Part F - The Unification Suite (U-Suite): Concept Sets, SenseCells, and System-Role Kinds and Assignments
Preface node
heading:part-f-the-unification-suite-u-suite-concept-sets-sensecells-and-system-role-kinds-and-assignments:126396
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
Source-Local Meaning Recovery
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative
Problem frame
One-sentence summary. Recover what one expression means in one exact source passage before comparing, translating, or reusing it elsewhere.
Use F.0.1 when a word in an already selected source may be read in more than one way and the reading can change the present answer or action. Name the exact source and edition, state the source-local meaning in ordinary language, and point to the passage that supports that reading.
The first useful result is one source-backed meaning statement. For example: “In OMG BPMN 2.0.2 (January 2014), Process here means the designed sequence or flow of Activities in an organization; this reading comes from §10.1.” If that answers the question, use it and stop.
What changes in practice. The meaning and its source are inspectable from one statement. Add a formal result only when a named receiving use needs it.
Not this pattern when. Use an already clear source claim directly when no lexical distinction changes the work. Use F.1 when the open question is which sources can change the answer, F.0.2 when several source ontologies must be compared for one receiving claim, and F.18 when the problem is selecting an FPF term after the subject is settled. When the problem is not lexical, use the rule that defines or tests the exact entity, relation, claim, measurement, permission, or Work question. Use F.9 only when an actual relation between already recovered source-local meanings is current.
Problem
A string does not identify one meaning across sources. The same spelling can denote different kinds of subject; for example, process can name a designed structure or a performed occurrence. If the source, edition, and passage disappear, a later claim can silently change its entity, relation, time stance, or intended use.
Requiring a full source survey, durable cell, provenance record, reliance disposition, or assurance claim for every word makes ordinary reading needlessly expensive. Recover one local meaning first and show when further work is needed.
Forces
Solution — recover locally, strengthen only for use
Recover one source-local meaning
- Name the question. What does the reader need to understand or do, and what would reading this expression differently change? The question may concern the wording itself or something the passage describes.
- Identify the source. Give the exact source and edition. A discipline or shelf label is not enough.
- Locate the passage. Point to the claim, definition, example, rule, or other passage used.
- State the meaning plainly. Say what the expression denotes or claims in that passage. Keep designed descriptions and performed occurrences distinct when the source does.
- Return the meaning to the question. If the reader asked what the expression means here, give the source-backed reading. If the passage supplies a premise about the subject, use it in the inference the reader needs. To decide whether work followed a described process, for instance, compare the applicable requirements with evidence of that work (F.11).
The source, edition, expression, passage, and plain meaning are the ordinary minimum. A brief scoped heading may keep an already established source in view; repeated source tags are unnecessary when the reading remains unambiguous.
Add a durable address only when it earns its cost
Create one F.17 SchemeSenseCell <ReferenceScheme, LocalExpression, LocalSenseClaim> only when stable reuse, a later claim, a named receiver, or an actual relation to another local sense needs an exact address. The effective ReferenceScheme keeps the source and edition recoverable; the LocalSenseClaim states the meaning rather than hiding it in a label.
State an obtaining LocalSenseBasisRelation only when the support relation from an exact basis episteme to that cell is current and supported. Identifying a source, stating that support relation, relying on it in a later use, and assuring that reliance are four different claims. Use A.10 and B.3 only for the latter questions.
When a source fixes a designed-versus-performed distinction or another time stance, put it in the local sense claim or its exact basis. Do not make an edition label or separate container stand in for the distinction.
Relate different local meanings only when the relation is current
A shared label, close paraphrase, common superclass, table row, embedding score, or family membership does not establish identity or another relation. First recover each source-local meaning separately. Then use F.9 when a receiving use needs an actual relation between the two F.17 cells.
The F.9 result states which two cells are related, what kind of relation obtains, how its endpoints are oriented, and what relation profile makes it true. When a receiving use is current, state a separate C.2.1 claim: what action is proposed, in which direction, under which correspondence rule, how much loss it tolerates, and whether the Bridge is suitable for that use. Changing this use claim does not change the Bridge. Neither the relation nor the claim by itself permits translation, substitution, or row membership, establishes reliance or authorization, or shows that the action occurred. A chain alone does not establish a direct relation between its endpoints.
Recover old Context-shaped artifacts only for a current reliance
An old Context Card or two-part SenseCell remains a historical episteme or representation under its original edition. Do not relabel it as a current F.17 cell.
When a current claim or action relies on it, recover only the values that use needs: for example, the exact source and edition, expression, source-local claim, passage, effective scheme, claim scope, or obtaining relation. If a needed value cannot be recovered, return the exact unresolved value and reopen only the dependent claim or action. Keeping an artifact in an archive does not require migration.
Minimal conceptual objects
Invariants
- Every load-bearing local meaning has a recoverable source, edition, and passage.
- A plain source-backed statement may be the complete result.
- A
SchemeSenseCellis created only for a named durable use and keeps scheme, expression, and claim distinct. - A basis relation is stated only when it obtains; reliance and assurance remain separate.
- Keep the local meanings separately identifiable. An established relation supports only the correspondence its profile specifies.
- Designed and performed readings do not become identical through shared wording.
- A changed edition, passage, or relation reopens only claims and uses that relied on the changed premise.
- Historical artifacts remain historical; current recovery does not require corpus-wide relabelling.
Readable reasoning moves
- Local reading. “This source passage uses t to mean m.”
- Durable address. “This receiver will reuse that reading, so record it as an F.17 cell under the effective source scheme.”
- Basis. “This exact source episteme supports the cell through a current
LocalSenseBasisRelation.” - Cross-source relation. “The two recovered cells stand in this stated F.9 relation under this relation profile.” When a receiving use is current: “A separate C.2.1 claim says whether that Bridge is suitable for this action, direction, correspondence rule, and tolerated loss.”
- No transitive shortcut. Two established relations through an intermediate meaning do not establish a direct third relation.
- Affected-only reopening. A changed source premise reopens the claims that used it, not every claim that cites the edition.
These examples illustrate the reasoning; they prescribe no storage format or workflow record.
Archetypal Grounding
Entry, stop, and continuation
Start with one troublesome use in one already selected source. Return one ordinary sentence plus a source pointer. Stop when it answers the question.
Continue only for the next result you need:
F.1for a question-relative source cut;F.17for a durable local-sense address and, when current, its basis relation;F.9for an actual relation between distinct recovered cells;F.0.2for a bounded comparison or synthesis among source ontologies;F.18for naming after the subject distinction is settled; or- the rule that defines or tests the exact entity, relation, claim, measurement, permission, or Work question, with the pattern ID as locator.
Compact worked results and recognition cues
A worked result below names the exact source, edition, passage, and meaning used. A recognition cue only marks a likely false friend; it establishes no local meaning until the practitioner supplies those four values. This distinction lets readers use examples from several fields while seeing which readings still need source work.
process and activity
- Worked result — OMG BPMN 2.0.2 (January 2014), §10.1, Processes. Process denotes the designed sequence or flow of Activities in an organization.
- Worked result — W3C PROV-O Recommendation (30 April 2013), §3.1, Starting Point Terms. Activity denotes something that occurs over a period of time and acts upon or with entities, including using or generating them.
A request about the two expressions may end with their local meanings. If the reader needs a reusable cross-source correspondence, recover the two F.17 cells and test the proposed F.9 relation, including what it preserves about concurrency or traces. To decide whether Tuesday’s work followed the process, use F.11 and A.15.1. Find the applicable process specification and evidence of the actions and ordering in that execution. The source terms can retain their meanings whether the work followed the plan or departed from it.
actuation and control output
- Recognition cue — control theory. If actuation may mean a signal applied to plant actuators, select one exact control-theory publication, edition, and passage before using that reading.
- Recognition cue — IEC 61131-3. If control output may mean a program-produced value sent to field I/O, identify the exact edition and clause before using that reading.
For a comparison of the two expressions, recover both readings before testing a semantic correspondence under F.9. To establish that a particular program output commanded an actuator, trace the output through the I/O configuration and operating regime. To establish the actuator’s response, examine the corresponding observation. Hardware and scan-cycle limits can affect those conclusions.
observation and service metric
- Worked result — W3C SOSA/SSN Recommendation (19 October 2017), §4.3.2.2,
sosa:Observation. Observation denotes the act of carrying out a procedure to estimate or calculate a value of a property of a feature of interest. - Recognition cue — ITIL 4. If service-level metric is used for a quantity that evaluates a service-level objective, identify the exact ITIL 4 publication, edition, and passage before using that reading.
The verified SOSA reading alone establishes no service-metric relation. Once the second reading is source-backed, a named use may ask whether the observation supplies evidence for that metric; that is a subject relation, not lexical identity or same-row membership.
subclass-of and is-a
- Worked result — W3C OWL 2 Structural Specification and Functional-Style Syntax, Second Edition (11 December 2012), §9.1.1, Subclass Axioms.
SubClassOf(CE1 CE2)states that the first class expression is a subclass of the second. - Recognition cue — engineering glossary. If is-a is being used as a less formal kind-of relation, identify the exact glossary, edition, and entry before relying on that reading.
Keep the verified formal reading and the unresolved cue separate. Relate them only when the receiving artifact needs the formal relation and an exact second passage supports the correspondence.
permission and RBAC role
- Worked result — W3C ODRL Information Model 2.2 Recommendation (15 February 2018), §2.6.1, Permission Class. A Permission allows an action on an Asset when its refinements and constraints are satisfied and its duties are fulfilled.
- Recognition cue — NIST RBAC. If role is being used for an access-control grouping through which permissions are assigned, identify the exact NIST publication, edition, and passage before using that reading.
Recover the RBAC role meaning from its source before relating it to the ODRL Permission meaning for a receiving access-control use.
Quick checks for later use
- String check. If the only evidence is the same spelling, no cross-source relation has been established.
- Stance check. If one source describes a design and another a performed occurrence, state that difference before any relation or row use.
- Direction check. Preserve the direction and limits of the actual relation; a reverse or broader reading needs its own support.
- Chain check. Keep intermediate meanings and accumulated loss visible; test a direct endpoint relation separately when needed.
- Contradiction check. Incompatible relation claims about the same cells remain explicit rather than being averaged into a vague alignment.
- Row check. A Concept-Set row needs the relation and bounded receiving-use judgment required by F.7 and F.9; a label or confidence level cannot admit a member.
Quick reference
- Ordinary result: one source-backed plain meaning statement.
- Durable local meaning: an optional F.17
SchemeSenseCell. - Current support: an optional obtaining
LocalSenseBasisRelation. - Different local meanings: separate cells; use F.9 only for an actual relation.
- Source selection: F.1; synthesis: F.0.2; naming: F.18; subject reasoning: the defining or testing rule for the recovered claim.
What does the reader need from this passage? Identify its source and edition, locate the expression and give its local meaning. If that settles the question, finish there. If the question concerns the subject described, use the recovered meaning to establish the relevant claim. Give the meaning a durable address when a later use needs one.
Bias-Annotation
- Gov: Source authorship, popularity, or standard status does not decide the receiving claim.
- Arch: The pattern favors recoverable local meanings and explicit relations over one global vocabulary or universal container.
- Onto/Epist: A word, a local-sense claim, its subject, its source, a basis relation, and a receiving claim remain different.
- Prag: Ordinary use is one sentence and one citation; stronger objects and assurance are conditional.
- Did: Plain worked cases come before formal designators. A scoped source heading may reduce repetition without hiding the source.
- Scope: The pattern recovers meaning. It does not establish source truth, source adequacy, relation obtaining, permission to substitute, or assurance.
Conformance Checklist
Static checks
- SCR-F01 (Recoverable source). Every load-bearing local meaning identifies the exact source, edition, and passage directly or through an unambiguous scoped reference.
- SCR-F02 (Plain first result). A readable source-backed statement answers a question about local meaning. Where the passage supplies a premise for a further conclusion, its contribution to that conclusion is identifiable.
- SCR-F03 (Conditional cell). A
SchemeSenseCellappears only for a named reuse, claim, receiver, or relation need and keeps ReferenceScheme, LocalExpression, and LocalSenseClaim distinct. - SCR-F04 (Separate support). A
LocalSenseBasisRelationis asserted only when current; source identification, reliance, and assurance are not inferred from it. - SCR-F05 (No string identity). Shared wording, family, score, or row does not establish a relation between local senses.
- SCR-F06 (Explicit relation and use). Every claimed cross-local relation uses exact F.17 endpoints and the applicable F.9 relation profile. When a receiving use is current, a separate C.2.1 claim names the proposed action, use direction, correspondence rule, tolerated loss, and polarity.
- SCR-F07 (Temporal honesty). Designed descriptions and performed occurrences remain distinct wherever the source fixes that difference.
- SCR-F08 (No subject capture). The local gloss does not redefine the subject's behaviour, deontics, measurement, kind, proof, or work rules.
Regression and evolution checks
- RSCR-F01 (Affected edition change). A changed edition or passage reopens only local claims and later uses that relied on the changed content.
- RSCR-F02 (Endpoint change). A changed cell claim or stance triggers recheck of relations and uses that cite that endpoint.
- RSCR-F03 (Composition guard). A relation chain alone establishes neither a direct endpoint relation nor unrestricted substitution.
- RSCR-F04 (Source-cut relevance). Reopen F.1 only when the receiving question or use, a relied source role, known rival, counterexample, or transfer limit changes.
- RSCR-F05 (Historical recovery). An old Context-shaped artifact is recovered only for a current reliance; missing values return as exact unresolved inputs.
Common Anti-Patterns and How to Avoid Them
Consequences
Local meanings become inspectable and revisable without forcing a global vocabulary. A reader sees when source selection, synthesis, naming, subject reasoning, or a cross-local relation is the next question. When one sentence and a citation answer the question, the reader can stop before preparing a reusable cell.
Making the source, edition, passage, and meaning explicit takes judgment. Creating a reusable cell or establishing a relation takes additional work. Undertake it only when a later receiver needs it.
Rationale
Recover the meaning from the passage where the source uses the expression. When the passage settles the question, this avoids the cost of a wider comparison. Starting with the source passage prevents a later comparison from rewriting the source before the commonality or difference is known. Keeping durable cell identity in F.17 and cross-local relations in F.9 prevents local recovery from becoming a second ontology or relation catalogue.
Choose any further work by what remains to establish. F.1 helps select a source basis, F.0.2 develops an account from consequentially different source claims, and F.9 tests a proposed relation between local meanings. Several sources may give the same sufficient answer.
SoTA-Echoing
Together, these choices support a plain local answer, with a durable address or explicit relation added only when the receiving use needs it. A Bridge's suitability for that use remains a separate judgment.
Relations
F.17definesSchemeSenseCellandLocalSenseBasisRelation.- use
F.1to select exact sources and editions for one receiving question and use. F.0.2compares or synthesizes several source ontologies after their meanings are recovered.F.2andF.3support term evidence and local-sense clustering when those heavier moves are needed.- Use
F.7for Concept-Set rows andF.9for actual cross-local relations and their bounded uses. - Use
E.15for edition continuity, supersession, and affected-premise reopening. - Use
A.10for evidence use and reliance andB.3for assurance. - Use
F.18for term choice after the subject distinction is settled.
F.0.1:End
Semantic Synthesis across Source Ontologies
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative
Problem frame
Use this pattern when what someone can conclude or do depends on how different source accounts fit together or disagree. The sources can address the same question or supply different premises needed to answer it. The recipient may need the comparison itself, a way to act that depends on it, or a contribution to a framework pattern.
In this pattern, conceptual synthesis and semantic synthesis name the same work. It means comparing and developing claims through the concepts, relations and boundaries used by their sources. It can concern a theory of action or any other subject. When undertaken to construct or revise a method, this conceptual work is part of methodological synthesis. Its conclusions can also answer an inquiry in their own right.
Begin with the question the recipient actually asked. What would two different answers to it lead them to believe or do? Use that difference to choose the source claims worth comparing. A familiar word can lead you to a passage about only one part of the subject. Recover its local meaning through F.0.1 when it matters, and use F.1 when the choice of sources needs justification.
What goes wrong if missed. An author either leaves a crosswalk or literature narrative where a positive contribution is needed, or merges similar words before checking whether the sources describe the same entity, relation, explanation, and use. A large research package may also be demanded for a question that one bounded comparison can settle.
What changes in practice. The recipient can use claims from different sources to explain a relation, distinguish accounts that lead to different conclusions, or identify what still prevents an answer. When developing a method, use the result to build or change the operation that needs it.
Not this pattern when. Cite one source directly when one source claim closes the question. Use F.1 when only source selection is current; F.0.1 when only local meaning is unclear; F.9 when one relation between already recovered cross-context senses is current; and the subject pattern when it already supplies the needed action and result. Use G.2 instead when the work needs a broad, refreshable SoTA Synthesis Pack@CG-Frame and its downstream Part G handoffs.
Problem
Researchers, practitioners and framework authors often receive accounts that use similar terms for different relations or explanations. A correspondence between terms identifies a comparison to make. To decide what follows, recover what each source says about the same inquiry, under which conditions and on what grounds.
Three failures follow:
- a correspondence, shared label, similarity score, or framework fit is treated as permission to merge;
- a source difference that changes action is hidden as terminology variation; or
- missing or weak source material is reported as if the disputed domain claim were false.
Semantic synthesis must preserve source-local meanings while returning a positive next move or a bounded stop.
Forces
Solution
Bounded synthesis method
- Start from the difference the recipient needs explained or decided. Identify what could have the property, make the choice or undergo the change in question. What must a satisfactory account let this recipient distinguish, and what choices are theirs? A comparison of terms can finish with their meanings. A question about how several parties accomplish a job calls for an account of their contributions to that job. When the question leaves the choice of wholes, parts or model distinctions to you, use :4.1.1 to choose the reach of the comparison.
- Locate material for that question. Reuse known passages that address it. If a source, rival explanation or connecting premise is missing, use F.1 to locate material you can inspect. A source may matter because it connects two other contributions whose meanings are clear.
- Recover and select the claims that matter. Identify what each source asserts, about which subject and under which interpretation and conditions. Keep the edition and its role in the argument recoverable. Use F.0.1 for a disputed local meaning, and F.9 if the argument needs a specific relation between local senses. Select sufficient grounds for the receiving question, including a consequential rival or limit. Reopen source selection if the comparison reveals a gap.
- Construct and test the account. For each source, ask which part of the recipient’s question it can answer in its own terms. What grounds allow a conclusion from one account to enter the other? When each source answers a different part, follow its contribution to the point where the recipient needs them together. Try leaving their separate claims true while changing how the described things are connected. If the answer changes, construct a candidate connection. Identify the participants each source distinguishes and propose what would have to hold between them in the same case. Derive its consequence for the recipient. Check the connection against the source conditions and any case that could distinguish it from a competing account. Keep a conditional explanation when the grounds support only that much. If the accounts rival one another, use a difference in their consequences to decide between them where it matters (B.5.TC). Keep the conclusion they share when that already answers the question.
- Give the resulting account. A provisional synthesis states what the source contributions warrant together and why. A contrast explains a consequential difference that survives the comparison. When a source gap prevents either conclusion, identify the unresolved inquiry and carry forward whatever the available material already establishes.
- Return the result to its intended use. Show which conclusion the recipient can now draw or qualify and on what grounds. When the work informs construction of a method, identify the operation or condition it changes. For a framework contribution, name the content or placement decision it informs. Keep the source edition recoverable and give the result a stable locator when it will be cited or replayed. C.2.1 governs an explicitly identified claim episteme.
