-participant meanings, obtaining, applicability, and occurrence identity, andA.6.RCDfor a residual needed claim or a derived-or-primitive candidate with its proposed direct subject settlement;A.6.0andA.6.5govern reusableRelationSignatureandSlotSpecdeclarations, andC.2.1governs decision, assertion, predicate-definition, and description epistemes. - Coordinates with:
E.8for pattern publication placement,E.10andE.10.ARCHfor wording-use restoration, andF.18for durable naming after ontology is settled. - Coordinates with:
E.24.CDfor candidate detection before the ontic-introduction decision,E.24.UKfor theUKindAdmissionResultoutput of the one sharedE24FamilySettlementDecision, andE.24.PUBfor ontic-description and publication distinctions. When both a new ontic and a new public U-kind are needed, E.24 and E.24.UK consume the same atomic decision inputs and neither treats the other's output as prior evidence. - Coordinates with:
E.17.0forU.Viewmembership, A.6.3 for optional viewing construction,C.29for mathematical representation, C.13 for a needed inspectable construction account, and E.14/B.3.5 for elected or independently required construction-assurance coverage. Each ontic-introduction decision names any additional subject-specific subject patterns instead of treating this relation list as a registry. - Coordinates with:
A.19.ECSfor contested comparison of candidate dispositions,E.9andE.9.DAfor recording and evaluating the authoring decision, andE.21for evaluating the resulting pattern. Those evaluation results do not become part of the selected ontic. - Used by: FPF authors when repeated relation and declaration material may need one durable ontic rather than subject pattern use or claims coordinated only inside a bounded local episteme.
E.24:End
Ontic Candidate Detection and First-Use Disposition
Type: Part E FPF authoring discipline pattern Status: Stable Normativity: Normative unless a section is marked informative
Use This When
Use this pattern when a recurring expression or visible form, such as a card or schema, appears to name a new FPF subject and the author must decide what to do next.
Typical moments:
- one word such as "process", "source", "quality", "architecture", "problem", "view", the unqualified word "role", "function", "mechanism", or "method" points to several FPF objects or claims at once;
- several patterns repeat a similar declaration, participant list, or relation rule;
- a project data structure looks concept-shaped, although it may be only a claim-bearing episteme, publication form, representation, or local record;
- a draft ToC row names a family that no current pattern yet governs;
- a proposed
U.*kind feels useful, but it may duplicate a current kind or direct relation.
Primary EntityOfConcern. When the author records this choice in a C.2.1 episteme, its EntityOfConcern is the subject already identified under a direct pattern. If that subject cannot yet be identified, an independently identified source episteme or expression entity can carry the open inquiry. Without an identifiable subject, keep the question attached to the blocked work or source material. The visible form and the note recording the disposition are not substitutes.
First useful move. State what the intended recipient must establish, explain, design or do. Identify which claim behind the expression would help them do it. Use §4.1 to settle that claim. For a question about several contributions, carry its answer into the decision that needs them together.
What goes wrong if missed. FPF grows shadow ontology. A table becomes a kind; a field label is mistaken for a relation-participant meaning; a filled field is treated as an actual relation participant merely because it occupies a column; a card becomes the subject; or a convenient word creates a second ontology over values and relations that already have subject patterns.
What this buys. The author identifies an applicable pattern or the fact still needed to choose one without filling a candidate record or maintaining a registry. A proposal for a durable ontic must pass E.24's full identity and relation test; simpler cases stop with their subject pattern, local classification, description or publication handling, wording repair, or a precise unresolved question.
Not this pattern when.
- If one existing subject pattern already states the needed claim, use it directly.
- If a local kind, criterion, candidate judgment, or extension is already the question, use
C.3,C.3.1, andC.3.2. - If the question concerns a description episteme, use
C.2.1for its identity and the subject-specific description pattern. UseE.17.0for view membership andE.24.PUBfor publication form, occurrence and carrier. Representation correspondence stays with its applicable subject or representation rule; useA.6.3.RTfor a representation transition andC.29for an unresolved mathematical representation, transfer or reliance question. - If the subject and governing claim are clear and only the wording hides them, use
F.19for wording repair andE.10for any unresolved meaning. - If a durable ontic has already been selected, use
E.24; if a durable publicU.*kind is separately at issue, useE.24.UK. - If the work is comparing architecture alternatives, construct the evaluation through
A.19.ECS.
Problem Frame
An apparent ontology candidate usually arrives inside something visible: a label, form, record, diagram, source passage, or repeated field list. That visible thing may point to a real durable subject, but it may instead carry claims about several already governed objects, publish or represent them, classify them for one context, or merely use an imprecise word.
E.24.CD governs this first-use choice before E.24 opens. It neither admits a durable ontic nor creates a candidate object of its own.
Problem
Without an explicit first-use disposition:
- Publication forms become false subjects. A card, table, or schema receives ontology authority because it is visible.
- Local classification hardens into public ontology. A criterion useful in one context is treated as a durable FPF kind.
- Subject patterns are bypassed. Existing methods, work, relations, epistemes, structures, sources, and results are duplicated under a new head.
- Wording repair becomes ontology creation. A broad word is replaced with a new broad word while the actual subject and predicate remain hidden.
- Candidate work becomes a registry ritual. Authors fill fields or scores for possible ontics instead of deciding the current case.
Forces
Solution
Start from the work that is blocked, not from the shape of the source material.
Ask these four questions in order:
- What must the next person do or decide? Name the comparison, classification, publication, repair, decision, or other practical use.
- What is that use about? Name the subject, claim, or source expression without treating its card, row, filename, diagram, or field bundle as the answer.
- Which current pattern already governs the needed claim? Name the predicate or judgment that would let the work proceed.
- What follows for the original work? If the claim supplies the answer, give it. If it supplies one premise, establish what connects it to the other premises the recipient needs. A remaining gap can concern how the parts interact, which cases a model must distinguish, or an ordinary matter of fact. Select the disposition in §4.1 for the claim that is open.
Apply the first disposition that fits the case
Choose the first disposition that settles the claim you are considering. Several rows may supply different premises about the same undertaking. When the question depends on those premises together, establish how they fit and whether they suffice. The terms situation, incident, current configuration, operating <system>, and emergency are recognition cues, not kind names. Recover only the subjects, claims, and relations that the receiving work needs.
When a durable public U.* kind is also proposed, E.24.UK returns its separate admission result. If the ontic and kind are both new, use the atomic co-decision already defined by E.24 and E.24.UK; neither result proves the other.
Recover objects hidden by a visible form
For a project card, row, schema, or diagram, inspect only what the current work consumes:
- Which filled statements are claims, and what is each claim about?
- Which entities or non-entity values are independently identified under their direct patterns?
- Which direct predicates are asserted, what are their actual participants, and which independently established facts satisfy their obtaining conditions?
- Is the visible arrangement a publication form, a representation under an identified interpretation, a carrier or merely a local layout? Apply the relevant subject-description rule; use C.29 when a mathematical representation, transfer or reliance question remains unresolved.
- Does the work need local classification of a candidate, or only a claim about an already governed feature?
- For a suspected overread of the visible form, use
F.19's grounded-contribution test and state the correction that changes the needed claim or next action.
A field label is not a SlotSpec. A.6.5 governs the declaration: a reusable SlotSpec appears only inside a RelationSignature for an already recovered direct relation and only when a named later use needs that declaration. A row value is not an actual relation participant merely because it occupies a column.
Open E.24 when cross-pattern pressure is concrete
Open E.24 when these detection facts are recoverable:
- one blocked use or decision;
- one independently recoverable candidate entity, proposal episteme, or source-construct entity that carries the inquiry without presupposing a durable ontic;
- concrete duplication or disagreement pressure in more than one current pattern description;
- the named consumers that make shared coordination plausible; and
- why one obvious direct-pattern route does not already close the blocked use.
Transfer those facts to E.24. E.24 tests the complete identity or constitution rule, minimal direct-relation set, dependent reliance, non-duplication, practical gain, declared use, and its applicability limits.
A missing settlement condition may lead to E.24's unresolved result; a missing detection fact instead leaves the E.24.CD inquiry open. Detection therefore does not require the author to settle the candidate before opening the settlement pattern.
When the subject pattern clearly closes the case, stop there. Repeated words, several source forms, copied fields, or a useful schema can prompt inspection, but without a blocked use, an independently recoverable inquiry subject, concrete cross-pattern pressure, named consumers, and failure of an obvious direct route, they do not open E.24.
State one result and stop
A disposition for one claim can be reported in one sentence. If the person asked only which pattern governs it, naming the pattern and the decisive fact is sufficient. For a question about a practice, return what applying that pattern establishes for the practice, including a connection still to be made when the local result supplies one premise.
For
<work or decision>, apply<subject pattern>to<exact subject or claim>because<decisive fact>; next,<action or stop>.
When the available facts do not yet justify applying a pattern, say:
For
<work or decision>, leave<exact subject or claim question>unresolved because<missing subject, predicate, or subject pattern>; return when<missing fact becomes recoverable>.
Any denied reading must pass F.19's grounded-contribution test. Use a longer explanation only when another author must understand a disputed disposition.
Do not create an OnticCandidateCluster, candidate registry, scorecard, or mandatory disposition form. Continue at the applicable pattern or unresolved return; return to E.24.CD when the subject or practical use changes, or when a missing detection fact becomes available for the same inquiry.
Archetypal Grounding
A candidate that opens E.24
Consider a library under development in which “description”, “view”, “claim set”, and “publication” repeatedly point to a claim-bearing object used across many patterns. The named consumers need stable claim identity across description, evaluation, reference, and publication work, but their current rules leave conflicting identity claims. No obvious direct-pattern route closes that gap.
Those facts open E.24 to test the proposed shared identity and relation set.
FPF's episteme ontology illustrates the resulting division of responsibility: C.2.1 governs U.Episteme; E.17.0, E.24.PUB, F.18, and the applicable subject-description or representation pattern keep viewpoint conformance, publication, naming and representation separate. In current FPF, use those established rules directly when they close the case. Cards and files do not become the episteme.
Local cooling-pump classification
A maintenance team repeatedly asks whether Pump #14 counts as a cooling pump in plant slice S-14. Pump #14 and its flow, heat-transfer, and operating-state features already have direct governors. The needed outputs are a reusable local criterion and a candidate judgment, not a durable ontology unit.
Apply C.3.2. A KindSignature may declare the criterion for repeated use; the judgment can be true, false, or unknown; a current extension is materialized only for a named set-consuming use. A measurement supports a claim about Pump #14's features but does not create its membership.
Problem card
A ProblemCard@Context under C.22.2 is a problem-side episteme. It may carry a signal, hypothesis, forecast, scenario, anticipated-condition claim, affected-entity reference, evidence cue, constraint, proposed direction, assignment cue, source reference, or gate cue without creating an actual Problem.
An actual Problem is one obtaining ProblematicForRelation under C.22.PFR. A card may assert that predicate, but it may designate a current Problem occurrence only after C.22.PFR independently establishes the actual-condition relation, criterion-applicability relation, adverse truth, and occurrence identity.
Signals, hypotheses, forecasts, scenarios, anticipated conditions, and reviewable formulations remain under C.22.2 or their exact forecast, scenario, temporal, or causal governor.
For a repair decision, keep the affected entity, evidence-use relation, any current local system-role kind and classification judgment, any exact obtaining system-role assignment, any separately governed responsibility relation, source-use relation, and gate or decision claim under their subject patterns.
Apply E.18.1, E.23, and the exact Work, search, evaluation, or continuation pattern when repeated problematization or later action is current. Neither the card nor its acceptance or publication creates or ends an actual Problem.
Open E.24 only if a different reusable subject-identity or relation gap remains after these direct claims are recovered; do not rediscover the actual Problem as a new ontic.