- Decide what to do with the result. Use the options for its branch in §4.3. When the receiving use requires a record, state the decision and cite the exact result and source edition. A result awaiting the decision remains a proposal; silence or citation alone is not acceptance.
- Keep the source use revisable. Identify the source claims and editions relied on, why each matters to this result, what was deliberately left local or rejected, and which change reopens the comparison.
Stop when the result answers the receiving question to the extent the sources allow. If a material premise is missing, state what remains open and what would settle it. Enter clustering, Concept-Set development, naming, UTS or broad source harvesting when the next use needs that further result.
Choose the reach of the comparison
Start at the subject the question names. If it names only an effect, ask what could produce or change that effect. If the recipient asks about a whole, consider whether the same constituent contributions could be organized differently with a different outcome. If the question starts with a part, consider whether putting it to another use changes what follows for the enclosing whole. Trace the composition of each kind that matters. For example, the acts that made up one job can differ from the operations prescribed by its method. Evidence that an act took place can connect those two accounts when it also establishes which operation was performed. At each scale, compare cases a proposed model treats alike. If they require different answers or different explanations that the recipient needs, restore the distinction and test the inference again. To simplify the model, establish why the inference survives the variation you propose to leave open. C.39.SA:4 develops this choice. B.5.MR:4.2 tests a claim that one model strictly refines another for the same inquiry and interpretation.
A rule applying separately to each part, a conclusion that carries from one scale or model to another, and a conclusion about what the parts obtain together require different grounds. For a transfer, identify the condition preserved in the required direction. For a joint inference, work out what follows when the premises concern the same arrangement (C.37.MM:4.1.1). One feasible arrangement can establish possibility. To explain what happened, establish that the relevant conditions held together and give the grounds connecting them to that outcome. For a guarantee, show why the required result follows across the permitted arrangements. A direct argument about the whole can settle the question without reconstructing every part. If a source itself gives the argument for the entire range, use it directly.
Source-basis branches
Three result branches
A result identified as a claim-bearing episteme retains its C.2.1 identity <claim content, EntityOfConcern, effective ReferenceScheme>. Keep the result and source edition recoverable for the receiving use. A stable local locator can supply a needed address for later citation or replay; it does not identify the claim by itself.
A finding about what the sources establish can itself answer the receiving question. If two authoritative rules leave the choice of the next participant to the commissioning party, the comparison can establish that those rules assign nobody. Making the appointment then requires an authorized decision. If a needed premise remains genuinely unresolved, explain its effect and whether a further observation, source or decision could settle it. A later user who needs to cite this result needs an identifiable account of it.
Bounded method and broad SoTA-pack method
Use F.0.2 when a bounded comparison of source claims, their interpretations and their consequences can answer the recipient’s inquiry with a synthesis, a contrast or a statement of what remains unresolved. Use G.2 when a CG-Frame needs a broad, refreshable harvesting surface with a CorpusLedger, Claim Sheets, palette, inventories, alignment records, examples, and declared G.3-G.5 handoffs.
An author may use identified claims and provenance from a G.2 pack as inputs here. The author must compare retained differences, limits, receiving use, and reopen conditions before writing an F.0.2 result. The bounded comparison does not require the rest of the pack when the question does not use it.
Boundary to wording-use precision restoration
Semantic synthesis and wording restoration both pass through ontology, but they answer different questions. Semantic synthesis forms or revises a cross-source claim. Wording restoration starts from an already current subject claim and repairs a consequential wording use so that the intended entity, relation, claim kind, and admissible action can again be recovered.
A recurring wording failure may supply evidence that a subject distinction is missing or unstable. The system maintaining the affected FPF or DPF edition then uses that evidence in a named content decision. The findings can support a proposed revision. The content decision governs changes to the claim maintained in that edition. A DPF keeps its domain wording entries beside the domain patterns that use them; E.10.ARCH supplies the shared restoration method.
A DPF may need a reliable current domain ontology so practitioners can recognize situations, distinguish Methods and results, and use its solution moves. An author can use F.0.2 to synthesize or revise that ontology as a proposed contribution. Ontology alone is not a DPF: use E.4 and E.4.DPF to connect it to recurring problems, constructive Methods, a usable first cut, evidence practice, access, and maintenance.
Semantic and methodological results
A conceptual synthesis states what the source accounts warrant about an identified subject, which kinds and relations support the claim, and where a disagreement remains. This can be the result a recipient asked for, including a result about actions or methods.
When the requested result is a new or revised way of working, call the whole undertaking methodological synthesis. Use established concepts where they suffice. If the source accounts disagree about what an operation acts on, what it supplies or what counts as success, compare those accounts. An agreement can supply a shared premise. If different readings are warranted under different conditions, provide for each in the method. If a premise remains open, obtain the evidence, use a way that works under either reading, or withhold the conclusion that depends on it. Trying the method can in turn reopen a conceptual conclusion. When the comparison is also needed as a reusable explanation in its own right, name both deliverables and what supports each. One recipient can need both in a single commission.
Carry each result to the scope at which its recipient will use it. A transdisciplinary way of reasoning can guide the work of a single discipline. Identify the conclusion at stake, examine how the field's types and relations warrant it and revisit dependent conclusions when a premise changes. In interdisciplinary work, identify what each field supplies and who needs each result. If a field uses a result from another, establish the correspondence it relies on. A recipient may use independent findings together. Identify the subject, conditions or constraint they must share for that recipient’s conclusion to follow. When two contributions appear to conflict, establish whether they concern the same subject under the same conditions. If the accounts are consistent but the products cannot work together, identifying the incompatibility may answer the inquiry. When the purpose requires joint operation, consider a change to the arrangement or the products and test whether it resolves the incompatibility. If the claims still conflict, identify what further evidence or revised account could resolve the conflict. Carry forward the conclusions whose grounds hold. A discipline supplies the particular means by which those grounds can be obtained or tested. A field method can be complete at this scope before a project selects staff, tools or a schedule. In a project, choose an applicable way and recover the local evidence needed for the next action. Reuse a known way when it suffices. Where several pattern Solutions supply the Method, show how their operations contribute to the outcome. Establish which results feed another operation and which must fit together at the point of use. If a needed result is unavailable, give the affected user an alternative or a way to revisit the choice. This is how a language of patterns expresses a way to act across cases. Use the mantra in C.39.WA:4.3 when an inference changes what the claim concerns, extends it to other cases, or changes which way applies. Name the relation and the grounds for that particular move. C.39.RO:4.2–4.5 develops the grounds for generalizing an operation.
Use C.39 to construct, connect, explain or change ways of obtaining the needed results. E.4.CM applies that general methodological work to a framework author's public account of one composite Method or several independent Methods. Those constructions can use the semantic result returned here; trying a proposed way can expose a missing distinction or reopen source comparison. A sufficient existing Method can be used directly.
Archetypal Grounding
Systems Engineering use and system concept
A Systems Engineering DPF author must decide what a practitioner should do when a proposed system concept is internally attractive but its connection to outside use is unclear. The author compares a proposed outside-before-inside instruction with the receiving ontology and an external systems-engineering account:
A lexical crosswalk that maps both using system and containing system to one parent system entry erases the use-versus-parthood distinction. The author can then no longer compare changing the project system's construction with changing how an external system uses it.
The comparison concerns priority within an iterative process. Starting from an internal construction can narrow the alternatives before the outside behavior and operating conditions are known. Begin with the outside-use question. Once the information needed for a particular design question is explicit, explore construction and feasibility while refining the use account; leave a dependent construction choice open while its decisive use conditions remain unknown.
The author records the provisional synthesis branch at local locator SE-CS-USE-01 with this claim: “For one Systems Engineering concept decision, first establish the proposed outside behavior, operating conditions and system-in-hand sufficiently to compare internal constructions. Develop and test candidate constructions against that use; refine both accounts as new information arrives, and reopen the use, boundary or construction when realization evidence defeats it. State the subjects and direct relations that change the decision.” This retains the proposal's outside-before-inside priority, reconciled with dependency-aware concurrent and iterative work.
Replay the clock choice. A team begins comparing gear arrangements for a clock. In this constructed case, asking where and how the clock will be used reveals a requirement for millisecond timing beside a rocket engine, with vibration and 30 g acceleration. The next result is an explicit timing-and-environment use hypothesis; the team compares and tests constructions against it before selecting one. Timing accuracy and vibration resistance can be investigated in parallel once their required operating conditions are stated. If feasibility evidence defeats the proposed clock use, revise that use or the system concept and compare again.
A key/lock case changes another premise: investigating the intended change from a locked to an unlocked door shifts the immediate engineering subject from the key to the lock arrangement, with the key treated as a subsystem in the proposed decomposition. That change of focus does not by itself change the product SoI: recover the extended enterprise, its product and agreements under A.1.STM:4.1 when that designation matters. Recover the selected boundary before comparing mechanical, electronic or biometric realizations. The retained function stays visible while the construction alternatives change.
In this worked case, the later Systems Engineering DPF content decision cites SE-CS-USE-01 with this source edition and records used as proposed. The applied use-and-system-concept move stays in the Systems Engineering DPF; general claim, parthood and function-bearing rules stay in FPF. Reopen SE-CS-USE-01 when a source changes the linked action or boundary, a counterexample shows that the proposal cannot guide the decision, or the receiving decision changes its intended use.
Organization change contrast
An organization asks why moving the point where a branch accepts a job changes which office can correct it. One local account follows the work that transforms the job. Another follows the agreement that transfers responsibility for it. Comparing their words for handover can establish what each office means. The requested explanation also needs the rule connecting acceptance to the power to order a correction.
Suppose the offices agree that the originating office can order a correction until the receiving branch accepts the job, and the branch can do so afterwards. A job can leave the originating office on Monday and be accepted on Wednesday. The work history and the transfer rule together place Tuesday’s decision with the originating office. Move acceptance to Monday while keeping every processing step and its time fixed. Tuesday’s decision now belongs to the branch. The difference explains the effect of changing the arrangement wherever that rule governs. An authoritative record identifying the office on Tuesday could settle a request for its name. A proposal for a different allocation of responsibility requires a further choice and a rule the offices can use. For a proposal to identify the two offices’ uses of handover, the same comparison warrants keeping dispatch and transfer of authority distinguishable. Dispatch can remain fixed while responsibility changes hands on a different day.
Scientific classification inquiry
A cross-disciplinary data framework proposes one shared classification relation between a laboratory ontology and an administrative classification. The available alignment suggests candidate correspondences, but the current edition of a load-bearing clinical source is unavailable. The author records an unresolved-inquiry claim about that bounded comparison, keeps the administrative-only use available, and names inspection of the missing clinical edition as the next source action.
Cheap anti-case
An engineer consults a handbook and its later edition only to recover the current pump tolerance. The later edition gives the needed claim, and no ontological difference changes the authoring decision. The engineer uses the tolerance, cites the later edition and stops without semantic synthesis.
Compare causal accounts on one sensor case
An author is deciding whether an explanation may treat inference about a hidden condition and changing an observed output as the same operation. Recover the source-local operations first. In the Introduction and “The Generative Model”, Parr and Pezzulo (2021) describe hidden-state inference together with policy-conditioned transitions and preferred outcomes. Pearl, Glymour and Jewell's corrected primer, p.55 describes intervention as replacing a generating mechanism.
Use the binary load/alarm case in C.28:4.4. Under exact inference from the same stipulated joint distribution, either account obtains a high-load probability of 0.9 after an alarm and 0.1 after an ordinary low reading. In the separately specified intervention model, setting the alarm output low leaves the load mechanism unchanged, so the high-load probability is 0.5.
An active-inference action can be compared with that intervention after its state-transition and observation consequences are specified. An action that changes only the output agrees on this load prediction. An action that initiates cooling requires the later load transition and its time order in both models. Comparing which action to select also needs the objectives or preferences used in selection.
Both accounts give the stated observational inference from the same joint distribution. Agreement about an action requires a correspondence between the specified mechanisms.
The receiving explanation decision can adopt this contrast and keep the two questions explicit. Reopen the comparison when the receiving question, source account, inference approximation or action mechanism changes.
Bias-Annotation
Scope. Use this pattern for a bounded inquiry whose answer depends on how source accounts relate. When that answer informs a method or framework, carry the established relation and its conditions into the operation or explanation that uses it.
This pattern favors explicit source roles and positive integration over a single authoritative vocabulary. That stance can overvalue synthesis and make every disagreement look like a new framework contribution. The cheap exits, contrast branch, unresolved-inquiry branch, and subject-pattern placement test keep the work proportional. Domain knowledge, lived practice, institutional authority, and empirical evidence remain source-local until a receiving claim states how they are used.
Conformance Checklist
CC-F0.2-1Before the comparison, state the receiving question, who needs the answer and which possible distinction or connection could affect it. Report the agreement, complementary grounds, difference or remaining limit that the comparison actually establishes.CC-F0.2-2Every load-bearing source claim has a recoverable EntityOfConcern, effective ReferenceScheme, edition, and source role; locators are not treated as source-use relations.CC-F0.2-3The source cut covers the intended use, known rival explanations, action-changing counterexamples, transfer limits, and material non-fit by relevance rather than by a universal count or distance threshold.CC-F0.2-4Similarity, correspondence, Bridge, framework fit, orG.2pack conformance is used as input and not as synthesis truth or permission to merge.CC-F0.2-5The result is one provisional synthesis claim, contrast claim or unresolved-inquiry claim whose content, subject, interpretation basis and sources are recoverable. Add a stable local locator when later citation or replay needs it.CC-F0.2-6The result identifies the inference, action or framework decision its grounds support and what remains open for that use. Keep any decision record required by the recipient.CC-F0.2-7The result identifies the source claims and editions it relies on, why each matters, retained differences and limits, and material reopen conditions.CC-F0.2-8An unresolved-inquiry claim identifies the open premise, its consequence for the receiving question, what remains warranted and any attainable source or action that could settle it. A missing fact alone leaves the disputed claim open.CC-F0.2-9Domain roles, methods, examples, evidence bases, and refresh lines stay in their subject DPF; the reusable synthesis move remains independent of those DPFs.CC-F0.2-10The author stops after the smallest result that answers the receiving question and opensG.2or later Part F work only for a named additional use.
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits:
- framework authors can produce a useful first claim before a complete account or UTS exists;
- source differences that alter action remain visible instead of being normalized away;
- small and large synthesis investments have separate methods that can reuse the same identified source claims; and
- DPFs reuse one method while retaining their own doctrine, evidence, examples, and currentness.
Costs and trade-offs:
- authors must recover source claims and editions rather than cite files or bibliographies as wholes;
- a contrast or unresolved inquiry can be the correct result even when a positive synthesis was expected; and
- later reliance needs the evidence-use, decision, admission, assurance, or publication claims required by that use.
Rationale
A shared vocabulary can help locate the claims to compare. The comparison earns its cost when it establishes what the sources jointly warrant, where their accounts differ, or which evidence would settle a consequential disagreement. That result can answer an inquiry directly, change a method or inform a pattern. One bounded comparison often suffices.
The three result branches keep the method constructive without forcing agreement. A provisional synthesis opens a positive contribution, a contrast preserves a decision-relevant difference, and an unresolved inquiry identifies the open premise, what remains warranted and any attainable way to settle it. Ordinary C.2.1 claim identity keeps those results separately revisable.
SoTA-Echoing
For the authoring question in :5.1, a correspondence-only answer can identify similar terms but leaves the outside-before-inside contribution undecided. A full G.2 pack adds a reusable corpus, claim sheets, inventories, alignments and downstream handoffs; these become worth their cost when later questions need that broader result. The bounded method instead retains source-local claims, compares the decision-changing difference and produces one result for the named decision.
Jaakkola's question and source-role discipline, Okoli's extraction and iterative synthesis, and Carroll–Brunton's treatment of non-fit inform steps 1–5 of that bounded route. These contributions organize the comparison; the working case determines what can be integrated and what must remain different. The trade-off is limited coverage for a smaller first result. Reopen the source cut when a known rival or counterexample can change that result, and open G.2 when the receiving work needs its broader reusable outputs.
Relations
F.0.2.MA uses a synthesis, contrast or unresolved-inquiry result to allocate Method contributions without erasing useful differences. F.0.2.RC tests whether the receiving explanation preserves the source's constructive contribution and follows changed premises to affected uses. These operations can stop on an already adequate synthesis and do not impose corpus-wide extraction on a bounded question.
- Builds on:
F.0.1for source-local meaning andF.1for a relevance-based source cut. - Uses when current:
F.9for an asserted relation between exact cross-context SenseCells;C.2.1for every recorded result claim;A.2.4and the applicable direct relation pattern for a later evidence or decision use. - Coordinates with:
F.2-F.8,F.14,F.17, andF.18when the next authoring question requires harvesting, clustering, Concept-Set, naming, or UTS work. - Coordinates with:
G.2as the optional broad harvesting method. Identified claims and provenance from aG.2pack may supply inputs here; the pack does not replace the receiving comparison. - Coordinates with:
E.4.DPFfor DPF entry and result placement,E.10.ARCHfor DPF-local wording entries under the shared restoration method, andG.11for source-currentness and refresh work. - Coordinates with: C.39 constructs, connects and revises the operations of a method. E.4.CM develops the public explanation when a framework author undertakes this work. A comparison here can supply a premise or expose an unresolved condition for an operation. Trying the method can in turn reopen that comparison.
- Boundary: Domain claims produced through this method remain in the subject FPF pattern or named DPF selected by the later content decision.
F.0.2:End
Synthesize and Allocate Method Contributions
Type: Method pattern Status: Stable
Problem frame
Use this when. You have recovered useful Method-related material from several sources, but still have to decide which contributions belong together and where they should be used. Documentary sources and accounts recovered from practice may use the same words for different actions, or different words for the same useful move.
Begin with the work the intended reader must be able to do. Which part of their explanation or way of working depends on the disputed contribution? This identifies the description or pattern in which the reader needs to find it. If the commission asks only where one known contribution belongs, settle that allocation directly.
The first useful result is one justified allocation. For example: “Use the existing coherence-checking Method; keep this source's narrower qualification window with the affected description claim; leave the observed shortcut as a candidate account.” Stop there if it settles the receiving decision.
Use an already sufficient F.0.2 synthesis or existing pattern result directly when no Method-content allocation remains. A.3.1.DR recovers documentary contributions; A.3.1.MR, and conditionally RMP.10, recover accounts from Work evidence. F.0.2.MA starts from their results.
Problem
Recovered material rarely arrives in reusable units. One chapter mixes a Method, a tool recommendation and a permission condition. Two sources agree on the main action but disagree on when its result may be reused. A practice account records a shortcut without establishing that the shortcut is a reusable Method.
A vocabulary crosswalk or chapter-to-pattern table does not decide these differences. Premature merging hides conditions; excessive splitting creates duplicate accounts; assigning every remainder to a new pattern turns source coverage into a publishing quota. Later users then cannot explain why a contribution was included or which source change requires repair.
Forces
Solution
F.0.2 returns a source-grounded synthesis, contrast or unresolved inquiry. To allocate Method content, decide which existing operation answers the question, what constructive condition remains, and which body or direct source must carry it. Use the synthesis directly if it already supplies that allocation.
Reconcile meanings before assigning destinations. Preserve every difference that changes the receiving Method use, and show how to return to the sources behind the allocation.
Work one allocation
- Name the receiving use and contribution. Identify the description, Method question or pattern-language use being changed. Take the recovered source claim together with its difficulty or opportunity, proposed action, result, conditions and source location.
- Keep the input roles distinct. A documentary claim remains what that source asserts. A returned MR account keeps its candidate, rival and evidence limits. A current FPF/DPF result keeps its declared scope. A source's report that work occurred is not independent occurrence evidence.
- Resolve only the meaning difference that matters. Use F.0.1 when a source expression is ambiguous, F.1 when the source cut is insufficient, and F.0.2 when cross-source synthesis is needed. Use F.0.2’s provisional synthesis, contrast or unresolved-inquiry result, with its source edition, grounds and any disposition already made. Preserve a locator when later citation needs it; a similarity score does not replace the comparison.
- Subtract the current answer by value. Open the current FPF or DPF supplier that appears to cover the contribution. Compare the practitioner question, action, first result, applicability and stop. Use the supplier directly if it answers those positions. When a source adds a warranted Method-specific condition or move, retain that remainder and explain its receiving use.
- Decide whether the contributions can share an account. Compare reusable action, participant meanings, preconditions, result, variation and stop. Merge supported common content while retaining different conditions. Split incompatible actions or statuses. Keep an unresolved alternative when the evidence cannot distinguish them. If the reader must use several contributions together, explain where the outcome of each is used and which conditions must hold jointly. Independently produced contributions may have to refer to the same item or fit the same version. An account of the encompassing work can carry that explanation while each source retains its own method. Use A.3.1.QM and B.1.5.CM for an actual claim of Method identity or composition.
- Assign the contribution and its remainder. Identify the existing result to reuse, the addition or change to a Method description or pattern, the material to keep at its source and any unresolved question. Allocate the shared construction by its premises. A Method, corpus or publication can be the subject of a transdisciplinary operation. Retain a domain contribution where its action depends on that discipline's substantive knowledge; do not infer either Core placement or exclusion from the subject's name.
- Make the decision usable and revisable. Give the reader the contribution, reason, destination, retained difference, source return and next action. If a later user will cite the allocation, keep a stable local locator with its edition. Name the source or supplier change that reopens it.
One justified placement can finish an allocation. If the reader must choose or change an arrangement, show how its requirements determine which contributions to retain, replace or obtain elsewhere. Put the reason where the reader makes that choice. A list of destinations can remain the useful index to that account.
Preserve what the destination cannot carry
A pattern may present the reusable action and still leave a worked derivation, local convention, dataset or detailed instruction at its source. State what remains there and how to recover it. A promised self-contained use cannot depend on an undisclosed omission.
When contributions merge, retain the reason that their common action is usable under the resulting conditions. When they split, preserve the action-changing difference and the relation between the resulting accounts. Shared terminology, publication order and apparent granularity are not sufficient reasons for either decision.
The allocation concerns source contributions and their uses. It is not a Method-composition result. Method parts are independently identified Methods related through B.1.5; a pattern body, tool, source fragment or source-recovery step is not a part by position.
A compact retained answer
For a reusable allocation, this amount of information is normally enough:
Use ordinary prose instead of a table when it is clearer. A retained allocation is a claim-bearing account about the named allocation question; its claim content, subject and effective reference scheme determine its C.2.1 identity. Its locator is a recovery aid, not a Method identity or an acceptance decision.
Archetypal Grounding
One review action, two conditions and a candidate shortcut
In this constructed case, a Method engineer is preparing guidance for reusing an earlier review of a Method description.
- Handbook H, edition 3, §4, says to reuse the earlier review when only the description's display changes.
- Paper L, edition 1, §2, says that reuse also depends on the same reviewed claims, question and qualification window.
- Returned account W, version 2, reports a team that reused a review after shortening a warning. It retains two rival explanations: the warning was redundant, or the team missed an important condition.
The documentary dossier preserves H and L separately. W remains candidate evidence. Current A.3.1.CC already supplies the coherence check and owner-specific correction; A.3.1.MV distinguishes a description change from changed reusable Method semantics.
The engineer's allocation is:
The first useful result is the conditional reuse answer. A description author can state the condition and the return to A.3.1.CC.
A changed source does not erase a former result
Later, only excerpts of H are available. Neither they nor the retained dossier contain the argument needed for the new use; the dossier gives its locator. Report that the argument cannot be rechecked and limit the dependent use to what the available evidence supports. Preserve the earlier result with its historical basis, L's independently supported condition and unrelated allocations. Reopen only what relied on the missing argument.
If the complete argument had already been preserved with a suitable basis, loss of current website access alone would not establish that the earlier claim was wrong.
Bias-Annotation
A source familiar to the engineer can become the vocabulary into which every other source is translated. Counter this by stating each source's action-changing meaning before merging. A desire for a neat language can also make every contribution look like a new Method; retain the non-Method and unresolved material that the receiving use needs.
The worked case illustrates allocation decisions. It does not establish the effectiveness of review reuse in an external project.
Conformance Checklist
- The receiving use and the contribution being allocated are recognizable.
- Source assertions, recovered candidate accounts, current supply and occurrence evidence retain their different roles.
- Current supply was compared at the action, result, condition and stop that matter.
- Every merge or split preserves its semantic reason; unresolved differences remain visible.
- The destination states the useful change and the direct-source or instructional remainder.
- Method identity and composition claims have their own basis.
- A changed premise leads to an identifiable reconsideration or stop.
Common Anti-Patterns and How to Avoid Them
Consequences
The engineer gains a language or description whose source decisions can be challenged and repaired locally. Reuse becomes more selective. The new body omits material that a current supplier already provides and preserves differences that matter to the receiving use.
The cost is retaining the few source and dependency facts needed for later reconsideration. For a one-off question that F.0.2 or a direct source already closes, another allocation account adds no value.
Architectural Rationale
Source recovery answers what the material says. Conceptual synthesis answers what a cross-source comparison warrants. F.0.2.MA uses those results to settle a Method-specific distribution: what reusable action and condition belong in which description or pattern, and what must stay outside.
Keeping this result separate from A.3.1.DR lets documentary recovery remain faithful even when a later allocation rejects a source proposal. Keeping it separate from Method admission prevents a placement decision from creating a Method. Retaining one small allocation before a corpus-wide map keeps the first useful result affordable.
SoTA-Echoing
For the allocation question, combine source-local conceptual comparison with Method-content and use analysis. F.0.2 supplies the former; A.3.1.DR, A.3.1.QM, A.3.2.DE and the other direct Method-development results supply their respective subjects.
In the H/L/W case, this allocation preserves a usable conditional review answer, rejects an unsupported shortcut-Method identification and leaves tool instructions at their source. A generic synthesis note is sufficient when it already supplies those destinations and returns; allocation is useful when one of them remains missing.
Daalhuizen and Cash's Method Content Theory (2021) supplies questions about goal, procedure, rationale, framing and mindset. Stacey et al. (2025) reinforces the need to retain the knowledge and use conditions around procedural guidance. Adopt these as source questions; do not turn their different models into a universal field schema or FPF parthood test.
Reopen this allocation method when current supply makes a separate result unnecessary, a source difference defeats a merge, or maintaining the allocation costs more than the source reconstruction it saves.
Relations
- C.39.RP supplies repertoire and lineage; A.3.1.DR supplies documentary recovery.
- A.3.1.MR and conditionally RMP.10 supply recovered Work accounts with their limits.
- F.0.1/F.1/F.0.2 govern source meaning, selection and conceptual synthesis; E.4.DPF governs framework-contribution subtraction.
- A.3.1.QM/B.1.5.CM qualify the Method claims; A.3.2.DE authors a description or improves a candidate Method account for a named use, retaining its identification status.
- E.8.ML uses the allocation to architect a language; F.0.2.RC challenges its recoverability and refresh reach.
- A.3.1.CC and A.3.1.MV receive contradicted claims and variant or non-variant maintenance questions.
F.0.2.MA:End
Test Constructive Coverage and Refresh Dependent Explanations
Type: Method pattern Status: Stable
Problem frame
Use this when. A Method-description language claims to preserve useful source contributions, but you need to find out whether a user can recover the promised action, conditions and source basis. Use it again when a source or use changes and you must identify the affected repair without discarding supported work.
As a Method engineer maintaining a source-grounded description or pattern language, begin with one promise: which source contribution should a user be able to reconstruct for this action, and what would count as a consequential loss?
The first useful result is one reconstruction answer: the contribution recovered, a deliberate omission with a usable return, or the missing or contradicted element and its next repair. A known one-claim contradiction already owned by A.3.1.CC needs no additional reconstruction exercise.
Problem
A source-to-pattern map can appear complete because every chapter has a destination. Yet the language may omit a stop, merge rival Method accounts, lose an evidence limit or hide the material relation that makes a continuation usable.
When a source changes, follow its dependencies to determine which claims need revision. A blanket rewrite can erase earlier valid results when only source access changed; a paragraph left in place can still carry a defeated premise.
Forces
Solution
F.0.1/F.0.2 recover and reconcile source meaning, while A.10.1 revalidates an affected reliance after a source change. Test whether the reader can recover the promised action, condition, join and return from the receiving explanation. A source citation or unchanged bytes alone cannot establish that the reader can make that recovery.
Attempt the promised reconstruction, state what was lost or recovered in the tested scope, and follow changed premises to the claims and uses that depend on them.
Make the coverage claim testable
Recover the reader’s intended work from the promise before consulting the candidate’s list of entries. State the source edition or bounded corpus, the candidate language edition and the contribution that should survive. Include the intended action, first result, applicability, stop and source/evidence qualifications that can change that use.
Choose the source cut by the promise. A source-complete extraction commitment needs its stated full recovery basis; a gap-only claim needs a recoverable prior baseline and a bounded delta. F.1 can select a question-relative source cut, but it does not replace an exhaustive domain review when that is what the claim requires.
Keep the candidate's allocation and source returns available. An absent or stale baseline is a result to return, not a reason to assume that the visible subset is complete.
Reconstruct in the direction the question needs
From a source contribution, try to recover its usable meaning through the language. Follow the actual reader entry and full bodies. Can the reader obtain the action and result, recognize the condition or stop, and recover the source basis or retained instructional remainder?
From a promised language answer, return to its supporting sources and allocations. Does the cited material support that action and status, or only a similar topic? A pattern that combines sources must retain the reason for the combination and the differences that affect use.
One direction may settle the question. Use both when the language claims source coverage and the risk includes unsupported additions. A held-out passage or a source condition omitted from the authoring route can expose a loss that the destination map misses.
For a promise about performing a composite Method, also recover a constituent action within its ongoing whole and the conditions that make that combination possible. Change an encompassing requirement and ask what must change in the action. A reader who can repeat the steps but cannot recover that dependence has not yet recovered this promised use.
Available lower-level skills and a known whole procedure can conceal a missing intermediate coordination. A promised actionable account must expose that gap or give a usable return to the contribution that closes it. Retain genuine earlier-result dependencies; a list of simultaneous activities alone does not explain constitution.
If the promise includes choosing or constructing the way of working, start the reconstruction with the material that reader would have before the choice. What choices about the way of working did the commission leave to that reader? When several results must work together, can the reader choose the parts and retain the distinctions their joint use requires? Check whether the account gives grounds for the choice and its bearing on the outcome. A task that supplies the pieces, their connections and the model can test the reader’s use of that arrangement. For a promise confined to its use, that can suffice.
State the reconstruction result
Use the smallest accurate disposition:
These are ordinary result descriptions, not a universal status taxonomy. An omission becomes a defect when it blocks a promised use; an irrelevant source detail does not have to become a pattern.
A positive reconstruction concerns the inspected claim and use. It does not establish Method identity, effectiveness, transfer or complete coverage of an uninspected corpus.
Follow a changed premise
- Name the changed source or use fact. Distinguish changed content, changed qualification window, unavailable access, defeated evidence and a different receiving question.
- Recover what relied on it. Identify the allocation, Method/candidate claim, description, pattern relation or reader use that consumed the changed premise.
- Classify the resulting change at its owner. F.0.2.MA handles allocation; A.3.2.DE description content; A.3.1.CC a contradicted maintained claim; A.3.1.MV a reusable Method variant or a non-variant change; E.8.ML the language arrangement. Use the direct source, evidence or access result when the question concerns the source, its evidence or access to it.
- Retain independent results. State why an unaffected result's relied-on basis still holds. A different file or nearby paragraph alone does not establish independence.
- Repeat the affected reconstruction. Compare the repaired action, condition, result and source return with the same promised use. If the promise itself changes, identify the narrowed or different use rather than report the original test passed.
If the dependency boundary cannot be recovered, widen the affected inquiry. Localized refresh is a conclusion about dependence, not a default limit on effort.
Keep the first use light
For one contribution, a short answer can name the failed promise, supporting or missing passage, affected destination, retained results and next action. Retain a more detailed account only when another user must repeat, compare or rely on it. Its C.2.1 identity follows its claims, named coverage question and effective reference scheme, not a map identifier alone.
Use a fresh reader when familiarity could supply a missing choice or inference. Give the reader the original commission and the information normally available at the point of choice. Obtain their answer before supplying a choice the commission entrusted to them. A desk reconstruction remains useful evidence, but it is not a fresh-reader or actual-enactment result.
Archetypal Grounding
A covered topic with a missing stop
This constructed case starts with handbook H, edition 3, §4: “Reuse an earlier review for a display-only change.” Paper L, edition 1, §2, adds the condition that reviewed claims, question and qualification window must still match. The reconciled contribution also requires a new source-dependent recheck to stop if its required argument cannot be recovered.
A candidate language has a “Reuse review” entry and cites H and L, but its body says only “reuse the prior review for display changes”. The destination map marks H §4 and L §2 as covered.
The map locates the topic, and the reader can recover the review action. The entry gives no test for matching the reviewed claims, question and qualification window, or for stopping a source-dependent recheck when its argument is unavailable. Return those missing conditions to the body that promises the decision.
After repair, the body states the matching-claim/question/window condition and the local source stop. Repeat the same use: the reader can now distinguish permitted reuse from a recheck lacking its basis. This is a constructed content check, not a measurement of review effectiveness.
Change one source without rewriting the repertoire
The later facts are:
Report that A1 cannot be rechecked and limit only the current uses that depend on its missing argument to what the available evidence supports. Keep its earlier result with the historical basis; do not infer falsity from lost access. Retain A2's condition and A3's uncertainty. If another pattern copied A1's assumption, it is affected even if its file did not change.
The repair is to recover the required argument, use a separately justified sufficient basis, or narrow the current use. A full repertoire rewrite would add no supported correction to A2 or A3.
When gap-only is not justified
A team has no recoverable earlier allocation and cannot tell which source claims the language relied on. It cannot prove that only a newly changed paragraph matters. Return the missing baseline and broaden recovery through A.3.1.DR/F.0.2.MA. Do not call the language source-complete or all other results unaffected.
Bias-Annotation
Authors can reconstruct a missing condition from memory and mistake that success for a property of the language. Use the route available to the reader and retain which information was supplied externally.
A coverage percentage can reward destinations while hiding a decisive lost stop. Treat counted items as a discovery aid unless their relation to the receiving use is established. Constructed cases and small reader probes retain their limited scope.
Conformance Checklist
- The source/language editions, receiving use and coverage promise are identifiable.
- The reconstruction tests actionable meaning, conditions, stops and source qualifications, not just the presence of a topic.
- Unsupported additions and deliberate omissions are distinguished from missing promised content.
- The result states its tested scope and evidence limits.
- Every changed premise has an affected claim or an unresolved impact boundary.
- Preserved results retain their own supporting basis.
- The repaired or narrowed use receives an affected reconstruction before closure.
Common Anti-Patterns and How to Avoid Them
Consequences
Coverage becomes a challengeable claim about useful reconstruction. Source changes can lead to small, justified repairs while leaving supported work available.
The cost is preserving the basis and dependence needed to make that judgment. Where those facts are absent, wider recovery is needed to establish the affected scope. The Method saves work only when its localized return costs less than the repeated reconstruction it avoids.
Architectural Rationale
A.3.1.CC begins with a potentially contradicted claim and locates its owner. This pattern can start earlier: a promised source contribution is missing, and the language may not yet expose the claim or relation that should own it. Reconstruction finds that loss and supplies the repair question.
Combining falsification with affected refresh keeps the original promise and its change history connected. A.3.1.MV decides whether reusable Method semantics changed; a missing description condition is not automatically a new Method variant.
SoTA-Echoing
Riehle, Harutyunyan and Barcomb's pattern-discovery and validation work uses established research Methods rather than treating the “rule of three” as sufficient validation. Adopt the need for explicit evidence and discriminating cases; source reconstruction is only one bounded check, not a substitute for their empirical programme or for domain trials.
F.0.2 and F.0.2.MA preserve source meanings and allocation, while A.3.1.CC/A.3.1.MV and G.11 retain their correction and refresh results.
Compared with a chapter-to-pattern map, the H/L case exposes a missing qualification condition that changes the review decision. Compared with a full reread, the A1/A2/A3 case retains L's condition and W's unresolved account while returning the one unavailable recheck.
These examples justify the useful form of the result, not a general claim of faster maintenance. Reopen when a lost dependency defeats localization, a source promise changes, a direct correction makes reconstruction unnecessary, or observed maintenance burden outweighs its benefit.
Relations
- A.3.1.DR supplies documentary recovery; A.3.1.MR and conditionally RMP.10 retain Work-evidence recovery.
- F.0.2.MA supplies allocations and unresolved source differences; F.0.2 supplies conceptual-synthesis results.
- A.3.2.DE/A.3.1.CC receive description and coherence repairs; A.3.1.MV distinguishes variants from other maintained changes.
- E.8.ML receives a failed language relation or entry; C.39.MC lets a user continue from the next needed result.
- C.37.CF can compare description revisions when the content/form question remains.
- A.3.1.WT, A.3.1.FT and C.11.MV retain enactment, transfer and worth questions beyond source reconstruction.
F.0.2.RC:End
Find and Select Sources for a Current Question
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative
One-sentence summary. Find inspectable material for the current question, expose search and access limits, and select the smallest sufficient source set when its claims, rivals or counterexamples need a justified basis.
Historical titles (informative). Domain-Family Landscape Survey; Question-Relative Source Selection. Aliases (informative). Source cut; question-relative source selection; question-led source finding.
Problem frame
Use this when. A current question needs source material, but you do not yet know where to find it, which words its sources use, or whether the available access reaches the needed passage. Use :4.4 to obtain inspectable candidate material with its consequential search and access limits. A catalogue reference, familiar phrase or ranked fragment may be a useful lead while still omitting the condition that changes the answer.
When candidate sources are already inspectable and more than one may change the answer, use :4.2 to justify the smallest sufficient source cut. Ask how each exact source and edition can change the answer or its limits, and retain the distinct contributions that matter. Finding and source selection can be used separately: a sufficient known source can answer directly, and a justified cut can start without a new search.
The source-selection branch returns one SourceCutNote : U.Episteme. Before treating the note as that episteme, identify the receiving question as one exact project entity. That question is the note's EntityOfConcern; the intended use remains in the ClaimGraph rather than becoming a second subject.
For that note, name the effective reference scheme: it must resolve the question, the cited source-edition designators, and the answer-changing role claims. The ClaimGraph also names deliberate exclusions, known limits or source gaps, and reopen conditions. If the question or scheme is still missing, keep an ordinary working note and return that missing value instead of claiming a C.2.1 episteme. A one-screen source note may represent the same episteme when it carries the same claims; its compact form neither admits a source nor contains its local meaning.