Record-shaped false candidate
A project schema contains:
Use the schema as source material. Its fields do not establish an ontology. A proposal episteme, method, mechanism declaration, work plan, intended-work claim, performed-work occurrence, holder system, state claim, evidence item, result, affected referent, and source remain different objects. Recover only those that the meeting uses:
Not every row has every listed object, not every filled field is claim-bearing, and co-presence in one row does not constitute a larger subject. The filled row is one C.2.1 episteme only when one exact ClaimGraph forms a claim-bearing whole about one exact EntityOfConcern under one effective ReferenceScheme.
Otherwise keep the epistemes separate and state only the exact collection, publication, representation, or meeting-use relation that obtains.
The column arrangement is a publication form only when selected to express an identified episteme for the meeting. The form is not a U.ChangeItem, and its columns are not ontic slots. If the project later needs a local kind of records for a query, C.3.2 may classify those records as records.
If a later blocked use has an independently recoverable inquiry subject, named consumers and concrete cross-pattern duplication or disagreement that one obvious direct route cannot close, open E.24 to test the proposed shared subject. The schema's shape supplies none of those facts.
Current configuration around a holon
A maintenance review asks about “the current configuration around Pump #14.” Identify Pump #14 under its system governor, then recover only the characteristic or state claims, actual part relations, temporal phase, and other direct relations that the maintenance decision uses. If the work compares a possible configuration, identify the possible-state episteme and its direct state or configuration claims rather than asserting current actuality.
A separate C.2.1 description episteme may provide claim-bearing orientation. Its EntityOfConcern and any grounding holon in a separately current EpistemeEmpiricalGroundingRelation neither identify Pump #14 nor turn the surrounding claims into one world-side object. The holon and those current relations answer the question; their conjunction is not U.Situation.
Operating pump with connected parts
Pump #14 is operating while a sensor, valve, and controller are connected. Operating first cues a governed state claim; it does not establish U.Work or U.Transformation. Connectedness does not establish parthood. Identify the pump and connected entities, state the exact connection relations, and use A.14 only for part relations whose predicates obtain.
Add dated maintenance or control U.Work only after every precise performer has an A.13 core and A.15.1 independently identifies the Work's performance history, Method, time, containing System, and performers. Add F.6 only when the recovered situation account also needs precise assignment-bound attribution. A local system-role kind and its classification remain separate. A short situation-recovery sentence may omit identifiers its receiving use does not need.
If maintenance or control is merely intended, keep it as an A.15.2 WorkPlan or other modal claim; it creates neither Work nor assignment. Add an actual bounded change under A.3.4 only when its changed referent, boundary, conditions, and change facts obtain. No bundle of system, state, connection, work, and change becomes a situation entity.
Multi-party emergency
An emergency report mentions a leaking vessel, an overheated subsystem, a suppression system, and response teams. Start from the question the report must answer. Recover the participating Systems and actual changes it depends on. If the question concerns their interaction, identify what the change in one permits or prevents in another.
For every dated response claimed as U.Work, recover each precise performer's A.13 core and independently admit the occurrence under A.15.1 as stated in E.24.CD:5.6; add F.6 only when the current account also needs exact assignment-bound attribution. Keep any local system-role classification separate.
Keep an intended response as a plan or other modal content until it occurs. State temporal relations through their temporal patterns and a causal relation through C.28 only when that claim is current and supported.
Use a C.2.1 emergency-description episteme only when the receiving work needs claim-bearing orientation across those objects. The emergency word, the record, and the co-presence of several systems and works identify neither U.IncidentSituation nor another bundled whole. Stop decomposition once the response decision has the exact subjects and relations it needs.
Mathematical inconsistency under a declared formal substrate
Two specification epistemes state constraints that cannot both hold under one declared FormalSubstrate and applicability. Identify the exact claims or epistemes, name that formal substrate, and state the exact inconsistency or consequence relation under its direct formal governor. Use C.29 when using the formalism as a mathematical lens leaves a representation, transfer or reliance question unresolved.
The formal relation may guide a later decision or repair-work occurrence, but it establishes no project-world event, work, transformation, causal relation, adverse episode, actual Problem, or situation entity. Formal consequence is not causation.
Inconsistent descriptions do not make their world-side subjects inconsistent without a separately governed bridge claim. If the exact relation or substrate cannot be named, leave the formal claim unresolved rather than letting the word inconsistency stand for it.
Architecture diagram
An architecture diagram may carry claims about selected structures of one holon. If the diagram is selected as one claim-bearing whole, C.2.1 identifies that episteme.
The same episteme has U.View membership only when E.17.0 conformance obtains; its publication form and carrier use E.24.PUB; C.30.AD governs its architecture-description adequacy and correspondence to the independently recovered subject. Use A.6.3.RT if a representation transition is claimed, and C.29 for an unresolved mathematical representation, transfer or reliance question.
The diagram does not become the architecture, structure, or ontic by being visible. If the current work is to correct one architectural claim, apply the architecture and structure patterns directly.
Broad source word
A source says that a method “supports” production. If the author can recover a required-effect, method-use, work-enactment, capability, evidence-use, or other direct claim, apply its subject pattern. If the source word still compresses several claims, use E.10 and E.10.ARCH to retain it only with its bounded meaning or in quote-only or reduced use.
Do not open E.24 merely because support recurs, and do not invent SupportRelation as the candidate.
Score table and characteristic space
A score table can serve as the publication form of an evaluation-result episteme over a U.CharacteristicSpace, or it may be only a local report. Use A.19 when the characteristic space itself must be identified and A.19.ECS when the work is constructing the evaluation characteristics for a contested comparison. Use the table's adequate mathematical structure and operations directly. Use C.29 when a representation choice, transfer, or reliance question remains unresolved.
The table does not admit U.CharacteristicSpace by appearance and does not require another candidate ontology beside the current A.19 subject pattern.
Bias-Annotation
This pattern favors early recovery of the real subject and the blocked work. It resists:
- publication-form bias: treating a card, schema, table, or record as the subject matter;
- wording bias: treating a repeated word as a kind or relation decision;
- registry bias: collecting possible ontics instead of disposing the current case;
- scoring bias: rating a candidate before its subject, identity rule, direct relations, and practical use are known;
- semio-bias: discussing forms and labels while the governed subject and claim disappear.
Name the work, the subject and claim you can identify, and the applicable pattern or missing fact. State the disposition, next action and stop or return condition. Open E.24 from the detection facts in section 4.3; the identity rule and relation set remain its settlement question.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Positive consequences:
- authors reach a subject pattern from a recognizable work situation;
- durable ontics are proposed from an identity and reuse gap rather than from form or vocabulary;
- local classification, claim coordination, publication, representation, and wording remain cheaper dispositions;
- cards, records, tables, and schemas remain useful source material and detection cues without becoming ontology by appearance;
- no candidate registry or score ritual is added to routine authoring.
Costs:
- choosing a subject pattern requires identifying the subject and needed claim; an unresolved stop identifies what is missing;
- some attractive names are lowered to local kinds, source wording, epistemes, publication forms, or representations;
- a proposal for a durable ontic requires the full E.24 decision and, when current, a separate E.24.UK admission result.
Rationale
Ontic candidates rarely arrive as pure ontology. They appear through the forms people use: project tables, cards, schemas, diagrams, source packets, draft rows, examples, and repeated words. Those forms reveal working concerns, but they do not decide what exists, what relation obtains, or what FPF kind is needed.
The first-use disposition identifies where a particular claim can be settled. The original commission determines whether that identification finishes the work or supplies part of a wider explanation or design.
Keep direct claims in their direct patterns; use C.2.1 only when one exact ClaimGraph, one EntityOfConcern for those claims, and one effective ReferenceScheme constitute an episteme; and use C.3.2 for local classification.
Keep publication in E.24.PUB, subject-description correspondence with its applicable pattern, representation transition in A.6.3.RT and unresolved mathematical representation, transfer or reliance questions with C.29. Use C.2.P.DR to block action inferred from declarative form; resolve ambiguous wording through A.6.RSIR, A.6.F, A.6.P, or E.10; and name only an already governed value through F.18.
Open E.24 when the detection facts in section 4.3 make shared ontology a live question that those simpler applications do not close.
This order keeps the first move affordable and falsifiable. Another author can see which fact selected the applicable pattern or unresolved stop and what to do next. A list of candidate fields or scores would make the form look authoritative and invite optimization of the record instead of settlement of the subject.
SoTA-Echoing
Reopen this source comparison when a newer ontology-engineering or data-model-to-ontology source changes the selected criteria for reusable subject identity, minimal relation sets, bounded scope, validation, or source-form overread; do not reopen it merely because a new vocabulary, serialization, or KG tool appears.
Relations
- Builds on:
E.24for durable ontic settlement;E.24.UKfor separate public U-kind admission;C.2.1for exact episteme constitution;C.3,C.3.1, andC.3.2for local typed projection;E.24.PUBfor publication; the applicable subject-description pattern for representation correspondence,A.6.3.RTfor representation transition,C.29for unresolved mathematical representation, transfer or reliance andC.2.P.DRfor declarative-form overread;A.6.5forSlotSpecdeclaration and participant-designation discipline;A.6.RSIR,A.6.F,A.6.P,E.10, andE.10.ARCHfor bounded ambiguity repair; andF.18for naming after the governed value is settled. - Coordinates with:
A.6.RCDwhen the missing piece is a relation-bearing claim that no current direct predicate closes. A local compound claim or predicate-definition episteme is neither a relation kind nor an occurrence; only a separately justified kind candidate proceeds throughE.24andE.24.UK. It also coordinates withE.17.0for actual view membership;A.1,B.1,B.2,A.14, andC.13for a surviving constructed-whole question;A.3.4,A.15.1, the temporal patterns, andC.28for actual change, work, temporal, and causal claims;A.6.0andC.29for formal-substrate and mathematical-lens use;C.22.PFR,C.22.2,E.18.1, andE.23for actual Problem, problem-side formulation, and later problematization;A.19forU.CharacteristicSpace; andA.19.ECSfor evaluation-characteristic construction. - Used by: DRRs and authoring work that must decide whether a recurring construct uses an existing subject pattern, remains one or several bounded epistemes, becomes a local typed projection, applies description or publication handling, receives wording repair, opens E.24, or remains unresolved.
E.24.CD:End
Episteme Publication: Availability, Form and Carrier
Type: Part E FPF authoring discipline pattern Status: Stable Normativity: Normative unless a section is marked informative
Use This When
Use this pattern when an ontic or another entity is encountered through a card, table, diagram, file, pattern host, dashboard, or similar published expression and the current work depends on knowing what was described, what was made available, and what merely carries the expression.
Primary working reader. A practitioner or FPF author distinguishing a claim-bearing episteme, a U.View, its expression, a reusable publication form, a carrier or an exact publication occurrence. Open representation methods when the expression also supports reasoning through a correspondence to its subject.
First useful move. Ask what the reader needs to obtain and what they will do with it. For a publication being designed, respect the specified edition, audience and conditions, and use the reader’s work to choose the claims, expression and carrier where those choices remain open. For an existing publication, recover its declared use and the routes actually available. Match those conditions to the reader’s task when its suitability is at issue.
Name a reusable form when its constraints matter. Cite a separate plan, decision question or U.WorkPlan only when it independently exists and changes the publication claim. Distinguish availability from a claim of actual access or use. When a team needs access, decide who needs to do what, by when. Apply §4.2.1 when different routes or conditions could change that answer.
Open fuller relation detail when needed to resolve disputed availability, its boundary or publication-occurrence identity.
What goes wrong if missed. A visible layout is treated as the described subject, a file is treated as the claims it carries, a diagram is treated as a view merely because it is graphical, or publication availability alone is used to admit a durable U.EpistemePublication kind. The receiving work then cannot tell which object changed when claims, layout, carrier, audience, or use changes.