What this buys. You can reach material under unfamiliar wording, explain what the available search could have missed, and return passages whose contribution to the current question can be inspected. Later source use can begin from an inspected reason for selecting each source.
Not this pattern when. If one already identified source closes the question, use its claim directly or use F.0.1 for one source-local meaning. Use F.9 for an actual relation between recovered local senses and F.0.2 for a bounded conceptual comparison or synthesis. Use G.2 for a broader refreshable SoTA pack. When the receiving work requires an exhaustive, protocol-defined, or appraisal-bearing evidence review, use its domain method; F.1 does not replace that search, appraisal, or reporting discipline. Building or maintaining an access system is a separate engineering question; this pattern uses the finding means already available.
Primary results. Finding returns inspectable candidate material, why it matters, and consequential search or access limits; it may instead return the precise missing material. A passage with a short explanation can suffice; no special record is required. Source selection returns one question-relative SourceCutNote. Neither result establishes source truth, applicability, completeness, reliance or assurance.
Problem
The following problems can arise when people search without a guiding question or omit a source cut that the work needs.
- Word drift. Familiar terms silently change meaning across sources and editions.
- Canon lock. One influential standard is mistaken for the whole answer.
- Recency blur. An old edition remains load-bearing merely because it was used first.
- Category bleed. Designed structures, performed occurrences, system roles, statuses, permissions, and measurements are merged before their source claims are inspected.
- Source bloat. A large reading list hides which sources can change the current answer.
- Search substitution. Rank, similarity, or family membership is mistaken for relevance, truth, or completeness.
- Vocabulary lock. The question's first wording misses a source that uses another name or explains the needed relation differently.
- Access blur. A title, excerpt or failed search over one subset is treated as if the full source had been inspected.
Forces
Solution — find inspectable material and select by answer-changing role
Minimal vocabulary
- Exact source and edition. The identified publication, canon, corpus, practice source, or other claim-bearing basis being considered.
- Receiving question and use. What needs an answer and what that answer will serve. A plain concern can start finding. For source selection, independently identify the exact question addressed by the source-selection claims; it is the
SourceCutNote's EntityOfConcern, while the intended use remains claim content. - Candidate material. A passage or other source content that can be inspected for a contribution to the question. A source address without accessible content remains a lead.
- Query variant. Words tried to reach material for the same original question; any added interpretation is a hypothesis, not a replacement question.
- Answer-changing role. The inspected claim, limit, rival explanation, counterexample, transfer condition, material non-fit, or other stated contribution by which a source can change the answer.
- Source cut. The finite set of sources retained for that question and use.
SourceCutNote. One C.2.1 episteme identified by its source-selection ClaimGraph, the exact receiving question as EntityOfConcern, and the named effective ReferenceScheme under which that question, source editions, and role claims are read.- Domain family. An informative discovery label, such as workflow, provenance, services, sensing, types, or control. It carries no semantics and cannot admit, merge, or replace a source.
- Local search policy. An optional named way to prioritize candidate inspection using search terms, descriptors, citations, embeddings, distances, active learning, or portfolio readings. Its output is attention guidance, not a source-selection verdict.
Source selection, source identity, source-local meaning, source truth, source adequacy, cross-source relation, synthesis, reliance, and assurance are separate questions.
Source-selection method
Step 1 — State the receiving question and use. Say who needs the answer and what it will let them conclude or do. Where the question concerns an organized whole, consider what could change its answer while the separate accounts of its parts remain true. Seek grounds for the relation or common condition when the answer depends on it. A direct source about the whole can already cover that need.
Step 2 — Name candidate sources by exact edition. Use identified sources whose relevant claims can be inspected. A discipline or domain-family label may help discovery but is not a source.
Step 3 — Inspect the answer-changing role of each candidate. Ask what claim or limit from that edition can change the answer. Use F.0.1's ordinary branch only when that role depends on a disputed expression. Pause selection just long enough to name the exact source, edition, and passage, and say in plain language what the expression means there. Then return to source selection. Look deliberately for intended-use contributions, rival explanations, action-changing counterexamples, transfer limits, and material non-fit. One source may serve several roles.
Step 4 — Take the smallest sufficient cut. Retain the sources that together cover the needed grounds. Two passages may be useful only together, for instance when one identifies an item and another states a condition that applies to that item. Check what links their claims. Exclude a candidate when its inspected content adds nothing the receiving use needs. Do not optimize for a fixed count, diversity score, canonical status, or exhaustive appearance.
Step 5 — Record exclusions, gaps, limits, and reopen conditions. Name deliberate exclusions and why they do not change this answer. State a load-bearing source gap rather than treating an unavailable source as irrelevant. Say which question, use, edition, rival, counterexample, or transfer-boundary change would reopen the cut.
Step 6 — Return one SourceCutNote.
Identify it under C.2.1 by three values: the ClaimGraph produced by Steps 1–5, the exact receiving question from Step 1 as EntityOfConcern, and the named effective ReferenceScheme that resolves the question, cited editions, and role claims. Keep the intended use in the ClaimGraph. Use a one-screen representation when it helps the receiver hold the result in view. Detailed source analysis stays with the receiving comparison, synthesis, or evidence method.
Step 7 — Recover only the meanings the work needs. After the cut is stable, create an F.17 local-sense cell only for a retained expression that later reuse, a claim, a named receiver, or an actual relation needs. Do not postpone a meaning needed by Step 3 to this stage, and do not manufacture a cell for every retained source.
When the cut will support a SoTA claim
A source cut used to claim SoTA has a stricter role test because source relevance and source currentness do not establish the best-known answer. Use E.8:11 for the FPF definition of SoTA and its comparison roles. F.1 records the sources’ roles and leaves selection of the best-known line to that comparison. For each retained source, record one of those roles in plain wording: best-known-line candidate, serious current rival, failure or counterexample evidence, official or popular comparator, lineage only, or identity/currentness only.
The first three roles can supply answer-changing evidence. An official or popular comparator may stay only when its named defect is necessary to the comparison. Roles are assigned by contribution, not institution: an official standard or widely used practice may instead be a best-known-line candidate when its substantive answer wins against the serious alternatives. Official status, prevalence, maintenance, citation, freshness, or academic praise contributes nothing to that rank. An official catalogue, publisher page, or registry entry may verify the source's identity, edition, publication date, or maintenance state; it establishes neither truth, adequacy, nor SoTA rank.
For a SoTA claim, one source suffices only when it is a current critical synthesis that compares the serious alternatives for the named question and the author can state why no known action-changing rival or counterexample remains hidden. Otherwise retain the necessary rival and failure evidence or return an unresolved source gap. A one-source cut alone does not warrant confidence in the answer.
An original FPF or DPF answer is compared under E.8:11 on the same basis as a source-derived answer. Select the sources needed to expose serious alternatives, their limits and relevant failure evidence; do not require a prior external publication containing the authored answer. The original answer still needs that comparison, so its authorship does not restore the generic one-source exit.
The SourceCutNote records the E.8:11 roles and the missing comparison, but it does not select the best-known line. Use F.0.2 when the receiving work requires a cross-source synthesis claim. Use G.2 only when a broader refreshable evidence pack is justified; a bounded comparison does not require that apparatus by default.
The SourceCutNote
The note's exact EntityOfConcern is the independently identified receiving question from Step 1. It is not the note, its file or one-screen form, the retained-source list, or a question-and-use bundle. The intended use remains part of what the note claims. Name an effective ReferenceScheme that resolves the question, every exact source-edition reference, and every answer-changing role statement. If either the question or scheme is unresolved, return that gap and keep the text as an ordinary working note.
Its ClaimGraph states:
- the receiving question and intended use;
- every retained source and exact edition;
- the answer-changing role of each retained source;
- for a SoTA-supporting cut, each retained source's plain role and any unresolved rival, counterexample, or synthesis gap;
- deliberate exclusions and their reasons;
- known limits and load-bearing source gaps;
- the distinction between designed and performed material when it affects the answer; and
- the conditions that reopen the cut.
A one-screen source note is a compact representation of the same episteme when it carries the same claims. If its ClaimGraph differs, it is another C.2.1 episteme rather than a second F.1 result kind. A short form does not establish source truth, adequacy, or local meaning.
Find candidate material when sources or terminology are uncertain
Use this branch when the current question lacks material that can be inspected: its source, wording, location or available finding route is uncertain. Begin with the means you have. A paper catalogue, available files, a source's references, a colleague or a search service can supply a route. An already identified sufficient source can be used directly.
Keep the question and bound the available material
Keep the original question, what an answer would change, and any unresolved condition. A plain concern can start finding; it need not yet have the independent identity required for a SourceCutNote. Separate that concern from the words tried in a search. If you add an interpretation, mark it as a hypothesis to test against the material.
Source reading can change what you understand the question to be. Keep the original receiving concern and explain what the new material changes: a different needed result, a mistaken premise or an additional condition. Use B.5.PI:4.4 when that inquiry needs reconstruction, then return with the revised question. A query variant tries another way to find material for a question; a revised question can require different material.
Establish what the available access lets you inspect: which publications, collections or archive portions, which editions, and what access they provide. Full bodies, titles, excerpts and an index have different reach. Record a consequential missing collection or unavailable original. This boundary makes a later no-hit result interpretable and prevents an available subset from silently standing for the whole repertoire.
Choose an attainable first route
Choose a route that could reach the needed material at acceptable effort. Use a known address or a cheap direct text lookup when it can answer the question. When terminology is uncertain, use a likely contents area, an analogous case or an inspected passage to obtain words the source itself uses.
These are alternative starting points. Choose an additional route only when a consequential gap remains and that route could recover material the first route could not reach. For example, a body search can bypass a title-only index; an older term can bypass a current-name catalogue; a referenced original can expose a condition missing from an excerpt. Repeating the same limited representation with another tool does not repair that limit.
Keep a material search aid's policy recoverable in ordinary wording: searched corpus and editions, route, relevant filter or model, and how its output is interpreted. Retain a score's scale or threshold when it changes what is inspected. No special record is required for an ordinary lookup.
Try question-preserving variants and repair a blind spot
Form variants from the concern: the object being worked on, needed action or result, failure, and condition that changes the answer. Start with the parts likely to occur in the available text. Trying every detail as one phrase can miss a passage that explains the needed relation in different words.
Keep the original question alongside each consequential variant. Terms suggested by a colleague, an assistant or a promising source remain search hypotheses until their local use is inspected. If a variant drops a constraint to broaden discovery, restore that constraint while reading candidates. Broadening the query permits finding more material; it does not broaden what would answer the question.
After an unhelpful result, ask what that route could have missed. Change the particular term, scope or access path responsible for the suspected omission. When no available route can inspect a load-bearing original, return that access gap rather than inventing its content. Add another route for a reason, not to satisfy a fixed number of searches.
Inspect the source around the candidate
Open the candidate with its enclosing heading and the definitions, conditions, examples or dependencies needed to understand the passage. Say how its content could contribute to the current question. A useful candidate can supply a direct answer, an operation, a rival, a counterexample or a limit. Word similarity alone establishes none of those roles.
Compare the material obtained with the contributions the next use needs. New wording, another search route or a different document can still repeat the same contribution. When several aspects matter, identify the one still unsupported and why its absence could change the receiving answer; seek material for that gap. Several documents elaborating one aspect do not supply another. Keep an unresolved aspect visible when its source cannot be inspected. A brief statement of the remaining need suffices.
Distinguish an inspected body passage from a title or reference to an inaccessible original. An excerpt can be returned as inspected material with its missing context stated; it cannot supply an uninspected original's omitted conditions. Recover the source and edition as far as access permits, and retain an unresolved edition identity when it affects use. If a source-role judgement turns on a disputed term, use the ordinary F.0.1 source-local reading and return here or to :4.2.
For a candidate FPF Method, retrieve the relevant complete pattern and the conditions needed for its use through the available publication. Material finding supplies that text; E.11.PUR judges a proposed use and E.11.PUA guides application of a selected pattern. A promising fragment can be enough to continue finding while remaining insufficient to perform the Method.
Return useful material, its limits and the next move
Return the inspected candidates with source/edition and passage locations, the reason each matters, and any missing condition or access limit. A passage with a short explanation can be enough. Keep inaccessible leads and uninspected candidates visibly separate from material already inspected. State the searched scope and consequential route limits when they affect what the receiver may conclude.
Choose the continuation from the receiving need:
- Use a sufficient identified source directly for the subject question. Finding creates no source-cut obligation.
- Use :4.2 when several inspected sources could change the answer and a justified source cut is needed. Source selection then returns its own SourceCutNote.
- Use the relevant subject method for truth, applicability, synthesis, reliance or assurance. When the inspected contributions still leave an obtaining way unexplained, C.39 develops that connection or returns its precise gap. For proposed FPF uses, continue with E.11.PUR only when the receiving work requires that judgement.
- Return the exact missing material or access condition when the next use cannot yet be supplied.
Stop finding when the next use has the material it needs, or no affordable available route could presently change that result. Judge another search by the missing contribution it could supply, not by the availability of more terms or documents. A limited search with no useful hit returns “no useful material found by these routes in this accessible scope”, with the remaining gap. It establishes no claim of universal absence or completeness.
Reopen the affected finding when the question, needed result, source edition or access changes. Reuse a passage whose contribution and conditions still hold. A renamed file or new edition number alone does not require repeating the whole search; a changed relied passage or newly accessible original can. If a search representation changes, check whether it now reaches or hides material needed by the question before changing the received result.
Invariants
For finding, keep the original concern distinct from search variants, identify the material inspected, and return consequential access limits. A no-hit result is relative to the searched scope and routes. The following invariants govern the source cut when the work calls for source selection:
- Every retained source has an inspected answer-changing role for the stated question and use.
- The cut is finite, inspectable, and revisable; no universal count establishes sufficiency.
- Exact sources and editions remain recoverable.
- Source-local meanings remain local; F.1 does not merge or relate them.
- The result contains no F.9 relation, synthesis conclusion, truth verdict, reliance decision, or assurance claim.
- Domain families and search readings guide discovery only.
- Designed descriptions and performed occurrences stay distinct when that source difference matters.
- A changed relied premise reopens the affected cut claims; an unrelated edition-number change does not.
- One already identified sufficient source is the ordinary cheap exit; for a SoTA claim, the stricter critical-synthesis condition in
F.1:4.2.1applies. - A protocol-defined evidence review continues to follow its domain method.
Self-checks
For finding:
- Question-preservation test. Could a changed or omitted query term change what would answer the original question? Restore that condition when inspecting candidates.
- Reach test. What material could you inspect through the available route, and which consequential blind spot remains? Use a complementary route only if it could repair that gap.
- Candidate-return test. Can the receiver obtain the inspected passage, see why it matters and distinguish missing context or an unavailable original?
- Stop test. Does the next use already have sufficient material, or is its exact missing basis returned? Do not turn a scoped no-hit result into a claim of absence.
For source selection:
- Answer-change test. What can each retained source change in the answer or action? If nothing, exclude it or state the missing role.
- Rival-and-limit test. Is a known rival explanation, counterexample, or transfer limit still hidden? Add the source that makes it inspectable or return the source gap.
- One-source test. Does one already identified source close the ordinary question? If yes, stop without additional F.1 steps. If the cut will support a SoTA claim, take this exit only when that source is a current critical synthesis of the serious alternatives and no known action-changing rival or counterexample remains hidden.
- SoTA-role test. Has each retained source been classified under one of the comparison roles defined in
E.8:11? If not, classify it there before using the cut for a SoTA claim; do not recreate the role meanings in F.1. - Rank test. Did official status, popularity, maintenance state, date, or a catalogue check raise a source's SoTA role? Remove that inference; keep the check only as source identity/currentness evidence.
- Gap test. If the best-known line cannot be established, does the note return the missing rival, counterexample, or synthesis need as a source gap rather than naming the newest available source?
- Locality test. Have two sources been treated as saying the same thing merely because they use the same word? If so, use the ordinary F.0.1 branch on the exact passages before deciding their roles, then return to selection.
- Search-policy test. Did a score decide membership before source claims were inspected? Undo that decision.
- Memory test. Can the receiver hold the cut and each source's role in view? Remove non-changing material or split genuinely different questions.
- Reopen test. Can the note say what future change would require selection again?
Archetypal Grounding
Three heterogeneous source cuts
Role assignment, performed work, sensing, and execution
Candidate sources include BPMN 2.0 (designed workflow structures), PROV-O (performed activities and provenance), ITIL 4 (service vocabulary), ODRL 2.2 (permissions, prohibitions, and duties), SOSA/SSN (observations and results), and IEC 61131-3 (control-program execution). Retain only those whose exact claims change the receiving question.
The cut keeps false friends visible: a BPMN participant is not an RBAC role; a PROV activity is not a BPMN process; an SOSA observation is an act, not a status; an ITIL incident is not automatically a plant fault. F.1 exposes these source roles but does not settle their relations.
Methods, types, and measurement
SPEM 2.0 or ISO 24744 may change the reading of Method and MethodDescription; OWL 2 and formal concept analysis may change kind reasoning; SOSA/SSN and ISO 80000-1 may change measurement and quantity claims. The cut preserves the source-fixed differences rather than making method, concept, or measurement globally uniform.
Control, actuation, and services
Control-theory sources, IEC 61131-3, ISA-95, ITIL 4, and SOSA/SSN can contribute different claims about controller design, program execution, integration, service commitments, and observations. A current question may need only a subset. Keep an actuation claim separate from a service commitment, and inspect whether an incident report concerns a plant fault.
Minimal worked source cut
The project brief already identifies this receiving question: Can one contribution use “process” for both a workflow description and a performed occurrence? The exact question, not the note or source list, is the SourceCutNote's EntityOfConcern. The project names its effective ReferenceScheme Workflow and occurrence source cut, August 2026; that scheme fixes the three editions below and the ordinary-English role statements. The intended use stays in the ClaimGraph.
The resulting SourceCutNote is identified by that ClaimGraph, the stated receiving question, and the named reading scheme. Because this cut turns on the difference between a designed workflow and a performed occurrence, the project recovers any disputed process reading through ordinary F.0.1 while inspecting source roles, before stabilizing the cut. It adds an F.17 cell afterward only if later reuse, a claim, a named receiver, or an actual relation needs one.
Portable first-hour case — function-to-module allocation
“First hour” means a compact entry slice, not a deadline or completeness claim.
Receiving question. Which current FPF sources must a Systems Engineering author use to prepare a first function-to-module allocation teaching slice without collapsing functions into modules? The project identifies this sentence as one exact question before treating the note as a C.2.1 episteme; that question is its EntityOfConcern.
Receiving use. Prepare a slice that helps a practitioner generate and compare candidate bearers, modules, allocations, interfaces, and trade-offs instead of copying functions into a component list or selecting familiar bearers first. This use stays in the note's ClaimGraph.
Effective ReferenceScheme. FPFCoreReferenceScheme, applied to the FPF August 2026 edition and the exact section locators below.
Retained sources and answer-changing roles. All five are exact passages in that edition.
- A.6.F §§4.2 and 4.5. Separates the possible subjects hidden by function-like wording and requires function, bearer, flow, module allocation, and interface claims to remain distinct. This blocks the one-function-one-component shortcut.
- A.6.M §§4 and 4.3. Treats module use as claim content over exact holons, allows many-to-many or still-unallocated functional claims, and requires an actual interface specification when compatibility or substitution is claimed. This changes what an allocation row must show.
- C.30.TFS-REL §4.2. Relates functional structure to selected transformation-flow structure without identifying them. This exposes candidate flow topology, crossings, and correspondence limits that can change bearer placement.
- C.31 §4.5. Makes function-module alignment, interface burden, and flow-boundary alignment separate characteristics rather than one modularity score. This changes the trade-offs the comparison must expose.
- C.32 §§4–5. Starts candidate synthesis from functional demand and candidate bearers, keeps materially different configurations visible, and records expected gain, known loss, constraints, and source-return conditions. This prevents the source cut from pretending to choose the architecture.
Deliberate limits. The cut claims neither an exhaustive survey of allocation algorithms nor one module taxonomy or cross-sector optimum. It does not decide the final DPF pattern identity.
First result. One SourceCutNote whose ClaimGraph contains the five roles, use, limits, and reopen conditions; whose EntityOfConcern is the stated question; and whose effective scheme is FPFCoreReferenceScheme for FPF August 2026. The later comparison must expose unsupported capabilities, unallocated functions, unresolved interfaces, alternatives, trade-offs, and accepted losses.
Reopen when. A relied FPF passage changes, the question expands to external allocation algorithms or another domain, or the intended DPF use changes the required answer or boundary.
Readable reasoning moves
- Retain: “This exact source stays because this inspected claim changes the answer in this way.”
- Exclude: “No inspected claim from this source changes the stated use; record the exclusion and what would reopen it.”
- Return a gap: “This unavailable or uninspectable source may be load-bearing; the cut remains limited here.”
- Keep meanings local: “Selection puts both sources in view; it does not say their terms are identical or related.”
- Use search assistance: “The ranking tells us what to inspect next; we decide whether to retain the source from the claim we inspect.”
- Reopen: “This question, use, relied claim, rival, counterexample, or transfer boundary changed, so select again.”
Didactic distillation of source selection
State and independently identify the question; keep the later use in the claims. Keep a source because an inspected claim can change the answer, expose a rival or counterexample, or mark a transfer limit. Return one finite
SourceCutNotewhose ClaimGraph carries those roles, exclusions, limits, and reopen conditions, whose EntityOfConcern is the exact question, and whose named effective scheme resolves the question and exact editions. Search scores may guide inspection; they do not admit or exclude a source. Finish selection when the sources together cover the grounds the question needs, or return the specific gap that limits the answer.
Manual archive: a rule under an unfamiliar name
A workshop reader asks, “May a shared bench be used while an acceptance run is cooling?” The available fictional archive has three paper binders labelled Setup, Drawings and Bookings. Its card catalogue lists current equipment names. The reader has no searchable index or retrieval service.
The reader preserves the whole question, including the cooling interval, and bounds the search to those binders. The catalogue has no “cooling” entry. That is a limitation of this route. Browsing Setup produces a reference to an older equipment name and a drawing whose legend says “reserved interval”. The reader tries that term in the drawing register and obtains drawing D-14, edition 2, note 6. The note reserves the bench through the cooling interval; its exception allows sharing only before the run begins.
The result is that inspected note, its drawing and edition address, why the interval and exception matter, and the scope of the archive searched. It supplies a rule for the subject practitioner to use with the actual run conditions. It does not establish the present bench state or permission by itself. The reader can return this sufficient source without producing a SourceCutNote.
Change the question to “Which condition for sharing the bench changed between editions?” The reader obtains the available D-14 edition 1 note and compares it with edition 2 around the exception. Both passages become candidates. If the later comparison needs a justified two-edition basis, :4.2 selects their roles and returns a SourceCutNote. Change the question instead to “What word do these drawings use for the cooling interval?” The inspected wording can answer directly; further source selection adds nothing to that use.
The first failed route mattered because its catalogue vocabulary excluded the older name. The complementary route recovered source wording. This construction works with paper and ordinary reading; it requires no index engineering or automated ranking.
An uncertain method-library vocabulary
A project reader asks, “Two proposed pattern uses each fit our concern, but both need the same exclusive instrument. What helps us say which use should precede the other?” They can search the available full FPF publication but know no PatternID for that concern.
The first lookup, “suitable methods”, yields only general fit cues. The reader preserves the shared-instrument condition and tries variants such as “coordination”, “shared constraint” and “precedence”. A contents row points to E.11.PUR. The reader opens the complete pattern, including :4.3–:4.4 and :5.5, rather than treating the row as an applicable Method.
The returned material explains why exclusion alone establishes conflict without selecting A-before-B. An applicable schedule or priority rule can supply a direction; when it is unknown, the direction remains unresolved. The reader returns that passage and the missing-rule question. The later judgement uses E.11.PUR; a work schedule or performed work claim belongs to A.15. Finding the passage creates neither kind of judgement.
A title-only service that cannot search or return those body conditions would require another accessible route or return its limit. If the reader already has a current adequate coordination judgement, finding can stop there. The example's useful search terms are variants for this concern, not fixed names every corpus must use.
A changed edition with a known source
An engineer already has the needed fictional WS-4 procedure, edition 2, clause 7. An edition-3 notice names a changed inspection exception. The question remains whether the same inspection condition applies. The engineer opens edition 3 at that clause and reads the exception with its surrounding applicability conditions.
If the relied condition is unchanged, the passage remains usable for its earlier contribution, with the updated edition address. If the exception changed, the engineer returns the changed passage and the earlier passage when the comparison needs it, then reopens the affected subject judgement or source-cut claim. A changed cover gives no reason to search unchanged chapters again.
Now change the question to whether a particular attachment can be accepted. The retrieved clause does not settle an unknown material or operating condition. The subject engineer receives the passage and that missing premise. A found rule and a justified source cut supply different contributions from engineering applicability.
The original cannot be inspected
A review cites a 2025 original, but the reader can obtain only the review's excerpt and the original's catalogue record. The catalogue establishes a lead. The excerpt is inspectable at its own review address; the original's body and omitted conditions remain inaccessible.
Return those distinctions and identify the needed original or authorized access. Continue a use the excerpt can support; keep a dependent judgement open when its missing context can change the answer. Another search over the same catalogue cannot turn the original into inspected material. A later copy of the original reopens that gap; it does not require discarding other unchanged findings.
A changed question with two needed contributions
A workshop reader asks, “Which training material will reduce mistakes at handover?” The available fictional archive contains a course catalogue, an incident note, the current equipment register R-4, a handover procedure H-2 and several equipment-identification guides. The immediate concern is a receiver being sent to the wrong pump or being unable to tell whether its preparation is complete.
The catalogue points to a note about equipment labels. Reading it reveals that an old label is used for two different pumps. This gives the reader a more useful local inquiry: what must the handover distinguish so its receiver can identify the equipment and its preparation state? B.5.PI:4.4 permits this revision while retaining the original concern. Searching for another synonym of “course” would keep the earlier hypothesis in control.
The equipment register explains how current identifiers distinguish the pumps. Further guides found through different terms give more identification examples, but leave the preparation-state question unanswered. A reference in the register gives a route to H-2. The next search opens its completion condition in section 5, with the definition of the reported state and interval. The new contribution is that condition, not merely a new document.
The reader now returns R-4:2 for the equipment-identification basis and H-2:5 for the condition on reporting completed preparation. These supply the current requirement question. Stop searching and use them to examine the proposed handover. Choosing a course still requires judging the learning need. Identifying the handover requirements did not settle that question. If the available operations do not explain how to produce the required handover, use C.39 to develop the missing connection, with those passages as inputs.
If the procedure is inaccessible, return the inspected identification passage and the missing preparation-condition source. More identification guides do not close that gap. If both passages were already known and sufficient, open them directly. No additional route or SourceCutNote is needed for that use.
Bias-Annotation
- Gov: Authority, popularity, international status, or sponsorship does not decide membership; the receiving question and inspected claims do.
- Arch: The method keeps source roles explicit and the active cut small without claiming exhaustive coverage.
- Onto/Epist: A source, edition, inspected passage, inaccessible lead, source claim, local meaning, source-cut episteme and representation remain different.
- Prag: Direct use of one identified sufficient source is the cheap exit. Add a search route only when it can repair a consequential gap at acceptable effort.
- Did: A one-screen representation protects working memory; detailed analysis stays with the receiving work.
- Scope: Practitioners use F.1 to find material and justify source selection. Truth, adequacy, semantic relations, synthesis, reliance and assurance remain receiving questions.
Conformance Checklist
Static checks
Checks that refer to a cut or SourceCutNote apply when the work calls for source selection; ordinary finding needs no note identity.
-
SCR-F1-S01 (Question and use). The receiving question is independently identified as the note's exact EntityOfConcern; the intended use and the source difference that can change the answer remain in its ClaimGraph.
-
SCR-F1-S02 (Exact sources). Every retained source and edition is recoverable.
-
SCR-F1-S03 (Answer-changing roles). Every retained source has an inspected role; exclusions and source gaps are explicit.
-
SCR-F1-S04 (Finite and inspectable). The cut can be held in view without dropping a known answer-changing source merely to satisfy a count.
-
SCR-F1-S05 (C.2.1 identity of the source-selection result). The
SourceCutNote's ClaimGraph, exact receiving-question EntityOfConcern, and effective ReferenceScheme are all recoverable. The scheme resolves the question, exact source editions, and role claims. A file, source list, one-screen form, or question-and-use bundle supplies none of these values by form and does not become another source-selection result kind. -
SCR-F1-S06 (Locality). No cross-source equivalence, merge, or relation is asserted by selection.
-
SCR-F1-S07 (Temporal honesty). Designed and performed source claims remain distinct when the difference affects the answer.
-
SCR-F1-S08 (Family neutrality). No meaning, relation, or membership decision relies on a domain-family label.
-
SCR-F1-S09 (Named search policy). Any material search aid has a recoverable policy, corpus, method or model edition, and interpretation.
-
SCR-F1-S10 (No algorithmic gate). A search reading changes the cut only after inspection of source claims.
-
SCR-F1-S11 (Reopen conditions). The result says which question, use, edition, rival, counterexample, or transfer-boundary change reopens it.
-
SCR-F1-S12 (Domain-method boundary). A required systematic or appraisal-bearing review remains with its domain method.
-
SCR-F1-S13 (SoTA source roles). A cut used for a SoTA claim assigns every retained source one of the comparison roles defined in
E.8:11, records it in plain wording, and says which retained contributions can change the answer. F.1 does not restate or supersede the E.8 definition. -
SCR-F1-S14 (No authority or currentness laundering). Official status, popularity, maintained status, citation count, publication date, freshness, or academic praise does not promote a source into the best-known line. An official or widespread source may still earn that role from its substantive comparison. Catalogue and publisher pages support identity/currentness only unless their substantive claims independently enter the comparison.
-
SCR-F1-S15 (SoTA one-source guard and gap). A one-source SoTA cut is used only when that source critically synthesizes the serious alternatives and no known action-changing rival or counterexample remains hidden. Otherwise the cut retains the necessary comparison or returns a source gap.
-
SCR-F1-S16 (Finding return). The original concern, inspected candidate material, source/edition locations, contribution and consequential search/access limits remain recoverable; leads remain distinct from inspected content.
-
SCR-F1-S17 (Scoped stop). A sufficient known source can end ordinary finding; a no-hit result states its searched scope and routes and leaves a consequential access gap open.
Regression checks
- RSCR-F1-E01 (Edition change). Recheck only answer-changing claims that used the changed edition.
- RSCR-F1-E02 (Use change). Recheck which sources, rivals, counterexamples, and limits matter when the question or use changes.
- RSCR-F1-E03 (New false friend). Add a concise warning when recurrent wording confusion changes the receiving answer.
- RSCR-F1-E04 (Bounded cut). Remove non-changing sources or split genuinely different questions when the active cut can no longer be inspected together.
- RSCR-F1-E05 (Search drift). A changed taxonomy, corpus, model, descriptor, scale, distance, threshold, or rank cannot silently change membership.
- RSCR-F1-E06 (Finding reach). Recheck affected material and limits when the question, access or search representation changes; retain findings whose contribution and conditions still hold.
Common Anti-Patterns and How to Avoid Them
Consequences
An unfamiliar name or incomplete search view need no longer end finding silently. The practitioner can construct a route from the available source, words or references, repair a consequential blind spot and return material with its limits. That result may support a direct answer without requiring source selection.
The source cut becomes small enough to inspect, while a decisive rival, counterexample, or transfer limit cannot be discarded merely to satisfy a count. Later work can recover why each source was present, which edition was used, what the cut deliberately omitted, and what change reopens it.
The cost is explicit judgment. The author must say what each source can change and disclose a load-bearing gap instead of treating absence as evidence against a claim. Unknown rivals can still be missed, so citation chasing, expert leads, search aids, or a domain review method may remain necessary. Their outputs are candidates until source claims are inspected.
The cut is not evidence that its sources are true, sufficient, representative, mutually compatible, or safe to rely on. Those conclusions belong to the receiving synthesis, domain evidence method, F.9 relation, A.10 reliance account, or B.3 assurance claim.
Changes that reopen or revise a cut
- A source edition changes a relied claim, limit, or distinction.
- The receiving question or intended use changes.
- A new rival explanation, action-changing counterexample, or transfer limit appears.
- A formerly retained source no longer changes the answer.
- A language edition changes distinctions used by the result; treat editions separately unless the source supplies a mapping that preserves them.
- Search-policy drift reveals a candidate but never changes membership without source inspection.
- A later F.9 relation is proposed; keep source selection unchanged unless that proposal changes the source roles themselves.
Rationale
Finding is useful when the question and the source's wording do not yet meet. An exact address or lexical lookup is sufficient when it reaches the needed content. Variants and complementary routes earn their extra cost by exposing material that the first route could miss. Restoring the original conditions during inspection prevents broader discovery from changing the question. State what the reader could inspect and return the passage in context so the receiver can tell which claims it supports.
The receiving question determines which source differences matter. A source belongs in the active cut when an inspected claim can change the answer, expose a rival or counterexample, or mark where transfer fails.
This keeps the useful part of purposive and iterative sampling—selection for a stated analytic purpose—without importing another field's population, richness scale, saturation claim, or count convention. It also keeps the useful part of systematic-search and AI-assisted screening—recoverable searches and efficient candidate discovery—without treating search coverage, rank, or reporting conformance as source adequacy.
SoTA-Echoing
Choose the cheapest attainable route that supplies the needed material and its consequential limits. Source selection still leaves synthesis and assurance to their own methods.
Relations
E.8:11owns the canonical SoTA definition, comparison-role meanings, and positive comparison contract. F.1 prepares only the question-relative source cut and records those roles for the receiving comparison.
Builds on.
C.2.1identifies theSourceCutNotefrom its ClaimGraph, the exact receiving question as EntityOfConcern, and its effective ReferenceScheme, and distinguishes that episteme from its representation.F.0.1supplies a plain local reading during selection when a candidate's answer-changing role depends on a disputed expression.F.17supplies a durable local-sense cell after selection only when later reuse, a claim, a named receiver, or an actual relation needs one.B.5.PI:4.4helps reconstruct a question revealed by source reading while returning to the original receiving concern.A.11supplies the parsimony test; finiteness does not replace relevance with a fixed count.E.15supplies affected-premise reopening across source editions.
Related later uses.
F.0.2can compare or synthesize the exact sources, roles, limits, and reopen conditions returned here.C.39develops a missing obtaining way from inspected contributions; finding those contributions does not supply their connection.F.9defines any actual relation between distinct recovered local senses; F.1 asserts none.G.2supplies the broader refreshable SoTA-pack method when that heavier result is required.- Use
A.10for reliance andB.3for assurance; source selection establishes neither. E.11supplies public finding cues;E.11.PURjudges proposed pattern uses andE.11.PUAguides one selected use. Finding supplies the source material needed for those judgements without making them.- An author of a subject pattern or DPF may use the source cut directly when an inspectable source basis is needed but no cross-source synthesis is current.
F.1:End
F.2 — Source-Local Term Harvesting & Normalisation
“Harvest the source’s own words, recover what they mean there, and stop before comparison.” Type: Architectural pattern.
Status: Stable Depends on. F.1 for question-relative source selection; F.0.1 Source-Local Meaning Recovery; E.10.D1 Recovering What “Context” Means in Use; A.7 Strict Distinction; A.11 Ontological Parsimony. Coordinates with. F.3 Source-Local Sense Clustering; F.17 for an optional durable cell and basis relation; F.4 SystemRoleKindDescription when the recovered subject is a local system-role kind; F.9 only for an actual relation between distinct local meanings. Aliases (informative). source-local harvesting; local normalisation.
Intent & applicability
Intent. Turn usage in an exact source and edition into a small, auditable set of source-local expressions and meaning claims. Keep the source's idiom and the explanation the intended reader needs, while leaving cross-source conclusions to their own comparison.
Use this when. A source is known and the reader needs its own expressions to understand or use its contribution. Use F.1 if choosing which sources to retain is still part of the work. Re-enter when the selected edition, passage, language, or interpretation basis changes enough to change meaning.
Do not use this when. The words are already recovered precisely enough for the receiving question, or the current task is to relate two local meanings; use F.9 for that relation. F.2 creates no global term, kind, assignment, behaviour, obligation, or storage scheme.
Problem Frame
Three mistakes recur even after the right sources are selected:
- Word-centrism. A string such as process, role, or service is treated as though it carried one meaning everywhere.
- Over-normalisation. Spelling or morphology is forced into a house style and source-specific cues disappear.
- Premature structure. A short lexical note acquires behavioural, deontic, measurement, or kind claims that the source passage does not establish.
F.2 prevents these mistakes by tying each expression and local-sense claim to the exact source basis that supports it.
Forces
Core idea (didactic)
For each needed use, name the exact source and edition, the relevant passage, and the effective reference scheme under which the passage is being read. Harvest an attested LocalExpression, choose a minimally edited Local Normal Form (LNF), and explain its LocalSenseClaim at the detail needed by the receiving question. Add a separate label when a naming or didactic use needs it.
The ordinary result is a source-local lexical note. When later work needs a durable address for the meaning, F.17 can supply a SchemeSenseCell = <ReferenceScheme, LocalExpression, LocalSenseClaim>. State a basis relation when its support is established and that relation matters to the use. Keep the source, edition, passage, claim and any needed labels recoverable with the note. A relation to a meaning in another source is a separate F.9 question.
Minimal vocabulary (this pattern only)
- Exact source basis — the selected source, edition or version, relevant passage, and effective reference scheme needed to recover this use.
- Attested phrase — a short verbatim cue showing how the expression is used in that source.
- LocalExpression — the exact expression whose source-local meaning is being recovered.
- Local Normal Form (LNF) — the minimally edited surface used to cite that expression while preserving the source’s spelling, hyphenation, and casing.
- Additional labels, when needed — a source-faithful Tech designation and, for a stated didactic use, a Plain designation for the same local meaning.
- LocalSenseClaim — the claim saying what the expression means in this source use, at the least generality and sufficient detail needed by the receiving question.
- Source-local lexical note — the exact source basis, expression and LNF, meaning claim, supporting passages, and any labels needed by the receiving use. This is F.2's ordinary outcome.
- Homonymy signal — notice that the same string supports different local claims; this notice is not a relation between those claims.
- SchemeSenseCell — F.17’s optional three-part address, used only when durable reuse needs it.
Solution — three mental moves
Move A — Recover the basis
Ask which expressions in this source could change the answer the reader needs. For a glossary, recover how the source uses the words in its declared scope. For a question about the practice described, look for the wording of the relevant action or relation, including an expression that connects several passages.