What this buys. The user can change a claim, view, form, carrier, audience, or declared bounded use without silently changing all the others. A publication can be inspected and repaired while the subject pattern remains centered on its subject.
Not this pattern when.
- Use
C.2.1when the question is the identity or content of the episteme. - Use
E.17.0when the question is whether an exact episteme conforms to an exact viewpoint episteme and therefore hasU.Viewmembership. UseA.6.3separately when source-to-receiving viewing construction is current. - Use
A.6.3.RTandA.6.3.RT.OEto construct or change an expression and recover its correspondence; useC.29when a mathematical lens and its inference or transfer conditions are current. - Use
E.24.CD, thenE.24, when a durable ontic is still being considered. - Use
E.24.UKwhen a publicU.*kind or dependent-kind disposition is unsettled. - Use the subject pattern directly when publication does not affect the receiving use.
Problem Frame
Ontics and other entities are usually encountered through expressions of epistemes and physical or digital carriers.
In a completed inspection card, the claims about the inspected subject can constitute a U.Episteme; the particular signs express those claims; reusable arrangement rules supply a publication form; and a file or sheet bears the expression. A publication occurrence makes the selected episteme available through that expression and carrier to a declared audience for a bounded use.
These relations connect separately identified objects. E.24.PUB defines exact publication through expression and bearing, and reusable-form conformance when a form is selected. It also supplies form-mediated claims and broader availability for uses that do not require one fixed expression. Ordinary publication can be stated in one sentence.
Plain published episteme names an episteme while it participates as the selected edition in a current publication occurrence. C.2.1 continues to govern that episteme's identity. If the receiving use needs classification by publication participation, use C.3.2 with its declared criterion and slice. E.24.UK:5.0 rejects the additional public kind U.EpistemePublication for the availability-only use; a different classification need follows its admission tree.
Here a publication occurrence is an occurrence of EpistemePublicationRelation: an availability relation that can endure. It is not the instantaneous rendering, printing, uploading, or access-control work that may establish or restore that availability.
Problem
When a reader sees only “the diagram was updated” or “the model was published”, it is hard to tell what changed. Five different changes may be hidden:
- the selected episteme edition may have changed because its claim content, EntityOfConcern, or effective reference scheme changed;
- another episteme edition may have been constructed, or the exact episteme may conform to a different viewpoint edition;
- the expression, reusable form or representation correspondence may have changed while the claims stayed the same;
- the presentation carrier or its availability may have changed;
- the declared audience or bounded use may have changed while the same episteme, expression and carrier remained.
Without the distinction, the receiving use cannot identify the smallest object or relation to inspect, revise, republish, or stop relying on.
Forces
Solution
Start with this readable publication statement:
Publication occurrence
<P>makes episteme<E>available to the audience identified by<A>for the use bounded by<U>, through expression<X>recoverable from carrier<C>.
Use that statement to identify a publication occurrence. When designing a publication, follow each needed claim to the point where its recipient will use it. If several claims are needed together, establish how the recipient can recover their connection and the conditions under which they apply. Their sources can remain separate. Test whether a permitted change of carrier, audience or situation could leave the declared publication available while the reader loses a distinction that affects the work. Retain the existing arrangement when it already supports the intended use.
When the selected expression follows reusable form F and that conformance matters, name F separately. For disputed availability, identity or change, recover the predicates below.
Identify the publication occurrence
EpistemePublicationRelation is the direct relation kind whose occurrence makes one selected episteme edition available to a declared audience for a declared bounded use.
Its actual participants are:
The U.Entity ranges leave X, F and C with their independently established domain identities. The positive predicates below determine expression, conformance and bearing. E.8 supplies FPF pattern-form rules, E.17 selects multi-view publication forms, and a domain publication pattern can supply another form.
Obtain X with an identity that answers the comparison. X is an independently identified sign construction: specified elements organized under the formation and equality rules of an expression domain. The domain tells a receiver which differences change that construction. A construction can be compared before anyone establishes which episteme it expresses. Two useful domains are:
- An ordered Unicode text has the declared codepoint sequence as its value. Changing a codepoint or its position changes X. Encoding the same sequence in UTF-8 or UTF-16 can preserve X while changing the stored bytes. Any normalization of line endings or Unicode variants must be part of the selected domain's equality or its decoding rule before comparison; it is not inferred from visual similarity.
- A keyed typed table has its column definitions, key-to-row mapping and typed cell values as its value. Keys are unique; missing, unknown and numeric zero have separately declared representations. Under an unordered-row domain, displaying rows in another order preserves X. A domain that includes row order instead gives a different X after that permutation. If the receiving use or selected F relies on order, the expression domain must retain it.
These are independently specified domains, not a closed taxonomy. Another notation supplies its own elements, organization, equality and formation conditions.
The chosen domain remains fixed in a comparison. A move between domains requires an explicit correspondence and cannot obtain identity merely by choosing a coarser equality after the change. Concrete copies of a construction do not multiply X when the domain gives them the same value; their carrier tokens can differ.
Obtain F as reusable rules. F is an independently identified collection of rules for constructing or arranging eligible expressions. Its identity is the interpreted rules, their application domains and imported dependencies. A layout-only F can leave subject claims unconstrained; another F can require fields, units, coverage or particular interpretations. The episteme that describes F retains its own C.2.1 identity.
An expression X conforms to F when it belongs to the applicable expression domain and satisfies each applicable rule and dependency. A rule referring to R uses C.2.1:4.1.1's distinction; a label such as format does not exclude a semantic requirement.
Obtain C through its own domain. C is a physical or digital entity from which X can actually be recovered under the stated conditions.
For a continuing file object, a write can preserve C while changing its state and X. For an immutable digital snapshot identified by its complete byte value, different bytes identify another C. For a physical sheet or copy token, another copy has another identity.
A route such as a URL is the designation or access path used to reach the chosen carrier; it does not determine that carrier's continuity. State which domain the current comparison uses before judging it.
A snapshot value alone establishes no actual availability. Its bearing and publication claims require a retrievable realization with a decoding path to X. When several copies realize the same snapshot value, a value-based availability claim can remain true while one copy disappears. A claim about which physical or managed file copy was supplied instead selects that copy as C. These choices answer different questions about the carrier.
The reusable declaration is:
These SlotKinds name participant meanings only inside this RelationSignature. They do not create five new U-kinds, and a card field with a similar label does not become one of these SlotSpecs.
The audience-declaration episteme identifies the audience criterion; it is not the audience and does not prove access by any particular system. A concrete system's access, reading, reliance, or later work is another direct relation or work occurrence. This lets one publication be available to every entity satisfying a stable criterion without inventing U.Audience or treating a changing set of readers as changing participants of the same publication occurrence.
EpistemePublicationRelation obtains while all of the following are true:
PublicationExpressionRelation(E,X,U)relates the selected episteme, exact expression and bounded-use declaration;PublicationExpressionBearingRelation(C,X)relates the exact carrier and that same expression;- entities admitted by the audience declaration can obtain X from C and recover the required content of E under the conditions stated by the bounded-use declaration;
- E, A, U, X and C remain the identified participants of this occurrence.
One occurrence is reidentified by those five fixed participants and their maximal continuous interval of availability. If availability obtains with any different participant, it is another publication occurrence. Demonstrated loss of availability followed by restoration yields a later occurrence.
Missing or stale evidence leaves current obtaining unresolved; it does not prove a gap.
Rendering, printing, uploading, indexing, or granting access are activities separate from the publication occurrence.
When a precise performed-Work claim is independently current, recover each actual performer through A.13 and let A.15.1 independently admit the dated Work; add F.6 only when that claim or its receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment. F.6 identifies neither assignment nor performer, and missing or failed F.6 alone leaves independently admitted Work intact.
Follow A.6.1 for an operation binding, C.11 for a ChoiceResult, or the exact access, reliance, or use relation without reproducing its test here. The activity and any result remain separate from the publication-relation participants.
Recover expression and bearing only when needed
PublicationExpressionRelation(E,X,U)obtains when interpreting X through its specified decoding conventions and E's effective R yields the claims of that selected E needed for U, with their operative participants, qualifications, modal force and asserted dependencies, and X supports the operations named by U. The recovered content must be traceable to E. If the bounded use requires only a selection, identify that selection and retain the conditions needed to use it; A.6.3's direct construction or loss rule governs any stronger claim of viewing or coarsening. A newly introduced claim belongs to the resulting episteme under C.2.1. It cannot be counted as content expressed from unchanged E.PublicationExpressionBearingRelation(C,X)obtains while an actual state or realization of C provides a recovery procedure that yields that exact X under the specified decoding conditions. For a digital value, the procedure reaches an actual accessible realization. The predicate retains the distinction between a continuing carrier and its changing state.PublicationFormConformanceRelation(X,F)obtains when the independently identified X satisfies the applicable rules, domains and dependencies of F. It establishes neither which E is expressed nor audience access.
Each of these occurrences is identified by its ordered participant values and maximal continuous interval of predicate truth. Missing evidence leaves obtaining unresolved; it does not itself establish an interruption.
Their reusable declarations are:
Form-mediated uses. PublicationFormExpressionRelation(E,F,U) obtains while at least one independently identified X conforms to F and satisfies PublicationExpressionRelation(E,X,U). At least one expression of this E for U must meet the selected form’s rules. The existence of a template alone is insufficient.
PublicationFormBearingRelation(C,F) obtains while C bears at least one independently identified X that conforms to F. It establishes the recoverability of a conforming construction; the expressed episteme and audience access require their own predicates.
Each of these form-mediated occurrences retains its fixed participants and maximal continuous interval of predicate truth. X is a witness for the quantified claim. Replacing X1 by X2 can preserve either occurrence when that predicate remains continuously true. Changing a filename or storage address does not by itself settle carrier identity; apply the carrier's direct identity and currentness pattern.
Their reusable declarations are:
For form-mediated availability through C, the same X must witness expression, bearing and F-conformance. An adequate X1 elsewhere and a different X2 on C do not establish that E is available through C. Recover the shared witness whenever the availability claim depends on both form-mediated relations.
Keep the granularity of the availability claim
If the undertaking is to provide access by a deadline, establish that each intended reader can use an allowed expression and carrier in time. To conclude that assigned reading was completed, establish that the assigned readers actually read the required material by the deadline. For continuous access, each reader must have such a route at every relevant time. Routes may differ between readers and may change during the interval. On each form-mediated route, the same X must express E, be borne by C and satisfy the required form. Keep any fixed edition, expression, form or carrier specified for the job. Delivery to a delegate can satisfy an undertaking to the group when its receiving rule grants that role. An entitlement for each member requires coverage of each member.
If everyone must use one carrier, identify a carrier accessible to all as required. Its accessible expressions can differ when the job permits. If the carrier must stay the same throughout, identify that same carrier across the interval. A single exclusive login can serve two readers in turn. It fails a requirement for simultaneous independent sessions. Joint reading of a display can suffice when the group needs one shared session.
For a joint log, identify the agreed record and its accepted starting version. Check who may enter or merge changes and whether the intended readers can see the agreed entries when needed, including how competing edits are resolved. A shared read-only file and disconnected personal annotations allow reading but leave the joint entries unavailable. Authorized replicas that reconcile by 18:00 can serve an end-of-shift log. A team that needs the changes at noon must establish visibility by noon. Require one physical log only if that is part of the job.
One established route suffices wherever it covers the requirement, even with an unknown backup. A failover guarantee with established starting conditions, capacity and transitions can cover the interval, including the switch. If a reader loses one route and a permitted alternative is unverified, coverage for that reader and time remains open. Established failure of every allowed route for that reader at that time refutes continuous access for everyone. Retain the readers, uses and times for which access is established.