Name the source and edition, locate the passages that support the meaning, and recover the effective reference scheme with F.0.1. Quote representative wording when it helps the reader recover that meaning. If the basis cannot be identified, leave the interpretation unresolved.
Move B — Name the local meaning faithfully
Ask: “How does this source express the meaning?”
Choose the LNF with minimal editing and explain the LocalSenseClaim. When the receiving use needs an additional label, keep the Tech designation in the source's idiom; add a Plain designation when it helps the intended reader recover the same meaning. Preserve meaningful head nouns and modifiers. Include a source-grounded equation, obligation or kind criterion when it is part of that meaning. Establishing a factual instance, performance, obligation or classification remains a separate claim under its direct pattern.
Move C — Fence the result
Ask: “What has this source use not established?”
A local reading establishes what the expression denotes in its source. Take a proposed relation between source meanings to F.9. When the question is whether particular work followed a process, recover the method and the performed actions through F.11. Matching words across two records cannot establish performance.
Normative guard-rails
- Basis visible. Every harvested expression MUST name the exact source and edition and the effective reference scheme needed to recover its meaning.
- Idiomatic normalisation. LNF MUST preserve meaningful source spelling, hyphenation, casing, and modifiers with minimal edits.
- Registers when used. A Tech designation remains source-faithful. A Plain designation is added for a stated didactic use and must explain the same meaning without broadening it.
- Minimal generality. The LocalSenseClaim MUST say no more than the source use and receiving question require.
- Category hygiene. A lexical note MUST NOT stand in for behaviour, obligation, measurement, kind, assignment, Work, or evidence.
- No cross-source claim. F.2 MUST NOT assert equivalence, subsumption, similarity, permission, or transfer between local meanings.
- Edition and language honesty. When edition or language changes meaning, recover a distinct source-local claim and interpretation basis.
- Parsimony. Keep expressions that the receiving use depends on, including a qualifying phrase that changes the meaning of a familiar head word.
Micro-examples
One worked note below supplies the exact source basis. The other items are recognition cues for candidate expressions; complete their source, edition, passage, effective scheme and attested use before relying on them as lexical notes. Their proximity asserts no relation.
- BPMN 2.0 (2011), effective BPMN scheme — expression and LNF
process; Tech process; Plain workflow graph; claim: “A graph of flow nodes and sequence flows specifying orchestration among participants.” - Worked note — W3C PROV-O Recommendation, 30 April 2013, §3.1, Starting Point Terms. Receiving question: what does
prov:Activitymean in this source? Effective scheme: the PROV-O vocabulary of that edition, withprovdenotinghttp://www.w3.org/ns/prov#. Expression and LNFprov:Activity; Tech activity (PROV-O); Plain occurrence that acts on or with entities; LocalSenseClaim: “Something that happens over a time interval and involves acting on or with entities.” Attested cue: “occurs over a period of time”. The note answers this local question; no durable cell or cross-source relation is needed. - ITIL 4 (2020), selected service-management use — expression and LNF
service-level objective; Tech SLO; Plain service target; claim: “A target value or range for a service characteristic.” - NIST RBAC (2004), RBAC scheme — expression and LNF
role; Tech access role; Plain permission grouping; claim: “A named grouping of permissions used in access-control assignment.” - SOSA/SSN (2017), SOSA scheme — expression and LNF
observation; Tech observation; Plain act of observing; claim: “An act applying a procedure to a feature of interest to obtain a result.” - IEC 61131-3, cited edition and runtime passage — expression and LNF
task; Tech task; Plain unit that calls programs on a cycle or in response to an event; claim: “A cyclic or event-driven runtime unit that invokes programs.”
Didactic heuristics (informative)
- Keep the source in your inner speech: say “process in BPMN 2.0” and “activity in PROV-O”.
- Prefer the source’s head noun and meaningful modifier: do not shorten service-level objective to objective.
- Preserve spelling or case when it carries source signal.
- Compress from attested use, not from a neighbouring theory.
- When role, function, process, or another trigger word appears, identify what the source passage is about before choosing a label.
Anti-patterns & remedies
Further recognition cues
The entries in this section are possible readings to inspect, not completed lexical notes. Recover each exact source basis before relying on the proposed gloss.
Enactment and sensing
Use the completed PROV-O note as one source-backed input. The BPMN, SOSA/SSN and ITIL cues still need their exact source bases before they can be used as lexical notes. After that recovery, a writer can ask how an SOSA observation result relates to an ITIL service target; listing the expressions together does not establish sameness between BPMN process and PROV activity.
Control and services
- State-space control source, cited passage and scheme —
actuation; Plain signal applied to the plant; claim: “A signal applied to influence plant state or output.” - IEC 61131-3, cited runtime passage and scheme —
task; Plain unit that calls programs on a cycle or in response to an event; claim as above. - ITIL Foundation: ITIL 4 Edition (AXELOS, 2019), §5.2.5, Incident management —
incident. The source defines it as “An unplanned interruption to a service or reduction in the quality of a service.” A qualifying interruption may be unreported. If a contract starts its response clock at notification, establish that event separately.
For the reported fault, establish whether the service suffered the interruption or reduction in quality described by the source. A record of an outage alone can fit both a planned stop and an unplanned interruption. If that difference decides the classification, use the schedule or another record that establishes it. For a contractual question about the response clock, apply its agreed start condition.
Kind, method, and knowledge sources
- OWL 2 profile source —
subClassOf; Plain class inclusion; claim: “Every instance of the first class is an instance of the second.” - FCA source —
formal concept; Plain extent–intent pair; claim: “A maximal object–attribute pair under the stated Galois connection.” - SPEM 2.0 and ISO 24744 sources —
method; Plain way of doing; claim recovered from the cited method passage. - SOSA/SSN (2017) —
procedure; Plain observation recipe; claim: “A description of how an observation may be carried out.”
Classifying an FCA concept as a root kind or identifying a Method from a procedure description requires a separate claim under the relevant pattern. Once recovered, the source meanings can inform those judgements.
Reasoning from a source-local meaning
- Localise. Hear the expression only in the exact source use now under examination.
- Recover. Identify the edition, passage, effective scheme, and source-local claim.
- Normalise minimally. Choose an LNF that preserves the source’s idiom.
- Label when needed. Keep the source-faithful designation; add a Plain label when the receiving reader needs it.
- Check scope. Tighten any sentence that says more than the passage supports.
- Signal homonymy. Note repeated spelling without claiming a relation.
- Stop. When the local note is clear enough for the receiving question, stop without adding a cell or comparison for the sake of completeness.
- Address only when needed. Use F.17’s three-part cell when later reuse requires stable addressability.
Relations
Builds on:
- Use the selected sources and editions, the receiving question and each source’s contribution to answering it. If the source cut still needs justification, use F.1 for that comparison.
- Use F.0.1 Source-Local Meaning Recovery to recover the exact source-local claim.
- Use E.10.D1 when vague context wording hides source, scheme, scope, situation, or use.
- Use F.17 for a durable
SchemeSenseCelland basis relation only when later reuse needs that address.
Constrains:
- F.3 clusters exact expressions and local claims under one effective scheme; it does not infer sameness across sources.
- F.4 may cite an exact F.17 cell when a local system-role-kind description needs it; a harvested expression establishes no kind.
- F.9 receives exact local meanings only when a proposed use needs an actual relation between them. F.2 supplies no Bridge, equivalence, direction, or use licence.
Used by. Part C patterns may cite the exact source expression and local-sense claim. The direct subject pattern still defines or constrains the value in the current claim.
Migration notes
- New edition. Recover the changed passage and scheme; keep the earlier note if its source use remains relevant.
- Idiomatic correction. Repair the LNF without silently changing the LocalSenseClaim.
- Ambiguity in one source. Recover two claims when selectional frames or entailments differ; F.3 tests the split.
- Language change. Determine whether the language edition changes source wording, reference scheme, or both; do not merge by translation alone.
- Tail pruning. Remove working notes that do not affect an active question; preserve cited source evidence elsewhere as required.
- Tool dialect. Keep it as one exact source basis and do not let it define other sources’ idiom.
Acceptance tests
Static conformance
- SCR-F2-S01 (basis). Every note names the exact source and edition and the effective scheme required to recover the claim.
- SCR-F2-S02 (idiomatic LNF). Each LNF preserves meaningful spelling, hyphenation, casing, and modifiers.
- SCR-F2-S03 (registers when used). Tech is faithful; an added Plain designation is explanatory without added scope.
- SCR-F2-S04 (lexical boundary). No note substitutes for behaviour, obligation, measurement, kind, assignment, Work, or evidence.
- SCR-F2-S05 (no cross-source claim). F.2 asserts no equivalence, hierarchy, transfer, or permission between local meanings.
- SCR-F2-S06 (minimal generality). Each LocalSenseClaim is no broader than its source use and receiving question.
Regression
- RSCR-F2-E01 (edition change). A changed edition produces a newly recovered claim only where meaning changed; earlier source identity remains visible.
- RSCR-F2-E02 (normaliser stability). LNF edits do not silently widen or narrow the claim.
- RSCR-F2-E03 (language honesty). Translation does not create unproved sameness.
- RSCR-F2-E04 (no stealth relation). New notes still contain no cross-source identity or use claim.
- RSCR-F2-E05 (head-term focus). The working set remains small and tied to the downstream questions it must answer.
Didactic distillation
“Name the exact source and edition, find the relevant passages, and say under which reference scheme you are reading them. Keep the source's own expression and explain its local meaning. Add a separate label when the receiving use needs it, preserving the same meaning. Stop when that answers the question. Use F.17 when a stable address is needed, and F.9 when a relation between two local meanings must be tested.”
F.2:End
Source-Local Sense Clustering
“Under one explicit interpretation basis, merge aliases that make the same local claim and split uses that do not.” Type: Architectural pattern.
Status: Stable Depends on. F.1 for question-relative source selection; F.2 Source-Local Term Harvesting & Normalisation; F.17 for the optional three-part local-sense cell; E.10.D1 Recovering What “Context” Means in Use; A.7 Strict Distinction; A.11 Ontological Parsimony. Coordinates with. F.4 SystemRoleKindDescription; F.7 Concept-Set Table; F.8 Mint or Reuse Decision; F.9 only when two exact local meanings need a tested relation. Aliases (informative). source-local clustering; sense consolidation.
Intent & applicability
Intent. Consolidate the expressions recovered by F.2 into a small set of source-local meaning claims under an explicit source and edition and an effective reference scheme. Merge aliases that the source uses interchangeably; split uses whose argument patterns, entailments, or practical consequences differ. The result stays local to its stated interpretation basis.
Use this when. Several expressions or uses from a selected source may carry one meaning, or one expression may carry several meanings that matter to the receiving question. Repeat when a changed passage, edition, or interpretation basis changes that partition.
Do not use this when. The local meaning is already clear enough, or the question is whether two meanings from distinct interpretation bases are related. That is an F.9 question. F.3 creates no kind, assignment, cross-source sameness, or permission.
Problem Frame
Source-local lexical notes often over- or under-differentiate meaning:
- Over-split: an abbreviation and its full form are treated as different meanings despite interchangeable source use.
- Under-split: one gloss covers incompatible argument patterns or conclusions.
- Source drift: different chapters, editions, or translations are blended without checking whether the interpretation basis changed.
- Didactic drift: the Tech and Plain labels begin to teach different things.
Test how each expression is used under the exact source basis. Counting strings cannot distinguish these cases.
Forces
Core idea (didactic)
Cluster by source use, not by spelling. Under one explicit interpretation basis:
- Same local meaning: expressions are interchangeable in the relevant source passages and no source-grounded test makes them support different conclusions.
- Different local meanings: their argument patterns, entailments, temporal stance, or practical consequences differ.
The cluster’s outcome is a LocalSenseClaim, with supporting expressions, any labels needed by the receiving use, and an optional counterexample. If durable reuse needs an address, use F.17’s SchemeSenseCell = <ReferenceScheme, LocalExpression, LocalSenseClaim>. The address preserves the claim's local interpretation basis.
Minimal vocabulary (this pattern only)
- Interpretation basis — the exact source and edition and the effective reference scheme used to understand the selected passages.
- Unit — a source-local lexical note from F.2.
- LocalSenseClaim — the claim that a cluster of source uses supports under that basis, explained at the detail needed by the receiving question.
- Supporting expressions — the F.2 expressions consolidated by the claim.
- Counterexample — a short source-grounded use that must not be covered by this claim.
- SchemeSenseCell — F.17’s optional stable address; it carries the scheme, an expression, and the local-sense claim.
- Usage cue — collocation, paraphrase, argument pattern, or entailment that suggests a merge or split; a cue is evidence to inspect, not a decision by itself.
Solution — how to think about clustering
Consolidate aliases supported by source usage
If spelling variants, abbreviations, or explicit synonyms are interchangeable in the relevant passages and do not change a conclusion, let one LocalSenseClaim cover them.
Example: ITIL’s service-level objective and SLO may support one local claim when the cited edition uses them interchangeably.
Split incompatible argument patterns
Split when the same head takes materially different participants or occupies a different place in the source’s propositions.
Example: in a BPMN diagram, event names a node. In a tutorial’s account of an outage, the same word can name an occurrence.
Split divergent entailments
If one use entails occurrence in time and another entails a design structure or capability, the uses support different claims.
Example: a PROV activity is a time-bounded occurrence; that claim does not describe a static algorithmic capability.
Prefer the coarsest adequate partition
Merge candidates when the recovered source uses establish that they are interchangeable for the receiving question and preserve every relevant conclusion. A failed attempt to find a distinction is insufficient when the source basis is too weak to decide. State which comparison remains unresolved. Split when a concrete counterexample would otherwise be admitted; do not split merely to fill a taxonomy.
Two distinct meanings can support the same action today. Ask whether, under the source’s own rules, one expression could correctly occupy the other’s place in the claim. A work order fixes the target at 20, and the observed value happens to be 20. A different observation changes the comparison while the order keeps the same target. One decision can use that comparison while keeping the two meanings distinct.
Keep labels honest
When the receiving use needs an additional label, keep the Tech designation in the source’s idiom. Add a Plain designation for a stated didactic use, explaining the same claim. Neither label is the value being described, and neither may widen the claim.
Address only when needed
Ordinary prose may cite the source, expression, and claim directly. Create an F.17 cell when reuse, comparison or traceability needs a stable address for the local meaning.
Outputs
For each interpretation basis used by the receiving question, F.3 yields a small set of local-sense claims. Each has:
- any additional labels needed by a naming or didactic use;
- the LocalSenseClaim and sufficient explanation of its meaning;
- the supporting expressions and passages;
- an optional counterexample that sharpens the boundary;
- an optional F.17 SchemeSenseCell when durable addressability is worth its cost.
Use these claims as reference points for reasoning. They can be stated in ordinary prose; make a record when the receiving use requires one. Displaying the claims together establishes no relation.
Invariants
- Basis explicit. Every LocalSenseClaim names the source and edition and the effective reference scheme needed to interpret it.
- Parsimony. Prefer the coarsest partition that preserves source-grounded differences relevant to use.
- Idiomatic Tech. A Tech designation, when used, remains source-faithful.
- Didactic Plain. A Plain designation, when used, aids comprehension without adding scope.
- Usage first. Claims follow source passages, not imported taxonomies.
- Counterexample rule. A source-grounded counterexample that matters to the receiving question forces a split or tighter claim.
- Category boundary. A local-sense claim does not establish behaviour, obligation, measurement, kindhood, assignment, or Work.
- No relation by clustering. F.3 establishes no cross-source identity, hierarchy, Bridge, substitution, or permission.
Self-checks
- Same-conclusion test. Do the source uses establish the same relevant meaning and preserve the conclusions needed here? If so, merge. If the available passages cannot settle the comparison, retain the unresolved difference and identify the missing basis.
- Argument probe. Substitute the participants from one use into the other. Split when this exposes incompatible source-required participant kinds or semantic roles. A false instance alone does not establish another sense: “Alice is licensed” and “Bob is licensed” can differ in truth while using
licensedwith the same meaning. - Entailment probe. Does one use imply an occurrence while the other implies a description, kind, or status? Split.
- Label inversion. When a Plain label is used, read it alone. If it invites a broader claim, tighten it.
- Test a counterexample. Use a source-grounded excluded use when it matters to the receiving question. If that use falls inside the claim, refine the boundary.
- Memory rule. Poor recall is a cue to inspect the partition for needless splits. Retain every source-grounded distinction that changes the receiving use.
Anti-patterns & remedies
Local-Sense Cards
An optional one-glance card may show:
- source and edition;
- effective reference scheme;
- Tech and Plain labels;
- LocalSenseClaim;
- supporting expressions and passages;
- counterexample;
- F.17 cell, only when its stable address is used for this claim.
The card is a display. Its fields do not create a new object or relation.
Worked example and recognition cues
Section 12.2 supplies one source-grounded clustering case. The other entries are recognition cues for possible clusters and boundaries; recover their exact editions, passages and interpretation bases before treating them as completed LocalSenseClaims.
BPMN 2.0
Process (workflow graph). Claim: a graph of flow nodes and sequence flows that specifies orchestration among participants. Supporting expressions may include process, process model, and business process where the cited passages use them for the diagram. Counterexample: “this process took five minutes” describes an occurrence, not this design claim.
Event (node). Claim: a typed diagram node marking starts, ends, or intermediates. Counterexample: “the outage event happened at 13:05” describes an occurrence.
PROV-O
Receiving question. Which of prov:Activity, http://www.w3.org/ns/prov#Activity and prov:Agent can share one local meaning?
Source basis. W3C PROV-O Recommendation, 30 April 2013, §1.3 fixes the prov namespace; §3.1 states the class meanings; §4.1 gives the class IRIs. The effective scheme is that edition's PROV-O vocabulary.
Activity. Consolidate the prefixed name and its full IRI: both address the class whose Tech label is activity (PROV-O) and Plain label is occurrence that acts on or with entities. LocalSenseClaim: an occurrence over an interval that acts on or with entities. A sorting algorithm as a reusable way of doing is not thereby such an occurrence.
Agent. Keep a separate local meaning. Tech agent (PROV-O); Plain responsible participant. LocalSenseClaim: something responsible for an activity, an entity's existence or another agent's activity. An RBAC permission role is not thereby a PROV agent.
The namespace expansion supports the alias consolidation; the temporal-occurrence and responsibility meanings remain separate. These two local claims answer the receiving question; add an F.17 address only when a later use needs that stable address.
ITIL 4
Service-level objective and SLO. One claim may consolidate the full form and abbreviation when the cited edition uses them interchangeably: a target value or range for a service characteristic. Counterexample: an observed availability value is evidence, not the target.
Incident. Claim: an unplanned interruption or reduction in service quality. Counterexample: a plant sensor fault that causes no service interruption or reduction in service quality.
SOSA/SSN
Observation. Claim: an act applying a procedure to a feature of interest to obtain a result. Counterexample: “20 °C” is a result value, not the observation act.
OWL 2
SubClassOf. Claim: every instance of one class is an instance of another. Counterexample: rdf:type relates an individual to a class.
EquivalentClasses. Claim: two class expressions have the same instances under the OWL semantics. Counterexample: owl:sameAs is individual identity.
IEC 61131-3
Task. In the cited usage a task is a cyclic or event-driven runtime unit that invokes programs. Identify the configured task when the question concerns its schedule, an activation when the question concerns one trigger, and the particular program execution when asking what work occurred. The same task can invoke a program many times.
Reasoning about a local-sense cluster
- Alias consolidation. Merge expressions only when the source uses them interchangeably for the current question.
- Argument split. Split uses whose required participants differ materially.
- Entailment split. Split when the meaning of an expression changes what the source claim entails. Two instances can yield different factual answers under the same meaning.
- Parsimony merge. Merge source-supported interchangeable uses when no relevant conclusion is lost; leave the comparison unresolved when the source basis cannot establish that result.
- Counterexample trigger. Tighten or split a claim that admits a concrete excluded use.
- Label check, when labels are used. Tech stays idiomatic; Plain explains without widening.
- Address check. Create an F.17 cell only when the receiving use needs a stable address.
- Edition check. Re-evaluate the interpretation basis before carrying a claim across an edition change.
- Check coverage. If a frequent source expression that matters to the receiving question has no local-sense claim, check whether one useful cluster is missing.
- Stop rule. Do not infer a relation to another local meaning from clustering alone.
Relations
Builds on:
- Use F.1 for the finite source cut and receiving question.
- Use F.2 for exact expressions and source-local lexical notes.
- Use F.17 for
<ReferenceScheme, LocalExpression, LocalSenseClaim>only when a stable address is needed. - Use E.10.D1 to keep source, scheme, claim scope, model use, and working situation distinct when context wording is encountered.
Constrains:
- F.4 may cite an exact cell but never infers a local system-role kind from it.
- F.7 displays exact local claims or cells and already obtaining relations for one stated comparison or use; F.3 clustering creates neither a row relation nor permission.
- F.8 compares proposed wording with exact existing designations and claims without treating either as the value being named.
- F.9 tests a relation only between exact local meanings whose interpretation bases differ. F.3 establishes no Bridge, cross-source sameness, substitution, or use licence.
Used by. Part C patterns may cite a local-sense claim under its exact source and scheme; the direct pattern defines or constrains the substantive value or relation in the example.
Migration notes
- Usage clarifies. Merge only when the recovered source uses support interchangeability for the receiving question; lack of a decisive comparison leaves it unresolved.
- Usage diverges. Split and add a counterexample when argument patterns or entailments pull apart.
- Edition changes. Recover the changed basis and claim.
- Labels drift. Repair Tech or Plain without silently changing the claim.
- Dormant claim. Omit it from the active comparison when it no longer changes the receiving answer; do not fold it into another claim without evidence.
- Bridge temptation. Record the question for F.9; do not answer it in F.3.
Acceptance tests
Static conformance
- SCR-F3-S01 (basis). Every LocalSenseClaim names its source and edition and the effective reference scheme.
- SCR-F3-S02 (labels when used). Tech and any Plain designation denote the same bounded claim.
- SCR-F3-S03 (fidelity and time stance). Each claim is grounded in cited source use, preserves any source-grounded design-time, run-time, or other temporal distinction, and contains no imported substantive calculus.
- SCR-F3-S04 (parsimony). The claim set is small enough for the receiving question.
- SCR-F3-S05 (counterexample). Ambiguous heads have a concrete boundary test.
- SCR-F3-S06 (no inferred relation). Clustering asserts no cross-source identity, hierarchy, transfer, or permission.
Regression
- RSCR-F3-E01 (merge soundness). Every merge has source-grounded support for the same relevant meaning and preserves the receiving conclusions.
- RSCR-F3-E02 (split necessity). Every split cites an argument, entailment, temporal, or counterexample difference.
- RSCR-F3-E03 (edition honesty). Changed editions are not silently absorbed into an old claim.
- RSCR-F3-E04 (label stability). Label changes do not change the claim unnoticed.
- RSCR-F3-E05 (downstream continuity). After a split or merge, direct citations and any F.17 cells remain unambiguous; no silent aliasing occurs.
Didactic close
“Start with one explicit source and interpretation basis. Merge aliases only when the source uses them interchangeably and no relevant conclusion changes. Split uses when their participants, entailments, or time stance differ. Add labels when a naming or didactic use needs them, preserving the same claim, and a short counterexample when a boundary test is needed. Use an F.17 cell only when the receiving work needs the address. To test a proposed relation between meanings from different sources, use F.9.”
F.3:End
SystemRoleKindDescription — Describing an Exact System-Role Kind
Type: Definitional (D) Status: Stable Normativity: Normative unless marked informative
Use This When
Plain name. Description of a system-role kind.
Use F.4 when a project needs a short, reusable description that makes one exact local system-role kind recognizable, teachable, and checkable. The described kind is a C.3 kind whose candidates must first be independently admitted as U.System. A candidate may be, for example, a person, team, organization, or non-human technical object; SystemRole does not mean “technical system only.”
Typical moments:
- a project has a durable kind name such as
InspectorSystemRole, but readers cannot recover which systems are candidates, what condition distinguishes members from relevant non-members, what change would make it another kind, the currentKindSignature, or the work-facing boundary; - a MethodDescription names a required system-role kind, but readers cannot tell which exact local kind must classify a candidate before an assignment can be checked;
- a kind name is starting to carry assignment, capability, Method, Work, permission, responsibility, evidence, publication, or status claims that belong elsewhere; or
- source prose says that a report, standard, dataset, theorem, dashboard, publication, or requirement has a “role”, and the writer must recover whether that wording denotes a system-role kind at all.
Primary EntityOfConcern. A SystemRoleKindDescription is one U.Episteme constituted under C.2.1. Its exact EntityOfConcern is one local system-role kind. Its ClaimGraph makes the C.3 recovery basis readable: the candidate domain, operative membership condition, intended member/non-member boundary, and continuity rule.
The description also names the current KindSignature edition, the description's effective reference scheme, useful source or practice provenance, and only the neighboring relations needed by the described use. Provenance helps readers locate and compare the definition; it does not identify the kind. The description is not the kind, a classification judgment, assignment occurrence, holder system, capability, MethodDescription, performed Work, status-use relation, or publication form.
Primary working reader. This description is for an engineer-manager, analyst, Method author or pattern author who needs to help people recognize the kind while keeping kind, candidate classification, assignment, capability, Method, Work, evidence use, status use, and publication use distinct.
First useful move. Name the exact local system-role kind, say in ordinary words which systems can count and what separates a member from a relevant non-member, cite the current KindSignature, and state the change that would make it another kind. Add source or practice provenance only to help readers find and compare the definition. Keep the recognition explanation no longer than the next classification, assignment, Method, Work, naming, or cross-local claim needs.
What goes wrong if missed. A description card becomes a hidden procedure, staffing record, access policy, permission badge, responsibility claim, evidence relation, status assertion, or Work log. The system-role kind then becomes confused with claims about its holders and with uses of epistemes or representations.
What this buys. A project gets a compact, readable description while each operational claim remains with the pattern that governs it. The kind stays recognizable; classification and assignment stay checkable; capability, Method, Work, evidence, status, responsibility, and publication claims stay inspectable instead of being smuggled into the name.
Not this pattern when.
- If the current question is whether a local system-role kind exists, how it is identified, or whether one candidate satisfies it, use A.2 with C.3 and C.3.2.
- If the current question is whether one admitted system and one exact local kind participate in an obtaining assignment, use A.2.1.
- If one assignment may satisfy one state condition during a window, use A.2.5.
- If the current question concerns admission substitution, incompatibility, qualification, a bundle, or another relation among system-role kinds, use A.2.7.
- If the current question is capability, use A.2.2.
- If it is about a Method, MethodDescription, WorkPlan, or performed Work, use A.3 or A.15 and the direct neighboring pattern.
- If an episteme is used as evidence, source, standard, requirement, publication, assurance input, status bearer, gate input, or decision input, use the direct relation. Do not classify the episteme as a system-role holder by wording.
- If only a durable name is needed, use F.18.
- If the current question relates two exact local system-role kinds, use C.3.3. If it relates two exact source-local senses, address them as F.17
SchemeSenseCellvalues and use F.9. Scheme difference or shared spelling alone triggers neither relation. - If bare role may mean a relation participant, declaration slot, representation position, ordinary wording, or another object, use E.10.ROLE and A.6.RSIR where relation recovery is needed.
Problem Frame
A local system-role kind often needs a recognizable description before people can classify a candidate, assign a system, compare kinds, or use the kind in a Method condition. A name such as InspectorSystemRole is not self-explanatory. Readers need to know which systems are candidates, what condition distinguishes intended members from relevant non-members, when that distinction continues, which KindSignature states it, and where neighboring claims begin. Source or practice provenance can help them locate and compare that definition; it cannot decide kind identity.
The recurring failure is to make the description carry too much. A compact card is tempting: put kind, status, permission, responsibility, evidence, capability, Method, assignment, Work, and publication cues into one “assignable” template. That convenience creates duplicate ontology. A standard used as a requirement source becomes a “standard role”; a report used as evidence becomes an “evidence role”; an access-control label becomes a system-role kind; a kind name becomes proof of capability or performed Work.
F.4 treats the description as an episteme about one exact local kind. It may cite neighboring relations but does not absorb them.
Problem
The following errors confuse a kind, its description or the claims made when that kind is used.
- Description and kind collapse. The description is treated as the local system-role kind.
- Description and classification collapse. A card is treated as proof that one candidate satisfies the kind criterion.
- Description and assignment collapse. A kind name or card is treated as proof that one system has an assignment.
- Description and capability collapse. A kind name is treated as evidence that a system can do the Work.
- Description and Method collapse. Kind criteria become a hidden procedure or MethodDescription.
- Description and performed Work collapse. A card is treated as evidence that Work happened.
- Status and episteme uses become system roles. Publications, standards, datasets, claims, and statuses acquire fake system-role classifications because they matter to reasoning.
- Relation positions become system roles. Participant meanings, declaration slots, interface places, and representation positions are mistaken for system-role kinds.
- Same-spelled kinds collapse or fragment. Shared spelling is treated as proof of one kind, or a changed practice or source as proof of two, without comparing the candidate domains, operative membership conditions, member/non-member boundaries, and continuity rules.
Forces
Solution
Constitute one SystemRoleKindDescription through C.2.1. Its ClaimGraph describes one exact local system-role kind, and that kind is its EntityOfConcern. It makes the kind's candidate domain, operative membership condition, intended member/non-member boundary and continuity rule recoverable, together with its current KindSignature. State the description's effective reference scheme for interpreting those claims.
The description may record source or practice provenance so readers can find and compare definitions, but provenance is not a kind-identity key. The description gives readers enough to recognize and check the kind while routing neighboring claims to their direct rules.
The following is a content checklist, not a relation signature or mandatory record.
Always make recoverable:
- the described local system-role kind;
- the candidate domain, operative membership condition, intended member/non-member boundary, and continuity rule;
- the current
KindSignatureedition and effective reference scheme; - a short recognition explanation;
- the candidate-system range, clarifying the broad A.1 meaning of System when the intended reader could otherwise misread it;
- the smallest direct-feature criteria or invariants needed by the current use; and
- the boundaries needed to distinguish this description from neighboring claims in the current use.
Add only when the current use depends on them:
- a
SystemRoleAssignmentStatePredicateor state-relation reference under A.2.5; - capability-condition references under A.2.2;
- Method or MethodDescription references under A.3 and A.15;
- durable-name or lineage references under F.18;
- C.3.3 or F.9 Bridge references; and
- a selected
BoundedModelUseStructureonly when that structure changes the described interpretation or receiving use.
These are claims or references in an episteme. They are not SlotSpec declarations and add no participant to U.SystemRoleAssignment or another relation. A card, table row, Method appendix, or pattern section may express the description. When availability matters, a separate E.24.PUB occurrence makes the exact description edition available through its publication form and carrier.
Content Meanings
A recognition description can finish here. If someone must decide whether to admit a candidate to work, use the description to identify the right criterion and test the candidate under it. Select any further questions from the actual admission conditions.
Description versus Neighboring Values
F.4 points to these loci; it does not copy their ontology.
Keep the description episteme, the exact local system-role kind it describes, the KindSignature that states the membership criterion, the effective scheme used to read the description, and any classification judgment about a candidate separate. Add an F.17 SchemeSenseCell only when a later use needs a stable local-sense address; cite a LocalSenseBasisRelation only when that relation obtains. An ordinary F.4 description requires neither.
Positive Construction Rule
Write a description in this order:
- Name the described local system-role kind and state its candidate domain, operative membership distinction, one useful member/non-member boundary, and continuity rule. Record source or practice provenance only when it helps the reader locate or compare the definition.
- Name the current
KindSignatureedition and effective reference scheme. - Give one short recognition paragraph, including the broad A.1 system range when a cold reader could narrow it incorrectly.
- State the smallest direct criteria or invariants that distinguish the kind.
- State an explicit boundary when the current use could otherwise confuse the description with classification, assignment, capability, Method, Work or another neighboring claim.
- Add neighboring references only when the receiving use depends on them.
- Use F.18 for a durable public name. Use C.3.3 only when the receiving claim concerns an obtaining relation between exact local kinds; use F.9 only when the receiving claim relates distinct F.17 local-sense cells.
Invariants
- One described kind. A
SystemRoleKindDescriptiondescribes exactly one local system-role kind. - Direct kind identity. The candidate domain, operative membership condition, intended member/non-member boundary, and continuity rule remain recoverable. A source, practice, taxonomy row, signature edition, or scheme helps locate, state, or interpret that basis; none decides identity by itself.
- Description boundary. The description is a
U.Episteme; it is not the kind, candidate, classification judgment, assignment, holder system, capability, Method, Work, or status-use relation. - System range. A candidate must independently pass A.1 as
U.System. No description or kind name performs that admission, andSystemRoledoes not narrow the candidate to non-human technical systems. - No hidden assignment. Classification under a local kind neither creates nor proves a
U.SystemRoleAssignmentoccurrence. - No hidden capability. Capability requirements may be cited, but the description proves no capability.
- No hidden Method. Method requirements may be cited, but the description is not a MethodDescription.
- No hidden Work. The description may support later Work-attribution checks, but it is not evidence that Work occurred.
- No status or episteme-use fusion. Status, evidence, source, requirement, publication, and assurance uses remain direct relations, not another description branch.
- Position discipline. Bare role that denotes participation, a declaration slot, interface place, or representation position is recovered through E.10.ROLE and A.6.RSIR rather than made a system-role kind.
- Name after meaning. Durable naming follows F.18 only after the exact kind, description, scheme, and local sense are recovered.
Reasoning Primitives
Use these schemas as thinking checks.
Worked Cases
Pump Inspector System Role
PumpInspectorSystemRoleKindDescription is a C.2.1 episteme whose EntityOfConcern is the kind currently named PumpInspectorSystemRole. An admitted system counts when it obtains readings for the named pump and declared condition characteristics in the applicable inspection situation and returns the named pre-maintenance judgment from those readings. A maintenance technician, inspection robot or service team can be a member. Apply both conditions to the system the question names. The team can meet the criterion when a technician takes readings in the team’s inspection and another member uses those readings to return the team’s judgement on that pump. To classify the technician personally, establish what the technician does. A report, or a system missing either condition, falls outside this kind.
PlantA-PumpInspector-KindSignature-v4 and Plant-A-Maintenance-Scheme state the current definition and interpretation. Plant A provenance locates that definition; it does not identify the kind. The same kind continues through an aligned edition while the candidate range and two-part distinction continue; a material change to either calls for another kind.
For this inspection, each predicate declaration supplies participant meanings and applicability, and the current case supplies the satisfying facts. Use A.6.F only if source wording first hides those claims behind function; it establishes neither predicate. If either predicate or its case facts cannot be recovered, record the exact A.6.RCD missing-governor or missing-information result instead of classifying the candidate.
The description says the kind concerns pump-condition inspection and does not denote repair. It may cite pump-inspection capability conditions or an inspection Method when a receiving Work claim needs them. Its boundary says that an inspection report is an episteme used through evaluation, evidence, source, or publication relations, not a system-role holder.
The description makes PumpInspectorSystemRole recognizable. It does not say that Robot-7 satisfies the kind, has an assignment, is capable of inspecting, has permission or readiness to inspect, enacted a Method, or performed Work. Those claims use C.3.2, A.2.1, A.2.2, A.2.8.PER, A.15, and the applicable evaluation or evidence relations.
Reviewer System Role and Review Report
ReviewerSystemRoleKindDescription may describe the kind currently named ReviewerSystemRole. An admitted system counts when it compares the named pattern claims with each selected scale in the applicable review situation and returns the named reasoned judgment with the assessed values or defects. A person, team, or review service satisfying both conditions can be a member; a review report, or a system that comments without applying the scales, is a relevant non-member.
PatternReview-2026-Reviewer-KindSignature-v2 states the current condition. The PatternReview source locates that definition; it does not identify the kind. The same kind continues only while the candidate range and substantive-review distinction continue under aligned editions.
In this review case, A.6.F is used only to unpack still-ambiguous function wording and establishes neither claim. If a predicate is missing, record the A.6.RCD missing-governor; if case facts are missing, record the corresponding unresolved result. This condition can be checked without asserting that any review appointment or dated review Work already exists.
Alice's classification under that kind, any review appointment she holds, any dated review Work she performs, and any report used as evidence remain four separate claims. This compact description names none of their occurrence identities.
Use:
- A.2 with C.3 for the local kind and direct classification;
- F.4 for the description episteme;
- A.2.1 when a particular review assignment must be identified;
- A.13 followed by independent A.15.1 admission when a particular dated review Work occurrence is identified, and F.6 afterward when the claim also identifies the assignment under which that Work was performed; and
- A.10, B.3, G.6, or another direct relation for the report's evidence or assurance use.
The report is not a system-role holder and does not acquire an “evidence role.”
Standard Used as a Specification or Source
The sentence “Standard S has the architecture-standard role in this Work” misclassifies the standard if it treats that episteme as a system-role holder. State the use instead: the exact edition of Standard S is used as a specification, external rule, premise, or source for named claims. A standard may constrain or support a claim through that direct relation. If a separate claim classifies the system applying the standard, establish that it satisfies the stated system-role kind. If an assignment is claimed, identify the assignment separately.
Access Role Is Not Automatically a System Role
RBAC role often names a permission grouping. If the current claim concerns permission or access standing, use the direct policy, deontic, access or status relation. Treat a local access term as a system-role kind only when A.2 and C.3 establish its candidate domain, work-facing membership distinction and identity. Classify a particular candidate when the receiving use needs that judgment. Permission and assignment remain separate.
Anti-Patterns and Repairs
Consequences
Benefits.
- Descriptions stay short enough for practice while preserving the ontology.
- Part F naming and Bridge patterns can cite descriptions without inheriting classification, assignment, capability, Method, Work, evidence, or status claims.
- Episteme-use relations stay direct and do not become a parallel system-role ontology.
- Method and Work checks may cite the description without treating it as Work evidence.
Costs.
- When adapting a template that mixes role and status, route each claim to F.10, A.2.4, B.3, A.10, E.17, G.6, or the direct relation that governs it.
- A stronger claim may require several neighboring patterns instead of one overloaded card.
- Public, Core-facing, or durable cross-local names require F.18.
SoTA Decision for a Readable Kind Description
The bounded question is: how can one work-facing kind be described for recognition without confusing the kind with its bearer, a classification judgment, assignment, capability, Method, Work, designation, or publication? The comparison assumes the effort of authoring one project pattern, not adopting a whole upper ontology.
Selected non-dominated contribution. gUFO and BFO offer richer foundational categorization, but at higher mapping effort and without deciding the project-local recognition use. ISO 704 and OntoLex keep description and designation separate at lower effort, but leave assignment, capability, Method, Work, and publication outside their answer.