Availability establishes an opportunity to act. If actual delivery, receipt, reading or amendment is commissioned, establish the specified event. For an agreed final log, verify the specified content in the accepted version. If a particular edit or author matters, check the amendment and any required attribution. These cases use the ordinary publication relations without a new kind.
Use progressive explicitness
Use the smallest statement that supports the current work:
- Ordinary use: name the selected episteme, audience, bounded use, expression and carrier in one sentence; add selected form conformance when needed.
- Changed-object use: identify the objects that changed and the relations those changes require you to re-evaluate.
- Contested availability: state the
EpistemePublicationRelationparticipants, obtaining evidence, and occurrence identity. - Contested expression: open
PublicationExpressionRelationand the applicable representation, preservation or loss pattern; inspect form conformance when the claim relies on it. - Contested carrier availability: open
PublicationExpressionBearingRelationplus the direct carrier-currentness or access pattern; preserve the fixed or quantified alternatives of the actual availability claim.
If a crew needs an operating limit from a cited instruction, an accessible route to the relevant passage or sufficient source-attributed content on its offline card can each support the decision. A locator can finish a locating task. Check that the needed condition can be recovered in time when the task is to decide how to operate.
Stop when the grounds settle the publication question you were asked. Identifying the object and relation can finish a classification. For the design of a publication, give the arrangement and the grounds on which it will serve the intended work. For a question about actual availability, establish the relevant route or coverage under §4.2.1.
Classify the encountered form by current use
Ask one question at a time:
The answers can be jointly positive because they concern different objects or relations. They do not follow from the words card, record, table, schema, diagram, view, file, or publication alone.
Keep direct verbs with their relations
- an episteme carries claims and designations;
- a
U.Viewis the same episteme individual for which E.17.0 conformance to at least one exact viewpoint episteme obtains; - an expression conveys the selected episteme's required claims for a bounded use;
- a reusable publication form constrains eligible expressions;
- a representation stands in an identified correspondence to its subject; C.29 governs the mathematical-lens branch;
- a carrier bears an expression through its actual state or realization;
- a publication occurrence makes one selected episteme edition available;
- a system may perform publication activity and may later access or rely on the published episteme, but those are separate claims under their direct patterns; publication availability establishes none of them;
A designator designates and a governed reference resolves to a referent. Neither operation publishes, bears, represents, or makes the subject-side predicate obtain.
Keep subject patterns subject-first
In a pattern about an ontic, structure, architecture, characteristic space, method, or another subject, explain the subject's identity, relations, practical problem, and solution before publication details. Add E.24.PUB only when the receiving use depends on distinguishing the description, selected edition, form, carrier, audience, or bounded use.
To use a publication as grounds for a conclusion about its subject, identify which claim the expression conveys and what makes that claim apply to the subject of the question. For a conclusion spanning scales, trace each required inference to the whole or constituents it concerns. The same published model can be used at several composition scales only where its assumptions and distinctions support the inference at those scales.
For example, an engineer checking the clearance of Pump #37 is asking about the pump. If the question becomes what its inspection card specifies, the card episteme becomes the subject. A verified reading of the card answers the second question. Using its value for the installed pump requires grounds that the measurement applies to that pump at the relevant time. A change of holonic scale requires a whole–part relation, such as the pump and its impeller. To compare levels of model abstraction, use B.5.MR on models of the same subject under the same inquiry and interpretation. If a summary drops a constraint, check which answers that change permits and whether the user relied on their exclusion. Publication of the card remains a separate relation.
Archetypal Grounding
Maintenance inspection card
A completed pump-inspection card states measured clearances and identified defects about Pump #37 under the maintenance reference scheme. Those organized claims constitute the card episteme E. The filled card supplies expression X; its reusable layout rules supply F. The PDF file is C under its selected file-object or snapshot identity.
One publication occurrence makes edition 4 of the card episteme available through X and C to the maintenance-planning team for planning the next repair.
Changing the PDF filename changes neither E nor necessarily C. Correcting a measured clearance changes the claim content and therefore E; a later-edition claim also needs C.2.1:4.5 continuity.
Replacing the layout changes F. If this changes X and the availability conditions hold for the new X, the resulting publication is a different occurrence. Broader availability of E can continue. Making the same E available to a supplier for quotation creates another exact publication because A or U changes.
Architecture diagram
An architecture diagram can express claims about selected structures of one holon. Those claims can constitute an architecture-description episteme.
When that episteme conforms to an architectural viewpoint under E.17.0, the same individual is a U.View; direct authoring and A.6.3 construction are independent construction routes.
Reusable graphical conventions can supply F; the constructed diagram supplies X. A correspondence connects its elements to the represented subject, with C.29 additionally governing any mathematical inference or transfer. A screen or sheet can supply C.
The diagram does not become the architecture by being published. C.30 governs the ArchitectureOf@Context claim and A.22 governs selected U.Structure values. E.24.PUB lets the architect locate a publication defect without replacing the architectural question with a discussion of diagrams.
Clinical procedure edition
A hospital procedure description is an episteme about how a procedure is performed. Treat it as a U.MethodDescription only when its EntityOfConcern is one independently admitted U.Method and its claims describe how that Method is carried out.
A wall poster expresses a selected edition for quick pre-procedure orientation; the laminated sheet is the carrier. A separate controlled publication makes the same edition available to clinicians for authoritative use during the procedure. The two publication occurrences differ in bounded use even if the words are identical.
Neither publication proves access, reliance, Method enactment, or clinical Work. If later clinical Work is independently claimed, use its direct Work pattern. Add the attribution pattern when the claim or receiving use needs precise assignment-bound attribution, as in §4.1. Keep the publication occurrence and every separately current access, reliance, assignment, Method, Work, or result claim distinct.
FPF pattern host
An E.24 pattern host contains an expression X of an ontic-description episteme about U.Ontic; its layout follows a reusable form F. The repository file is C. The selected episteme participates in an exact publication occurrence while X and C make it available to the declared FPF audience for the stated use. Changing C alone leaves the episteme's claim content, subject and interpretation unchanged.
Training availability and later choice work
One instruction edition is available to a training group for studying a method. That EpistemePublicationRelation occurrence establishes availability to the declared audience for that bounded use; it establishes neither that anyone read the instruction nor that adjustment, inspection, acceptance, or release work occurred. The same availability alone does not support an acceptance commission's choice about releasing one named lot.
If a commission later makes a release choice and that stronger claim is current, recover each actual choice-work performer through A.13 and let A.15.1 independently admit any dated choice Work.
For that choice Work, add F.6 only when the choice account or receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment; F.6 identifies neither assignment nor performer, and missing or failed F.6 alone leaves independently admitted choice Work intact.
Identify the resulting ChoiceResult separately under C.11. Keep both Work and result separate from the publication occurrence; the publication statement need not carry their identity, staffing, or omission rules.
When the later claim says that the published instruction was actually used, state that exact use under its direct relation, or under A.6.1 only when a declared operation application is current. If no such route is established, stop at publication availability and let the receiving pattern identify its own blocker. The ChoiceResult is neither the choice Work, the bounded-use declaration, nor a participant of the publication occurrence.
Bias Annotation
- Onto: direct relation occurrences and their actual participants remain primary; visible forms do not decide kinds.
- Epist: the selected edition, audience declaration, and bounded-use declaration retain C.2.1 identities.
- Semio: designation, expression, representation, bearing, and publication availability use different predicates.
- Arch: keep each subject claim with its applicable pattern; E.24.PUB defines only the neighboring publication architecture.
- Prag: progressive explicitness stops at the first statement sufficient for the receiving use.
- Did: a readable sentence can identify a publication. For a design or availability question, show how the reader can use the chosen expression and carrier under the conditions at issue.
Conformance Checklist
Common Misuses and Repairs
Consequences
The main gain is local repair. Replacing a stale carrier does not imply a change to the claims. A layout repair can preserve the model being presented. A revised claim can create another episteme edition without implying a change to the audience or form. A narrower audience or use can create another publication occurrence while preserving the edition.
The cost is that load-bearing publication claims need five identified participants and two supporting relations. Progressive explicitness contains that cost: the ordinary publication statement can be one sentence. Recover the fuller relation details when the receiving use needs them, including when availability, expression, or bearing is disputed.
Rationale
Publication does not change an episteme into a nested publication object. It is an availability relation supported by an expression relation and a bearing relation. That architecture explains why one encountered card, diagram, or file can matter in several ways without admitting one umbrella presentation kind.
The split also preserves agency. A System can render, upload, print, index, withdraw, or replace a carrier, but the publication occurrence is the enduring availability relation, not that activity.
The availability relation may obtain with no continuing publication Work, and an attempted publication activity can fail while no publication occurrence begins. If an actual Work claim matters, its direct patterns govern it; E.24.PUB only keeps it and its result outside the publication-relation participants.
In OWL, SKOS, RDF and triple-store implementations, one interface can expose labels, axioms, serializations, documents and queries together. Recover their functions for the intended use: an export operation by itself establishes neither the kind of the exported subject nor the truth of its claims.
SoTA-Echoing
The practical question is what an audience can obtain for a stated use, and which availability continues when the expression or carrier changes. Sections 4.1–4.2 use the following comparison.
For form-mediated availability, one and the same X must satisfy expression, bearing and form conformance. An adequate expression elsewhere and a different conforming expression on C do not establish that C makes E available. This shared witness is the binding that the two separate existential support claims leave open.
A change of X or C gives a different exact occurrence. The broader claim survives when the permitted routes still cover the promised work for every required reader and time. Loss of one reader’s only permitted route breaks a guarantee to everyone even while others can still read. For a claim about an actual delivery or edit, identify evidence of that event.
The cost of the exact comparison is selecting the expression domain and equality, carrier identity, and relevant time interval before comparing publications. A continuing managed file and an immutable snapshot answer different identity questions. For ordinary use, the readable publication sentence suffices; recover this detail when its alternatives can change the answer.
The following bounded comparisons support those choices. Their source models retain their own meanings.
The abstracts supply these bounded problem comparisons, not a validated performance ranking or the five-participant FPF predicate.
A.6.REL supplies the direct-relation and occurrence rules; C.2.1 supplies episteme identity. A.6.3, C.29, E.17 and E.24.UK retain their construction, mathematical-representation, view-membership and admission questions. Their combination here is the FPF construction justified by the publication questions above.
Reopen the choice when a real receiving use cannot distinguish the provided content under the declared identities, or when the broader-availability rule reports a gap despite continuous availability through its permitted alternatives. Reopen a source comparison when a newer result changes the distinction it supports. A different file format or storage tool alone supplies neither counterexample.
Relations
- Builds on:
A.6.RELfor direct obtaining and occurrence identity,C.2.1for the selected edition and declaration epistemes, andE.24for ontic-description boundaries. - Coordinates with:
E.17.0forU.Viewmembership andE.17for multi-view publication;A.6.3for optional viewing construction;A.6.3.RTandA.6.3.RT.OEfor representation construction;C.29for a mathematical lens;E.8for FPF pattern-form rules; and the direct carrier-currentness or access pattern when that question is current. - Coordinates with:
E.24.CDfor candidate detection andE.24.UKfor public U-kind and dependent-kind settlement. Its resultE24UK-NAR-EPUB-01rejectsU.EpistemePublicationfor the availability-only use; this pattern states that participation, while C.3.2 governs a receiving classification use. - Used by: subject patterns only when a receiving use depends on distinguishing the subject, description episteme, selected edition, view, representation, publication form, carrier, audience, bounded use, or publication occurrence.
E.24.PUB:End
U-kind Admission and Ontic Settlement
Type: Part E FPF authoring discipline pattern Status: Stable Normativity: Normative unless a section is marked informative
Use This When
Use this pattern when a public FPF expression proposes a U.*, type, kind, or subkind and the author must decide its status. When the grounds support a decision, the four choices are to reuse an admitted durable kind, declare a bounded C.3.2 local kind, admit a new durable kind that the use needs, or recover a non-kind object under the rule that defines or tests it. A title, filename, ToC row, table, or source spelling raises that question without settling it.
Typical moments:
- a direct relation family has stable occurrence identity and patterns for the next questions need one common kind for those occurrences;
- a proposed
U.*name appears in a pattern title, host filename, monolith heading, or ToC row; - a current pattern uses type, kind, or subkind wording and the governed object is unclear;
- a structural name looks useful for search, but may advertise a false root kind;
- a
RelationSignatureSlotKind, an assertion or description field, aC.29representation element, or anE.24.PUBreusable form has acquired aU.*spelling; - a single E.24 ontic settlement appears to govern one root U-kind plus several dependent durable U-kinds.
Primary EntityOfConcern. Identify the object the admission decision is about before filling the shared E.24-family decision: an already recoverable C.3 U.Kind, the proposal episteme for an unadmitted distinction, or the source-construct entity being translated.
Put the proposed criterion, candidate individuals, intended extent and non-member boundary, spelling, and dependent claims in the decision's ClaimGraph. If no decision subject is identifiable, keep the inquiry open. An extension, member list, rule bundle, title, or spelling cannot fill this position.
Primary working reader. The first reader is an FPF pattern author or reviewer deciding whether a public FPF name should remain U.*. The downstream reader is the practitioner who uses public pattern titles, headings, ToC rows, and names as orientation cues and needs those cues to point to the real governed object.
First useful move. Establish which distinction the practitioner needs and what turns on classifying the proposed individuals. Identify the local kind, proposal episteme or source construct to which the admission question applies. An inquiry with no identifiable subject remains open.
Then recover the proposed governed individuals, identity or membership rule, intended extent and non-member boundary, and the action-facing claim that needs the kind. Test whether existing U-kinds, direct relations, declaration SlotKinds, C.3 local kinds, or selected structures already preserve that distinction. Judge the public spelling only after the admission disposition is stable.
What goes wrong if missed. Treating a U.* spelling as admission creates a second ontology. A slot label is treated as a kind, a publication form as an ontic, and type and kind wording can compete with the settled ontology. A searchable title can survive even if it names the wrong object.
What this buys. Readers can trace an admitted public U.* name to its governing rule. A new public U-kind must pass the explicit test for its root or dependent form. If an existing rule covers the distinction, its PatternID serves only to locate that rule; no umbrella kind is introduced. If the grounds settle neither branch, the inquiry stays open.
Not this pattern when.
- If the question is whether FPF needs a durable ontic at all, use
E.24. - If the question is only detecting an ontic candidate before the durable decision, use
E.24.CD. - If the question is the difference among an ontic, its description episteme, publication, and publication form, use
E.24.PUB. - If the question is wording with no durable name pressure, use
F.19; openE.10,E.10.ARCH, or the direct subject pattern only for meaning that remains unresolved. - If the current governed object is already recovered and only its public label must be chosen, use
F.8,F.5,F.18, orF.17according to the naming use.
Problem Frame
FPF reserves U.* names for admitted durable U-kinds. Current source material and older corpus passages can still place that spelling on a declaration-local SlotKind, participant designation, selected structure, publication form, representation element, or unsettled candidate. The spelling is therefore evidence of admission pressure, not evidence of admission.
Section 4.2 separates exact accepted admission-result references from open prerequisites, blocked candidates, and non-admission exits. A public spelling, owner citation, or orientation row supplies no admission by itself. Existing root and same-individual-dependent kinds remain usable only through the accepted result reference recorded there. U.Capability is rejected for the recovered A.2.2 need because qualified holder-ability claims require no additional kind. An assertion cannot settle an unresolved prerequisite kind.
E.24.UK governs the following distinctions. A world-side relation participant keeps its independently governed kind. A RelationSignature SlotKind stays declaration-local. An assertion-side designation stays in its claim-bearing episteme. A publication form or C.29 representation keeps its direct use.
E.24.UK is an E.24 subpattern because kind admission and ontology disposition share one decision basis. A justified kind admission does not require an additional durable ontic; E.24 separately tests whether such coordination is needed. E.24 remains the head pattern for U.Ontic and ontic introduction. E.24.UK governs the detailed U-kind admission rules.
Problem
Without this pattern:
U.*spelling substitutes for admission. A public name is retained because it looks like a kind.- Unsettled type and kind wording competes with U-kind admission rules. Type, kind, subkind, Concept-Set rows, U-kind names, and E.24 ontics become overlapping ontologies.
- A dependent distinction becomes an independent root. A kind whose individuals retain root identity or depend on one root-kind individual is treated as if it had an independent root settlement.
- Structural names over-admit. A title, filename, heading, ToC row, bounded-context label, system, team, subsystem, view, diagram, publication, or named use is treated as if it created a base
U.Structureidentity or specialization membership. - Declaration and representation elements become U-kinds. A participant meaning in a direct relation, a SlotKind in its reusable declaration, an assertion field, or a
C.29representation element receives aU.*spelling even though its governing object is already known. - Naming patterns are asked to do ontology. F.5, F.8, F.18, or F.17 is used before the governed object has been recovered.
Forces
Solution
Durable U-kind admission is a claim-bearing decision about one identified entity. The decision is neither a relation between a public name and a settlement nor a bundle of future members, rules, boundaries and uses. Select the decision's EntityOfConcern by the entry rule above; keep the proposed kind criterion, extent, spelling, and use-enabling claims in its ClaimGraph.
Record the decision in a DRR or another claim-bearing episteme under E.9; the decision creates no project-side U.Relation occurrence.
Use the one E24FamilySettlementDecision in E.24:4.0a. The table locates its U-kind fields. They belong to that decision and follow all its requirements.
A short view is valid only when every displayed answer resolves back to that one shared decision. An inactive field may disappear from the view; a required field may not disappear from the decision.
The UKindAdmissionResultRef identifies the result, not the decision episteme. In an atomic decision, both results remain provisional while a required branch is unresolved. Neither supplies evidence for the other.
The shared decision selects exactly one positive form—root, same-individual-dependent, or identity-dependent—or one non-admission exit—reuse, local-kind, or reject. Every positive result cites its durable membership rule and scheme. Same-individual dependence also states the root and the implication to root membership for the same individual. Identity dependence instead cites an already governed relation to a distinct root-kind individual plus every discriminator. The three non-admission exits cite the exact reused kind, local C.3.2 declaration, or recovered non-kind object.
A public Tech label follows the accepted result through F.18 and F.17. Spelling improves retrieval but supplies neither membership nor extent. The decision, its output result, the proposed or admitted kind, and individuals classified by that kind remain different objects.
Positive Test For A Durable U-kind
Test a proposed new durable U-kind against these eight conditions. It may receive root, same-individual-dependent, or identity-dependent only if all eight hold:
-
Governed individuals. The candidate classifies independently identified governed individuals. Source expressions, declaration fields, table columns, reference suffixes, publication forms, and mathematical representation elements supply no such identity or kind admission merely by labeling or representing them.
-
Stable identity or membership. Cite an identity, grounding, recognition, or membership rule that reidentifies individuals and determines whether they enter the intended extent. The rule can stay fixed while an individual enters or leaves its extension as its state or the declared slice changes (C.3.2).
-
Reviewable witness. Cite the direct operational test. For a relation-kind candidate, cite the pattern passage that defines the relation; that passage must state participant meanings, obtaining, applicability, and occurrence identity. If no current direct relation closes the claim, an
A.6.RCDapplication may record a derived or primitive candidate with a proposed direct subject settlement; its local-claim and predicate-definition exits are not kind witnesses. Every other candidate cites its direct constructive, classificatory, or membership test. A signature, row, declaration, or mathematical trace counts only when its declaration or defining rule states the correspondence to the governed individuals. -
Action-facing need. FPF users need to state, compare, constrain, transform, or otherwise reason about those individuals under this kind; a wording preference alone does not qualify.
-
Non-duplication. Existing U-kinds, direct relations, declaration SlotKinds, local C.3 kinds, and selected structures cannot preserve the needed distinction without this durable kind.
-
Defining locus. One primary rule passage or accepted governed source set states the kind's identity or membership, intended extent, admissible use, and non-use boundary.
-
Shared E.24-family settlement. Fill
E.24:4.0awith the subject kind and identity rule, the smallest governed relation set needed by the named use, any direct relation that the subject’s identity predicate selects as identity-bearing (including its defined participants and occurrence-identity rule), the declarations the receiving use reuses, direct defining or testing rules, receiving use, and non-use and reopen boundaries. Also cite the durable-membership rule and scheme, the same-individual inclusion law or identity-dependence relation when applicable, and the exact result references. Record whether the receiving use needs an additional durable ontic.If direct rules suffice, return the direct-use or bounded-coordination disposition and no new ontic result. Only when both new outputs are independently needed does an atomic co-decision return separate outputs on independent grounds. Both remain provisional while a required branch is unresolved.
-
By-value dependence. Current or selected downstream uses cite the kind by value rather than only repeating its label.
An unestablished required condition leaves positive admission unresolved. When the available basis establishes a non-admission exit, select reuse if an admitted durable kind already covers the distinction, local-kind if bounded C.3.2 classification is sufficient, or reject if no classification result is needed for the receiving use.
A failed positive test alone supplies none of those results. Recover the direct relation, declaration component, selected structure, episteme, publication form, representation element, or source wording that carries the current claim. Only after disposition is settled may an author apply F.8, F.5, and F.18 naming criteria and constitute any public F.17 row.
Six Admission Dispositions
The typed AdmissionDisposition has exactly six values:
root. The candidate classifies individuals identified by one cited identity or membership rule whose extent and recognition conditions are explicit.same-individual-dependent. The candidate classifies individuals already admitted under one root U-kind. The root pattern keeps individual identity; the dependent pattern adds a stable membership condition and an action-facing use. The accepted settlement also states the implication: if that same individual satisfies the dependent condition, it is a member of the named root kind.identity-dependent. The candidate classifies a distinct individual whose identity cannot be stated without one named root-kind individual. The exact dependence relation between those two individuals and every additional discriminator must already have a defining rule. A holder or root reference without that relation does not close admission.reuse. The needed individuals and distinction are already covered by one admitted durable U-kind. Reuse that kind and its cited identity or membership rule; do not admit a duplicate root or dependent kind.local-kind. Record this non-admission exit only with one current C.3.2 declaration throughLocalKindDeclarationRef. The distinction remains local under the C.3 family and does not become a root or dependent durable U-kind; E.24.UK does not restate the declaration's internal mechanics.reject. The receiving use requires no durable or local classification result. Keep the exact relation, declaration component, selected structure, episteme, publication object, representation element, or source wording that carries the claim. When the receiving use needs only a statement of participation, use Plain relation-defined wording. When it needs classification by that participation, test the C.3.2 local-kind route before considering durable admission.
Only root, same-individual-dependent, and identity-dependent admit the candidate as a durable U-kind. reuse, local-kind, and reject are distinct exits, not weakened dependent admissions.
Read kind, individual, dependence, and part separately:
U.WorkPlanis a kind name.MaintenancePlan_Q3is one individual that may be classified by that kind. The name is not the plan individual, and neither is a declaration slot or record field.- Same-individual dependence adds membership, not another object. C.2.1 first identifies
MaintenancePlan_Q3as oneU.Episteme; when A.15.2's plan-membership predicate holds, that same episteme is also aU.WorkPlan. No second plan individual and no parthood claim follow. - Identity dependence concerns two distinct individuals joined by a governed relation that contributes to one individual's identity. A proposed capability individual distinct from its holder would need that relation and a receiving use that requires its continuity. A.2.2 instead states qualified ability about the holder; a holder reference in that assertion supplies no second individual.
- Dependence does not imply parthood. Even if a capability-to-holder dependence relation is governed later, that fact alone does not make the capability a part or characteristic of the holder system. A parthood conclusion needs its own direct part relation under A.1 and that relation's obtaining rule.
When the contrast is kind versus instance, say kind, individual, instance, or concrete governed object, not bare value. Reserve slot-filler wording for actual declaration slots and record-field wording for records.
Durable Membership and C.3 Projection
Durable U-kind membership and C.3 classification remain distinct; C.3 relies on an admitted meta-kind. E24UK-AR-UKIND-R5-01 admits U.Kind; its individuals are reusable intensional classification distinctions recovered through candidate domain, operative membership condition, intended member/non-member boundary, and continuity rule. A KindSignature, source or practice label, scheme, extension, assertion, or publication is not that kind individual.
For an independently identified candidate x, membership in an admitted durable subject kind K follows the direct predicate M_K under the accepted settlement. A C.3 U.Kind individual may declare or reuse that predicate for typed reasoning without admitting another public U.* kind. A row, spelling, record, or unresolved evaluation changes neither the direct predicate nor the world-side extent.
E24UK-AR-USUBKINDOF-R5-01 separately admits U.SubkindOf as a same-individual dependent kind under U.Relation. Its individuals are the same relation occurrences already admitted under U.Relation whose exact ordered kind participants satisfy C.3.1's criterion-entailment branch or exhaustive deliberately closed-domain branch within declared applicability.
Scheme and signature editions qualify the obtaining test and assertion; they are not participants or occurrence-identity discriminators.
For any other same-individual-dependent admission, the settlement states M_Kd(x) -> M_Kr(x) and the same individual keeps root identity. For identity-dependent, the cited rule defines or constrains a two-place dependence relation from the distinct dependent individual to one exact root-kind individual and supplies every additional discriminator. A root reference alone closes neither form.
The recovered A.2.2 capability need takes the reject exit in section 4.2c: ordinary holder-ability claims remain usable without a second kind or an invented dependence relation.
U.Structure follows the accepted A.22 architecture instead. A.22 identifies one context-independent selected organization from four and only four discriminators: exact independently identified constituents, exact selected obtaining relation occurrences, exact constraints as applied, and one named selection-use frame. E24UK-AR-USTRUCTURE-R12-01 records the root admission.
BoundedModelUseStructure and A.22's conditional crossing-analysis specialization are same-individual dependent predicates over already identified U.Structure individuals. The same structure individual keeps its A.22 identity; satisfying the corresponding A.22:4.1c condition adds the specialization and implies U.Structure membership. The bounded-model-use name has a current F.17 row.
For crossing analysis, identify the same U.Structure under A.22:4.1c. Its exact constituents include several independently identified BoundedModelUseStructure individuals. Its selected relations include exact crossings among those constituents that obtain under their own predicates. Establish the applied constraints and the frame for the named crossing-analysis use. An obtaining crossing can supply one of the selected relations. Membership belongs to the structure meeting all these conditions. The supplier-to-billing proposal in §5.0 supplies no such witness. The proposed public name remains pending under F.17.
An A.2.6 claim-scope value or membership fact affects the selection only when an exact applied constraint refers to it; that applied constraint, not the bare scope or membership outcome, occupies the third discriminator.
The three A.1.1 relation-kind designations consumed by the bounded-model-use test are current through UTS.ModelApplicabilityRelation.FPFCore.2026-07-25, UTS.ModelUseRelation.FPFCore.2026-07-25, and UTS.ModelExpressionCoherenceRelation.FPFCore.2026-07-25. Those F.17 rows publish only the names. A.1.1 defines each relation; the corresponding passage states its predicate, participants, obtaining condition, and occurrence-identity rule. A row, NameCard, matching token, or appearance in this registry makes no occurrence obtain and grants no BoundedModelUseStructure membership.
A project that needs bounded quantification may use an admitted U.Kind individual through C.3.2. If the kind's membership criterion cites an already governed durable subject-kind predicate, that projection neither admits another durable kind nor creates an automatic U.SubkindOf fact. A project-specific kind remains an individual of U.Kind without acquiring its own public U.* label; proposing such a label reopens E.24.UK for that subject kind.
Accepted Admission-Result Registry
Each E24UK-AR-* reference identifies one accepted UKindAdmissionResult, not the decision episteme that produced it. The registry indexes these results. The two R5 admission references resolve to their exact membership results in sections 4.2.2 and 4.2.3. The U.Kind ontology disposition is separate from its retained root admission. The bootstrap resolves to E24-CO-UONTIC-BOOT-01 and sibling result E24-OS-UONTIC-BOOT-01.
A row marked RG is a reconstructed by-value result whose exact subject-pattern passage and this row together state the disposition, membership test, reliance, and boundary. It records a reconstruction, without claiming that a new shared decision was made. No result reference gains a second result by appending a suffix.
Each R5 or bootstrap result is a C.2.1 episteme about the pre-judgment subject construct. Its own ClaimGraph states the exact shared decision reference, disposition, membership or identity basis, subject-pattern locator, branch result, reliance, and non-use/reopen boundary under FPFCoreReferenceScheme. An RG result instead uses the reconstructed basis stated above and carries no fabricated shared-decision or sibling-output reference. The decision episteme has its own ClaimGraph and identity. A consumer relies on the result reference and follows it to the decision when it needs common inputs or decision mode.
RG means reconstructed and grandfathered. Rely on the result identified by the reference. The row wording alone cannot support that reliance.
Reopen a row if its direct membership or identity predicate, intended extent or named reliance, nearest non-use boundary, or shared E.24 settlement law changes; carrier, layout, and spelling changes alone do not reopen it.
E24-CO-UONTIC-BOOT-01 takes the E.24 source construct, shared settlement rule, receiving use, and non-use boundary without presupposing U.Ontic. It returns E24-OS-UONTIC-BOOT-01 and E24UK-AR-UONTIC-BOOT-01; neither the schema, pattern, decision, nor kind thereby becomes an ontology-unit individual.
Open Prerequisites, Blocked Candidates, and Non-admission Results
When the shared decision encounters a public kind name without a resolvable accepted admission result, record the name as an open prerequisite. A new admission cannot assume that the named kind has already been accepted.
C.2.1:4.1.1 explains how to obtain claim content in writing, recover its effective interpretation and compare exact values. That may suffice without classifying all possible constituents. For the receiving ranges stated there, C.2.1:4.1.2 supplies the positive C.3 declarations EpistemeInterpretationScheme and NonemptyClaimContent. C.2.1:4.2.6 supplies exact constituent designations and their resolution without admitting another RefKind.
A receiver needing a different membership or substitution distinction must state its candidate domain and criterion under C.3.2. A reusable typed relation declaration must supply the independently governed ValueKinds required by A.6.5. ByValue is a designation mode. It does not waive either condition when that condition applies to the use.
These three language dispositions retain the same independently identified systems and formation, interpretation and reference predicates that the proposed public kinds would classify. The receiving comparisons use those identities and predicates; no further classification consequence is needed. Their ontology disposition is subject-pattern-use: apply E.5.2.L directly. No additional LanguageOntic or positive public-kind admission follows from these results.
E.24 may coordinate exact constitution and dependent relation rules in EpistemeOntic while the two constituent-kind proposals, U.ReferenceScheme and U.ClaimGraph, remain open. Its positive basis is the shared identity and the named cross-pattern dependencies, subject to E.24's participant-identification and predicate tests. That coordination admits neither proposed constituent kind and does not complete a typed interface whose required range is missing.
Generic reuse and local-kind are decision exits, not accepted example results. Close reuse only with a ReusedUKindRef that resolves to this registry; close local-kind only with one current C.3.2 LocalKindDeclarationRef. If either reference is absent, keep the candidate unsettled.
Use the disposition to decide how to repair each use of the name. Method-description claims retain U.MethodDescription; exact viewpoint and view claims retain U.Viewpoint and U.View only under E.17.0 membership.
For the availability-only use rejected by E24UK-NAR-EPUB-01, recover the selected U.Episteme, obtaining EpistemePublicationRelation occurrence, particular expression, optional reusable form or independently identified carrier that the claim concerns. When a source use instead requires classification by publication participation, preserve that need and follow the admission tree in section 4.3.
Here dependent describes how admission and identity are related. It does not apply merely because something is named in a record, participates in a relation, or qualifies the interpretation of an episteme.
Accepted Root Settlement For U.Relation
FPF has already admitted U.Relation; project users do not repeat this ontology decision. The root kind classifies individuable obtaining relation occurrences. A direct relation can obtain before a system individuates, names, describes, or references one occurrence, but admission under this root requires the direct relation pattern to supply an occurrence-identity rule.
The admission does not force explicit materialization of every obtaining relation. Ordinary engineering prose can stop at the direct relation sentence. A system performs explicit-individuation work only when a named receiving episteme, direct relation, or operation-application assertion depends on occurrence identity. The accepted Tech label U.Relation is governed separately through its F.18 NameCard; the label does not establish the extent.
Apply the positive extent rule before classifying a nearby object. Predicate content is a rule. An assertion or occurrence description is a C.2.1 episteme. F.18 governs a designator or reference, and E.24.PUB governs a reusable form. A row, graph edge or diagram element retains its interpretation under the applicable subject-description or representation rule.
Use C.29 for an unresolved mathematical representation, transfer or reliance question and A.6.3.RT for a representation transition. The rule, episteme, designator, reference, form and representation remain distinct from the obtaining occurrence. Connect the object under consideration to the occurrence only through its explicit assertion, description, designation, reference, publication, or representation relation.
An individuable publication-relation occurrence is a U.Relation when E.24.PUB defines that relation and states its obtaining and identity conditions. A row that represents the occurrence remains a representation element. Reidentify the current object by the rule that defines or tests it instead of inferring membership from words such as relation, edge, link, record, or reference.
Accepted Root Settlement for U.Kind
E24UK-AR-UKIND-R5-01 admits the root U.Kind. Its individuals are reusable intensional classification distinctions, identified through the C.3/C.3.1 candidate domain, operative membership condition, intended member/non-member boundary and continuity rule under FPFCoreReferenceScheme. A signature, source label, scheme, current extension, assertion or publication is not the kind individual. Existing individual subject kinds do not supply this cross-pattern meta-kind.
The root membership result is sufficient for recognizing such kind individuals. It does not require another durable ontology unit. C.3.1 comparison independently uses exact kind identities and the U.SubkindOf predicate; C.3.2 classification independently uses its admissibility and membership rules. Those receiving uses cite their direct rules without deriving them from an empty relation set. If several already defined claims must be read together for a bounded task, a C.2.1 episteme may coordinate them without becoming a durable ontic.
For these receiving uses, E24-CO-UKIND-R19-01 selects subject-pattern-use: apply the direct rules above without an additional durable ontic.
For example, C.3.1 can establish CoolingPumpKind as a subkind of PumpKind when the aligned cooling-pump criterion entails the pump criterion. An extension sample alone cannot do so. C.3.2 can classify an admissible Pump #14 under its declaration and slice; a missing required cooling-performance fact yields unknown, while an out-of-domain request is not-applicable before classification. Neither result needs a new ontology unit. A project-specific kind remains an individual of U.Kind without gaining its own durable public U.* name.
Reopen the additional-ontic question when a named dependent use requires persistent nonduplicative coordination beyond those direct rules and supplies a stable subject and positive contribution. Neither admitting the root nor adding SubkindOf merely because it is nearby establishes that need. E.24's positive EpistemeOntic case retains its independently justified constitution and neighboring-relation coordination.
Accepted Same-individual Dependent Settlement for U.SubkindOf
At the decision's entry, its subject was the unsettled C.3.1 proposal for an ordered kind-participant relation. The decision did not presuppose U.SubkindOf. The accepted relation occurrence keeps its U.Relation identity and gains the dependent membership only when C.3.1's obtaining rule holds.
The decision and its two outputs are three distinct C.2.1 epistemes. Each output has the same pre-judgment source construct as EntityOfConcern, its own ClaimGraph consisting of the applicable output claims above plus the exact decision reference, and FPFCoreReferenceScheme. The sibling ontic result is recorded for navigation, not used as prior evidence. Scheme and signature editions qualify interpretation, applicability, and assertions; they are neither participants nor occurrence-identity discriminators.
Practitioner-first Admission Tree
-
Recover the candidates and criterion. Begin with the decision the classification is to support. Identify the decision subject, who will classify which individuals and what inference the membership judgement must license. If several classifications are used together, establish the relationship required by that inference. When the use needs classification, identify the candidate individuals, stable membership or identity rule, intended extent and nearest non-member. For a relation kind, use the rule that defines its participant meanings, obtaining, applicability, and occurrence identity, and cite the PatternID that locates that rule; an
A.6.RCDapplication may record a derived or primitive candidate only with a proposed direct subject settlement. If the decision subject cannot be identified, keep the inquiry open. If classification is needed but its criterion is missing, the inquiry also stays open. A positively identified use that needs no classification can instead supportrejectunder step 6. -
Try an admitted durable kind. If one accepted result already preserves those individuals, the criterion, extent, boundary, and use, record
reusethrough that result and stop. -
Try bounded classification. If bounded classification supplies the required judgement or quantification, identify or make the C.3.2 declaration that supplies its criterion and record
local-kindfor that use. When another practice needs to rely on it, establish whether that criterion applies there, or which correspondence permits the intended inference. Several practices can share a local criterion. Different criteria can also serve them when the relationship between the resulting judgements is sufficient. -
Test the need for a new durable kind. Proceed to positive admission only when repeated cross-pattern use needs one stable membership law that existing durable kinds and direct relations cannot preserve. Run the eight tests and name each downstream question, its defining or testing rule, and the PatternID that locates that rule.
-
Choose the positive form. Use
rootfor independently identified individuals,same-individual-dependentwhen one root individual gains an additional stable membership predicate and inclusion law, oridentity-dependentwhen a distinct individual has an already governed dependence relation to one root individual plus all discriminators.Fill the shared E.24-family decision and separately test the need for additional durable coordination. Direct-rule sufficiency permits positive kind admission with an explicit no-additional-ontic conclusion; use atomic co-decision only when both new outputs are needed.
Apply A.11 and A.8 when kernel status is claimed. A fresh kernel proposal supplies A.8 with the independent candidate inputs of the shared decision; its projection finding contributes to that decision before step 6. An already admitted kind uses its applicable accepted result when its kernel placement is reconsidered.
-
Close or leave unresolved, then name. Record the positive admission when the selected form passes every condition. A missing branch law or fact leaves admission unresolved. Use
rejectonly when no classification result is needed for the receiving use, and recover the non-kind object under the rule that defines or tests it. Only after one disposition and governed object are stable may F.8, F.5, F.18, or F.17 expose a public name.
The PatternID locates the applicable rule content. C.3 states the membership and continuity rules for kinds. A.6.REL states the common relation-occurrence discipline. Each direct relation pattern states participant meanings, obtaining, applicability, and occurrence identity. A.6.0/A.6.5 define reusable declarations. E.24 defines ontic-settlement predicates. F.8/F.5/F.18/F.17 constrain names after ontology is settled.
Source Ontology Conversion Guide
Use this conversion guide when a source ontology, schema, standard, class hierarchy, or top-level ontology uses words such as type, class, category, object type, entity type, kind, or subtype. BFO-style, ISO-style, OWL/RDF, database-schema, programming-language, and discipline-local type systems are source ontologies or representation regimes; they do not become FPF U.* names by translation.
First recover the source construct by value:
- source name and source ontology or schema;
- source identity rule, membership rule, extent rule, or recognition rule;
- source relations such as is-a, part-of, realizes, participates-in, depends-on, or equivalent local relations;
- intended source use: classification, query, modeling, exchange, validation, reasoning, implementation, or documentation.
Recover what the receiving practitioner must decide from those source distinctions. For a question about membership, test which individuals satisfy the criterion in each practice. For a question about composition, establish which constituents form a whole of the specified kind. For a question about models at different levels of abstraction, fix the inquiry and identify what each model allows or rules out. A source hierarchy may combine several relations. Map each one used in the inference and give the grounds for any conclusion that crosses them. One local classification needs only its applicable criterion and facts.
Select the FPF object for each contribution the receiving work needs:
A source "type" may become an FPF kind and may require an ontic, but only after these tests. If the source construct only supplies local classification or exchange syntax, keep it as C.3 typed reasoning, bridge material, representation material, or source wording. Do not create a rival FPF type layer beside durable U-kind governance and E.24 ontic settlement.
Structural Location Rule
In a pattern title, host filename, monolith heading, or ToC row, readers use a U.* name to identify the governed object. The spelling there carries more weight than the same spelling in ordinary prose.
Use this rule:
- Prose occurrence: recover the local claim, the rule that defines or tests it, and that rule's PatternID locator.
- Table row or record field: recover whether it is one SlotSpec, one assertion or description field, one reusable-form element, or an already governed object.
- Heading: retain
U.*only when the section's primary EntityOfConcern is that object or the heading directly references an already admitted U-kind. - Pattern title or host filename: retain
U.*only when the pattern's primary EntityOfConcern is that root or dependent U-kind. - ToC row: retain
U.*only when the row points to the passage that carries the accepted settlement; otherwise name the direct governed object or repair the wording with E.10.
Do not keep a false U.* structural name for memory or search convenience. Use a Plain label, local heading, Name Card, Concept-Set row, relation name, record field, or quoted source wording when that is the actual object.
When positive U-kind admission fails
When positive admission fails, select a non-admission exit only when its own basis holds: reuse with one accepted result, local-kind with one C.3.2 declaration, or reject when no classification result is needed for the receiving use and the actual object is recovered under its defining or testing rule. If a missing law or fact prevents a truthful result, keep admission unresolved.
A participating entity keeps its intrinsic kind. A declaration component stays an A.6.5 SlotSpec. A designation or claim field stays in its episteme. A structure stays under A.22, and a publication form stays under E.24.PUB. A representation stays under its applicable subject-description or representation rule, with C.29 supplying unresolved mathematical representation, transfer or reliance work. A measure or source expression stays with its measurement or wording-use rule. Public naming waits until that recovery is complete.
Archetypal Grounding
Five Replays Through One Decision Sequence
Use the same five steps in every replay.
- Identify the decision’s EntityOfConcern and named use.
- Test an existing durable kind, direct relation, and bounded C.3 classification.
- For a classificatory distinction, state the governed individuals, membership or identity, intended extent, and nearest non-member. If the use needs no classification, identify what carries its claim.
- Test a proposed positive admission against all eight conditions. Use the shared E.24-family settlement and apply A.11/A.8 when the kernel question arises. A non-admission exit needs the grounds for that exit.
- For a closed replay, record its result reference, naming outcome, non-use boundary, and reopen condition. For a stopped replay, identify the missing grounds and leave the admission result open.
A new candidate must complete this sequence before its public name is admitted.
In each closed replay, the E24UK-* reference identifies the admission-result episteme or reconstructed result. The five steps summarize that result and, when a current shared decision exists, point to its separate ClaimGraph; the effective reference scheme is FPFCoreReferenceScheme. A stopped replay identifies what prevents admission. It records no admission result.
Reconstructed root — U.Relation.
- Subject and use. The EntityOfConcern is the A.6.REL source construct for the common kind of individuable obtaining relation occurrences. C.2.1 and receiving direct relations need to refer to one exact occurrence without turning an assertion, row, or graph edge into that occurrence.
- Coverage. No other admitted durable kind covers all and only those occurrences. A bounded C.3 kind would not supply the cross-pattern root used by direct relation patterns.
- Membership. An individual enters the extent only when its direct relation pattern establishes obtaining and supplies an occurrence-identity rule under A.6.REL. Predicate content, an assertion, description, designator, reference, tuple, or edge is the nearest non-member.
- Eight tests and settlement. Governed individuals, stable occurrence identity, direct-pattern witness, action-facing occurrence use, non-duplication, A.6.REL plus the direct relation pattern, accepted result
E24UK-AR-URELATION-R11-01, and by-value reliance are all present. A.11 retains one common root rather than duplicating it for every direct relation; A.8 does not promote relation-specific names into additional universal roots. This reconstructed result has no fabricated settlement suffix. - Result and flip.
E24UK-AR-URELATION-R11-01recordsroot;NC-U-RELATIONretains the Tech labelU.Relation. Reopen when the common occurrence criterion, direct identity discipline, dependent use, or settlement law changes. If an already admitted kind is found with the same governed extent and use, the disposition changes toreuse.
Same-individual dependent — U.WorkPlan.
- Subject and use. The EntityOfConcern is A.15.2's WorkPlan kind-source construct;
MaintenancePlan_Q3is a member witness, not the decision subject. Planning and readiness patterns need one durable way to recognize substantive intended-work epistemes. - Coverage.
U.Epistemealready supplies individual identity, but it does not by itself distinguish epistemes that substantively coordinate intended work. A one-project classification would be tested under C.3 before durable admission. - Membership. C.2.1 identifies
MaintenancePlan_Q3; A.15.2's plan-membership predicate classifies that same individual asU.WorkPlanand implies its rootU.Epistememembership. A calendar image or ticket title without substantive intended-work claims is the nearest non-member. - Eight tests and settlement. Identified epistemes, C.2.1 identity, the A.15.2 membership witness, planning use, non-duplication, A.15.2 as direct locus, accepted reconstructed result
E24UK-AR-UWORKPLAN-RG-01, and by-value A.15 reliance are present. Under A.11's test, the result is a same-individual dependent kind rather than a second root or plan object; no new A.8 universal root is claimed. This replay does not invent a separate settlement identifier. - Result and flip.
E24UK-AR-UWORKPLAN-RG-01recordssame-individual-dependent; the existing Tech labelU.WorkPlanis retained and this replay mints no new name. Reopen when C.2.1 identity, A.15.2 membership, the planning use, or settlement law changes. If only one bounded project needs the distinction and one exact C.3.2 declaration suffices, the disposition changes tolocal-kind.
Same-individual structure specializations — BoundedModelUseStructure and the A.22 conditional crossing-analysis rule.
-
Subject and use. The decision subjects are the A.22 source constructs for base
U.Structureand its two model-use specializations. A.1.1 and crossing-analysis consumers need durable membership without turning a context, team, subsystem, description, or view into another structure individual. -
Coverage.
U.Structuresupplies the one base identity. The two specialization conditions add stable action-facing membership to that same individual; neither needs an independent root or an identity-dependence relation to a context-like bearer. -
Membership and near-misses. A.22 first identifies
PressControlUse_Sfrom exact constituentsPressControlModel-5,Press-3, andPressControllerCode-17; selected obtainingModelApplicabilityRelation,ModelUseRelation, andModelExpressionCoherenceRelationoccurrences; an exact applied safety-control constraint claim whose proposition refers to the claim scope used by that selection; and the named use “decide whether operating use and controller-code maintenance belong to one bounded model-use organization.” That claim may state a proposition about the scope or its A.2.6 membership predicate; neither the bare scope nor the membership outcome is the constraint claim. Only then may the samePressControlUse_SsatisfyBoundedModelUseStructure.The supplier-to-billing proposal sets out
SupplierUse_S,BillingUse_S, direction, required fit, permitted loss and claim scope. It supplies neither an applicable predicate and facts establishing the crossingSupplierToBillingTranslation_Rnor a selection establishingSupplierBillingCrossing_Sas the structure for this analysis. Withhold those two assertions in this case.If a crossing obtains, it can serve as a selected relation. To establish positive membership, identify the structure itself by all four A.22 discriminators and show that it meets §4.2a, including the several bounded model-use constituents.
PressControlTeam, aBillingContextlabel,ContextMap_v3as aU.View, its diagram and its publication supply no such grounds on their own. -
Eight tests and settlement. Governed base-structure and bounded-model-use individuals, A.22 identity and positive bounded-model-use membership witnesses, action-facing model-use needs, non-duplication, A.22 as direct locus, the relevant settlements, and by-value reliance are present.
For the conditional crossing-analysis specialization, this replay establishes the rule and its action-facing need. The positive membership of the proposed supplier-to-billing structure remains unestablished, and the public naming decision remains pending.
A.2.6 contributes only when an applied constraint refers to an exact claim scope. That constraint, not the bare scope, membership outcome, or its representation, occupies the third discriminator.
-
Result and flip.
E24UK-AR-USTRUCTURE-R12-01recordsroot;E24UK-AR-BMUS-R12-01records the current namedsame-individual-dependentspecialization;E24UK-AR-A22-CROSSING-RULE-R12-01records only the conditionalsame-individual-dependentcrossing-analysis rule without asserting a current member or public term. Each specialization impliesU.Structuremembership only for the same individual that satisfies its exact A.22 condition.If base identity is unestablished, retain the warranted description or representation. If the base structure is established, retain it and any supported specialization. Reject a specialization for this candidate when a necessary condition is known to fail. If the condition is unknown, leave membership open. To conclude that the whole domain has no members, show that its conditions are incompatible there or check a demonstrably complete population.
Reopen only when the A.22 identity or specialization condition, the A.2.6 applied-scope interface, the named reliance, or the shared settlement law changes.
Rejected near-miss — a separate U.Capability individual.
- Subject and use. The source construct proposes
Pump37MaintenanceCapability_2026as an individual distinct from holderPump37. Recover what planning, promise, work-fit and development uses need: what the pump can do under stated conditions and bounds, with the warrant required for the receiving use. - Coverage. The holder retains
U.Systemidentity. An identified ability assertion uses C.2.1U.Epistemeidentity. The proposition, its assertion, support and demand-fit result need no second capability individual; no additional local classification is needed for those uses. - Discriminating cases. New evidence can change support without performance change. A stricter throughput demand can change fit. Calibration can change actual ability with unchanged paperwork. Loss and recovery change the qualified predicate at the stated times; replacing the holder changes the assertion's subject. None requires reidentifying a second individual.
- Tests and result. No additional classificatory distinction survives recovery, so
E24UK-NAR-U-CAPABILITY-01recordsreject. This rejects the new kind commitment, not the reality of the holder's ability. A holder-indexed tuple cannot substitute for an independently needed subject and identity rule. - Reopening condition. Reopen only when a concrete receiving use depends on continuity of a separate object and the available evidence supplies its required identity facts, including any obtaining dependence relation and its identity effect. A new record, evidence window, ability label or local-kind declaration alone does not supply that need.
Rejected near-miss — U.EpistemePublication.
-
Subject and use. The EntityOfConcern is the proposed kind source construct for an episteme made available to an audience; the use is to speak plainly about that availability.
-
Coverage.
U.Epistemealready identifies the episteme, while E.24.PUB defines the exact E/A/U/X/C publication occurrence, expression and bearing, with optional form conformance and separately quantified availability. -
Participation and classification. The same episteme can participate in a publication occurrence at one time and cease participating later. A fixed criterion can classify it by that participation at the declared slice. An episteme outside the required publication relation at that slice fails the criterion. The present use asks only for the availability claim, so it needs no additional classification.
-
Failed test and settlement. Non-duplication fails for the proposed public kind:
E24-OS-EPISTEME-ONTIC-02and the E.24.PUB rules already supply the objects and publication relation needed to state availability. A.11 returns that existing expression; the E.24.UK decision recordsrejectfor this proposal and use. No A.8 core-placement test or new name is needed. -
Result.
E24UK-NAR-EPUB-01recordsreject. Use Plain published episteme only in a claim that identifies or permits recovery of the obtainingEpistemePublicationRelation. This availability-only use needs noU.EpistemePublicationkind or public name.Reopen when a receiving use needs classification by publication participation. For a project query over available epistemes, first establish the C.3.2 declaration, including its criterion and slice; return
local-kindonly with itsLocalKindDeclarationRef. A proposed durable public kind has to meet all eight conditions.
Broad rule-content provision/support candidate
A proposal groups phrases such as "this pattern defines or constrains the result", "the framework supports the claim", and "service provision" under one public kind. The phrases do not identify common individuals: one locates defining claim content, another may describe evidence or source use, and the third may designate dated Work, a Method, promise content, or an operational bearer. They share neither an identity or membership rule nor a receiver that needs the proposed instances as one kind.
Recover each material occurrence first under C.2.1 and its exact subject relation. Then return reject for the broad candidate. Each result cites the corresponding assertion through RejectedCandidateRecoveryRef; it does not cite the shared word list or migration table.
Work, source, evidence, publication, authority, access, and direct-relation claims survive unchanged in their own meanings. An unrecovered occurrence stays unchanged. This rejection admits no U.Provision, U.Support, generic SupportRelation, governance occurrence, or rule-locus description kind.
Type And Kind Governance Passage
A passage that says a proposed type must pass A.8 or A.11 is a kernel-level U-kind admission question. A passage that says U.Kind and U.SubkindOf are used for typed reasoning remains under C.3 rules. A naming passage in F.5 or F.8 waits until the governed object and admission decision are stable.
Lower-level Heading
A lower-level heading containing U.* can name any of several objects. Its spelling alone cannot admit a U-kind. Recover whether the heading names an already admitted root or dependent U-kind, a declaration-local SlotKind, a claim-bearing U.Episteme, a relation-defined participant meaning, or a publication object. Keep the recovered object and the rule that defines or tests it; rename the heading when it advertises a different kind.
Bias-Annotation
A U.* spelling, title, filename, table row, or imported type word is insufficient to establish a durable FPF kind. First recover the governed individuals, their identity or membership rule, and the PatternID that locates that rule.
If the candidate names participation in a relation, a SlotSpec, an assertion or description field, a selected U.Structure, an E.24.PUB form, or a C.29 representation element, keep that object and its defining or testing rule. For a structure specialization, first identify the same base individual by A.22’s four discriminators, then test its specialization predicate. Only then decide whether a durable U-kind distinction remains.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Positive consequences:
- public
U.*names become reliable orientation signals; - dependent durable U-kinds can be named without pretending to be independent roots;
- model-use structure specializations can be named while retaining A.22 base identity;
- type and kind wording is handled through C.3, E.24.UK, A.8, A.11, F.8, and F.5 rather than preserved as overlapping ontology;
- structural names are settled before they become misleading public names.
Costs:
- pattern authors must read the governed object before keeping a convenient
U.*spelling; - some familiar host filenames, headings, and ToC rows must be renamed;
- structural inventory work becomes part of U-kind governance.
Rationale
FPF needs U-kind names to stay rare and load-bearing because they orient many patterns at once. Without a separate U-kind governance rule, ordinary type words, source-ontology classes, slot labels, filenames, and memorable headings create a second ontology beside E.24 ontic settlement and C.3 typed reasoning.
The admission rule keeps durable classification connected to direct ontology without making every local class public. E.24 and E.24.UK share one settlement; C.3 handles bounded classification. A same-individual dependent kind adds one membership predicate and root-inclusion law to an existing individual. An identity-dependent kind instead requires a governed relation to a distinct root-kind individual plus all discriminators. Missing branch evidence blocks admission, and no public name, reference field, or owner label substitutes for it.
SoTA Decision for Durable Kind Admission
When should a reusable distinction become a durable public kind, when is a local classifier or direct relation enough, and what identity commitment does each choice add? Compare the alternatives for the work of repairing or authoring one FPF pattern. Adoption of a whole enterprise ontology falls outside that comparison.
Contribution to the FPF admission decision. E.24.UK first tests reuse of an accepted kind and bounded C.3 classification, and recovers a direct relation or other non-kind object when no classification result is needed for the receiving use. A new durable kind requires repeated cross-pattern reliance and an exact membership or dependence rule. Where the simpler result already preserves the needed distinction, it avoids another ontology commitment while retaining identity, non-member, reliance, and reopen tests.
The shared E.24 decision is part of that gain. It prevents the ontic and public-kind questions from being answered by two incompatible cards.
The U.Kind case retains the independent root admission and returns direct rule use without an additional durable ontic; the separate U.SubkindOf settlement remains under its own direct relation and receiving-use grounds. In the capability case, recovery returns qualified holder-ability claims and rejects an unnecessary additional kind. In the WorkPlan case, the same episteme gains dependent membership without a second plan.
A role or phase near-miss stays with a direct relation or local kind; HighRiskPump@Turnaround2026 stays a local C.3 distinction; and public spelling waits for F.18 after the object is settled.
A source may model a pressure quality or an obligation-bearing relation as a separate dependent individual. FPF does so only when a direct rule identifies that individual and its exact dependence. A measurement value, assertion, participant pair, agreement document, or relation record cannot admit it by representation alone.
The progressive route can under-admit a distinction if a bounded use hides later cross-pattern reliance, or over-admit it if repeated wording is mistaken for stable membership. Reopen when a stronger current alternative offers the same identity assurance at lower effort, when an admitted kind loses its member/non-member or continuity test, or when a worked counterexample changes the same-individual, identity-dependent, local-kind, or non-kind decision.
Relations
- Shares settlement with:
E.24through the oneE24FamilySettlementDecisionschema inE.24:4.0a. E.24.UK records theUKindAdmissionResult; E.24 records the ontology disposition and anOnticSettlementResultonly when a durable ontic is selected. An existing result may be reused, while a case needing both new outputs is one atomic co-decision with neither output used as prior evidence. - Uses for relation admission:
A.6.RELstates the common occurrence discipline. Each direct relation pattern states participant meanings, obtaining, applicability, and occurrence identity. AnA.6.RCDapplication may record a residual claim or a derived-or-primitive candidate with its proposed direct subject settlement; local-claim and predicate-definition results remain claim content and do not admit a relation kind. - Uses for neighboring objects:
A.6.0defines reusable signature identity;A.6.5definesSlotSpecdeclarations;C.2.1defines admission-decision, assertion, and description episteme identity;F.18supplies the naming rule for selected Tech labels and designators; subject-description patterns govern their correspondence claims,A.6.3.RTgoverns representation transition, andC.29supplies unresolved mathematical representation, transfer or reliance work. - Coordinates with:
A.22for admitted structure identity and dependent structure predicates;A.1.1for bounded model-use relations;A.2.6for claim-scope membership;C.3,C.3.1, andC.3.2for admittedU.Kindindividuals, the admittedU.SubkindOfdirect species, declaration, admissibility, and judgment;E.24.CDfor candidate detection;E.24.PUBfor publication distinctions; direct kind patterns for every other admission; andE.10, F.8, F.17, and F.18 for wording and naming after settlement. - Does not replace: the rule that defines or constrains the classified individuals, their identity or membership, intended extent, and action-facing use, or the PatternID that locates that rule.
E.24.UK:End
Last Updated: 2026-11-22 — upstream FPF commit 7a93d8f8 (github.com/ailev/FPF)