Among the approaches considered at the effort of writing one project pattern, the smallest construction judged sufficient uses one C.2.1 episteme about one already recovered C.3 kind, a short recognition criterion in ordinary language, and references for stronger claims. It preserves the distinction between the description and the kind. The criterion tells a new project reader how to recognize the kind, and the references show where to check those claims.
Reopen F.4 when A.2, C.3, A.2.1, A.2.5, A.2.7, A.15, A.6.5, A.6.RSIR, C.2.1, F.9, F.10, F.18, or the accepted episteme-use discipline changes enough that the described-kind or non-inference boundary would be stated differently.
Relations
Builds on. A.2, C.3, C.3.2, A.6.5, A.6.RSIR, A.7, C.2.1, E.10.ROLE, E.10.D2, and E.24.
Coordinates with. A.2.1, A.2.2, A.2.5, A.2.7, A.15, A.15.1, A.15.2, F.5, F.9, F.10, F.14, F.15, F.18, and direct evidence, status, source, publication, requirement, permission, responsibility, and assurance relations.
Constrains.
- F.5 names a
SystemRoleKindDescriptiononly after the described local kind, current criterion, effective scheme, and local sense are recovered. - Use F.8 to decide durable name minting or reuse without turning status or episteme use into a system-role-kind description.
- F.14 keeps bundles and separation-of-duties relations separate from kind descriptions.
- Use F.15 to check the single-kind and non-inference boundaries.
Conformance Checklist
Phrasebook
Prefer:
- “description of the local kind currently named
ReviewerSystemRole; JournalReview-2026 locates the definition”; - “candidate-system admission is established under A.1; classification and any assignment are separate”;
- “capability requirement cited by the description”;
- “Method requirement cited by the description”;
- “review report used as evidence for this claim”;
- “standard used as a requirement source”; and
- “relation position declared by this
SlotSpec.”
Avoid as live Tech vocabulary:
- “evidence role” for an episteme;
- “status role” for a status or status-use relation;
- “standard role” for a standard used as a source;
- “holder” for a publication, report, standard, dataset, or theorem unless the exact entity is independently admitted as
U.Systemand an exactU.SystemRoleAssignmentnames it as holder; - “role” for a
SlotKind; and - “role description” for a Method, capability, Work record, access policy, or status-use relation.
Didactic Memory
A SystemRoleKindDescription is an episteme that explains one local system-role kind. Use its recognition account and current KindSignature to judge a candidate under C.3.2; establish any assignment separately under A.2.1. Follow :4.2 for capability, Method, Work or other neighboring claims.
F.4:End
Name U-kinds, Local System-Role Kinds and SystemRoleKindDescriptions
Type: Definitional (D) Status: Stable Normativity: Normative unless marked informative
Use This When
Plain name. Meaning-first naming discipline.
Use F.5 when a project needs a durable name for either:
- a public U-kind already admitted through E.24.UK, or another durable cross-local value already recovered through the direct rule for that kind of value—for example, episteme constitution or relation obtaining; a Concept-Set row may cite comparison evidence but does not recover the value; or
- one exact local system-role kind and, when needed, the separate
SystemRoleKindDescriptionepisteme that describes it.
Typical moments:
- a Concept-Set comparison has enough witnesses for a naming question and the reusable value is already admitted, but candidate names import one source tradition too strongly;
- an F.4 description names
ReviewerSystemRole, and the label must remain faithful to the exact local kind without smuggling assignment, capability, permission, Method, Work, evidence, status, or responsibility; - source wording with role must be named locally, but the project has not yet recovered its use—for example, a system-role kind, assignment, status or access relation, relation position, another object, or ordinary wording; or
- similar names threaten to collapse independently governed objects—for example, a kind, assignment, status, Method, Work occurrence, and description episteme.
Primary EntityOfConcern. The EntityOfConcern is the naming discipline for these name families. It relates a recovered meaning to a selected Tech designation and any Plain designation needed for a stated didactic use. Naming gives no basis for defining the U-kind or local system-role kind, constituting a description, classifying a candidate, creating an assignment, asserting status or responsibility, supplying evidence, or publishing a form.
Primary working reader. Use this discipline when you have a candidate meaning and need a name that lets readers recover the value’s kind and intended use.
First useful move. Recover the exact named value and its direct meaning source before choosing the label. For a U-kind, use its accepted E.24.UK admission result or its direct admission rule. For a local system-role kind, use its A.2 and C.3 identity and criterion; use F.4 for the separate description episteme. Then choose one Tech label that preserves the recovered meaning. Add a short Plain explanation when the intended reader needs it; its scope must not exceed that meaning.
Smallest useful result and stop. Stop with one already identified value and one Tech label as soon as the value and its meaning resolve unambiguously for the named local use. Add a Plain explanation when that use needs one. Do not create a NameCard, public row, Bridge, description episteme, or new kind merely to complete a form. If the value or kind is unresolved, apply its direct recovery rule. Use F.18 or F.17 only for the durable or public use they address. Use C.3.3 only when there is a relation between specific local kinds, and F.9 only when there is a relation between distinct F.17 cells. If the label starts carrying assignment, Work, result, provenance, assurance, responsibility, or publication claims, stop naming and recover those objects first.
What goes wrong if missed. Names become arguments. A system-role-kind label smuggles in neighboring claims—for example, assignment, permission, responsibility, or capability. A status phrase becomes a system-role kind. A U-kind name imports one practice's or source's private ontology. A shared word can hide disagreement among witnesses. Downstream patterns then repair semantics that naming already broke.
What this buys. Readers can use short names without guessing the ontology. U-kind names stay neutral across witnesses. Concrete ...SystemRole designations point to exact local kinds, and ...SystemRoleKindDescription designations point to their separate description epistemes. Names for neighboring claims—for example, status, evidence, access, requirement, source, publication, assurance, gate, and decision claims—remain with their direct relations.
Not this pattern when.
- If the problem is ordinary phrase repair, use E.10, E.10.ROLE, E.10.ARCH, A.6.P, A.6.RSIR, or the direct pattern.
- If the question is whether a
U.*spelling or structural name should survive as a durable U-kind, use E.24.UK before F.5. - If the broader local-first protocol, NameCards, candidate comparisons, lineage, or public naming is current, use F.18.
- If the current object is a
SystemRoleKindDescription, use F.4 to constitute it before naming it. - If the question concerns kind admission, classification, assignment, assignment extent, or performed-Work attribution, use A.2 with C.3, A.2.1, or F.6.
- If the current object is another governed value rather than a name—for example, a status, evidence use, source use, standard use, requirement use, publication use, assurance claim, gate result, or decision—use its direct pattern.
- If role denotes a relation position, recover the position under A.6.RSIR and A.6.5.
- If the naming use involves a cross-local relation, use C.3.3 for exact local kinds or F.9 for distinct F.17 cells.
Problem Frame
FPF needs names that humans can use without dragging the wrong ontology behind them. A good name is short enough for documents and conversation, but it belongs to a recovered meaning.
This pattern keeps two recurrent naming tasks separate.
First, a public U-kind gets a name only after E.24.UK admits the exact value. Another durable cross-local value gets a name only after its direct rule has identified or established it; kind membership is only one case. A Concept-Set row may preserve witness comparison and evidence; it neither admits nor identifies the value. The name should be neutral across witnesses and no wider than the recovered invariants.
Second, one concrete local system-role kind receives a ...SystemRole designation after A.2 and C.3 settle its candidate domain, operative membership condition, intended member/non-member boundary, and continuity rule. A practice or source reference may help readers find or compare that settlement; it does not identify the kind. SystemRole is common morphology, not a universal kind. An F.4 description episteme is another object and may receive a separate ...SystemRoleKindDescription name. Neither label creates the kind, description, classification, or assignment.
Using system-role descriptions for statuses and episteme uses can seem convenient because all need labels. Treating these uses as system-role kinds creates duplicate ontology. Another governed value—for example, a status, evidence use, permission, or publication—may need a name; none becomes a system-role kind because it is named.
Problem
Without a way to check names against their meanings, these errors can arise.
- Local terms look global.
Observation,Activity, orProcessbecomes a U-kind name although it carries one practice's or source's private commitments. - System-role names become hidden admissions. A label such as
ReviewerSystemRoleis treated as if the local kind or candidate classification already exists. - System-role names become hidden assignments. A concrete kind label is treated as if someone is already assigned.
- System-role names become capability claims. A candidate is assumed able because the kind label sounds competent.
- System-role names become Methods. A noun label hides a Method or Method family.
- Description and described kind collapse.
PumpInspectorSystemRoleKindDescriptionis treated asPumpInspectorSystemRoleitself. - Status names become system-role kinds. For example,
Approved,AccessRole,ModelFitEvidenceRole, orRequirementRolecreates a fake work-facing classification instead of the exact direct relation. - Relation positions become system-role kinds. Signature, relation, or argument-position names borrow role morphology even though they name participation or a declaration place.
- Names carry interpretation metadata.
Task-IEC61131,Participant-BPMN, orReviewerSystemRole-SchemeAfossilizes an edition, source, local boundary, or scheme in the label. - Aliases become silent renames. Several labels circulate for one meaning without lineage or Bridge discipline.
Forces
Solution
Name after meaning. Recover the value, its kind, direct meaning source, and intended use. Then choose designations that preserve them.
Make these facts recoverable in the prose, direct admission, F.4 description, Concept-Set row, or NameCard. This is a naming checklist, not a relation signature or mandatory record:
- the exact named value and its admitted kind;
- the direct source of its meaning;
- for a local system-role-kind designation, the candidate domain, operative membership condition, intended member/non-member boundary, continuity rule, current
KindSignature, and effective scheme, with source or practice provenance kept as a locator or comparison cue; - for a description name, the separate F.4
SystemRoleKindDescriptionand its exact EntityOfConcern; - the selected Tech designation and any Plain designation needed for a stated didactic use;
- aliases or predecessor labels with lineage;
- morphology, neutrality, and minimal-generality checks; and
- the boundary that prevents the name from absorbing classification, assignment, capability, Method, Work, status, evidence, permission, responsibility, publication, or relation-position claims.
Name Families Used Here
Keep four things separate: the chosen name, the local system-role kind it names, an optional F.4 description of that kind, and any assignment needed for this use. The name designates the kind; the description describes it.
An assignment is a separate A.2.1 occurrence of a directly declared species under U.SystemRoleAssignment. That species says which systems may be holders, which exact local kinds may fill the assigned-kind place, what the assignment predicate means, when it applies, how an uninterrupted occurrence keeps its identity, and whether the species requires another participant. The occurrence supplies the actual holder, assigned kind, and any other participant values.
When the use only needs a name for the kind, stop there. A spelling or suffix can designate the kind. A NameCard, row, description or citation can also refer to an independently established assignment, dated Work, result episteme, provenance record or publication occurrence. Each such claim needs grounds under its own rule.
Tech and Plain Designations
Choose a Tech designation for durable reuse. Add a Plain designation when a stated didactic use needs it:
For a concrete local system-role kind, the Tech designation normally ends in ...SystemRole, for example ReviewerSystemRole or PumpInspectorSystemRole. The Plain designation may remain ordinary, for example “reviewer” or “pump inspector”, when the named practice and criterion make the intended kind clear. Add “system role” only when it distinguishes the intended meaning from a plausible alternative. The compound does not imply non-human technical systems, kind admission, candidate classification, assignment, agency, capability, Method, or Work.
For the description episteme, name the description rather than the described kind: PumpInspectorSystemRoleKindDescription may have Plain designation “description of the pump-inspector system-role kind”. SystemRoleKindDescription identifies the construction; Kind identifies the EntityOfConcern and Description already identifies the episteme.
For a coupled system-role–Method phrase, recover the local kind and Method separately before naming either one. Recover and name a MethodDescription, WorkPlan, or dated Work only when that object has been admitted and the naming use needs it; a shared phrase does not require any of them to exist.
RoboticsEngineerSystemRole may designate one admitted local kind; RobotEngineeringMethod names a Method or Method family. Ordinary engineer-roboticist may remain the Plain expression when nearby project wording makes the intended kind clear and the C.3 candidate domain, membership distinction, boundary probes, and continuity remain recoverable. The wording helps the reader; it does not identify the kind. It replaces neither a qualifying MethodDescription nor any description of planned or performed Work.
If a later naming use needs the identity of a dated Work occurrence, the Work must be constituted before naming begins. Recover every exact actual performer through A.13, and let A.15.1 independently admit the Work from its semantic Method, time, containing System, and other required direct facts.
Add the assignment occurrence, holder equality, and F.6 relation only when the naming record or receiving use expressly represents precise assignment-bound attribution; missing or failed F.6 alone leaves independently admitted Work intact. If the same finding invalidates a premise consumed by its A.13/A.15.1 admission, recheck that premise and its dependent conclusions under F.6:4.2. When the naming use does not consume a dated Work identity, keep the activity in ordinary wording.
For a U-kind, the Tech designation should be neutral enough that no witness wins by vocabulary alone. If witnesses disagree between Observation, Reading, and MeasurementResult, a Concept-Set row preserves the comparison; the exact shared value and invariants must still pass E.24.UK admission or their direct defining rule before an author uses F.5 to choose a name.
Positive Naming Rules
- Recover the object first. State the governed kind or construction of the value—for example, a U-kind, local system-role kind, description episteme, classification judgment, assignment, relation, Method, Work, status, evidence use, slot, lens, or another object.
- Recover the meaning source. Use the exact E.24.UK or direct admission for a U-kind; A.2 with C.3 for a local system-role kind; F.4 for its description; A.2.7 for relations among kinds; A.3, A.15, G.5, or the exact composition pattern for Method and Work names; and the direct relation for status, evidence, source, requirement, publication, assurance, gate, decision, and relation-position names.
- Use minimal generality. The designation's scope is no wider than the admitted invariants.
- Keep interpretation metadata out of the label. Edition, source, witness, local boundary, reference scheme, and threshold belong in the direct declaration, description, relation, or NameCard.
- Make morphology object-sensitive. Concrete local system-role kinds use
...SystemRole; description epistemes use...SystemRoleKindDescription; states use state or level wording; slots saySlot,Argument,Endpoint, or another exact position head. - Keep coupled names typed. A compact phrase may help a reader, but one label must not carry several independently governed objects—for example, kind, assignment, capability, Method, Work, and description.
- Do not encode thresholds or windows in the name. Put time, state, threshold, capability envelope, or admission window in the direct claim.
- Use aliases only with lineage. A source term, predecessor term, symbol, or translation does not become a second selected Tech label.
- Escalate only when reuse calls for it. Use F.18 and F.17 for durable or public naming. When the use depends on a cross-local relation, name the exact obtaining C.3.3 relation between local kinds or F.9 Bridge between distinct F.17 cells and keep the separate C.2.1 claim that it suits the named use. For ordinary reliance, recover the descriptive A.10 evidence-provenance path and the independently established direct relations it cites. Judge evidence sufficiency under the direct rule for the relied-on claim.
RelianceDisposition=passsupports only the named bounded use;degradesupports only its named narrower use. Use B.3 only when the use involves a named assurance claim. When naming or reusing a value, check any claim about assignment, Work, a result, provenance, assurance or publication under the rule for that claim. The naming record may cite the grounds.
Neighboring Use Boundary
When a candidate contains a tempting word, recover the current claim instead of replacing words mechanically.
Select the name only after recovery. A cleaner string is not a repair if it hides the same ontological error.
Archetypal Grounding
Public or Cross-Local Kind Name
A Concept-Set row compares SOSA Observation, metrology measurement result, ML practice metric reading, and a dashboard value exported for comparison. The row is a comparison and evidence surface, not admission or identity of a common result value.
Keep the objects with the rules that establish them. Pump 14 was measured before the reading was recorded, but this naming example does not identify a dated Work occurrence. If a use needs that occurrence, recover its exact actual performer through A.13 and admit it independently under A.15.1. Attribute it under F.6 only when that use also consumes precise assignment-bound attribution.
C.16 constitutes the measurement result: a value attributed to the measurand together with the Characteristic, Scale, uncertainty, method, model, calibration basis, time stance, and measurement Work needed to interpret it. Pump14PressureReading_2026-07-14T10-42Z is one C.2.1 episteme that states that result; F.5 does not repeat either pattern's schema.
The result and its episteme are distinct from raw output, indication, Pump 14's actual state, a later diagnosis, a criterion verdict, evidence, or a dashboard display. Pump14CalibrationTrace_2026-07-14 is a provenance record. G.6 and A.10 make its calibration and source path recoverable by representing independently established direct relations. A dashboard publication may cite the reading, and the Concept-Set row may cite the reading and trace; neither is the result, its episteme, provenance, or a generic relation that establishes them.
Only E.24.UK or the direct result pattern can admit a shared value and its invariants. After admission, use F.5 to select Reading, Result, or another neutral head no wider than that value. The spelling still creates no result or provenance identity.
Local System-Role Kind and Its Description
Under Plant-A-Maintenance-Scheme, PumpInspectorSystemRole designates one exact local kind; it is not that kind. PumpInspectorSystemRoleKindDescription-v3 is a separate C.2.1 episteme whose EntityOfConcern is the kind. Its ClaimGraph states which systems are candidates, the reading-and-judgment condition that distinguishes members, useful member and non-member probes, the continuity rule, current KindSignature, and effective scheme. Plant-A maintenance provenance locates that definition; it does not identify the kind. The Tech designation is PumpInspectorSystemRole; the Plain designation is “pump inspector”.
The foreman also asks which appointment the July work list means by “Robot-7 as pump inspector”. The two names above already answer the naming question. For the foreman's question, Robot7-PumpInspector-Assignment-2026Q3 is one occurrence of the directly declared PlantAPumpInspectionAssignment species under U.SystemRoleAssignment. The species' holder slot admits a U.System; its declaration-local assigned-kind slot uses the exact PlantAMaintenanceSystemRoleKindDomain; and its predicate applies within the Plant A maintenance scheme and obtains while the fixed holder is assigned under PumpInspectorSystemRole to supply the pump-inspection contribution.
The occurrence identifies Robot-7 as holder and PumpInspectorSystemRole as assigned kind, and spans the maximal uninterrupted interval over which that predicate obtains for those values. This species declares no additional identity-bearing participant; a commission, position, or installation locus would become one only in a species whose predicate and identity require it.
This naming example does not identify Robot-7's inspection of Pump 14 as a dated Work occurrence. Pump14InspectionFinding_2026-07-14T11-18Z is a separate claim-bearing result episteme, and Pump14InspectionTrace_2026-07-14 is the provenance record; G.6 and A.10 represent its independently established source and provenance relations in a descriptive path.
The kind label helps readers recover the kind; the description episteme describes it. Neither says Robot-7 satisfies the kind, has an assignment, performed the inspection, produced the finding, or supplied its provenance. A NameCard, row, pattern section or citation can identify a separately established object or relation. Its mention, or the suffix on its name, does not establish that the relation holds or that the Work occurred.
Evidence Use Is Not a System-Role Name
Source text may say ModelFitEvidenceRole. This naming example does not identify the model-fit evaluation as a dated Work occurrence. It uses three objects: ModelFitResult_2026-07-15T09-22Z is a separately constituted domain-local result episteme; ModelFitTargetClaim-v5 is the target claim; and ModelFitRunTrace_2026-07-15 is the provenance record. G.6 and A.10 represent its independently established source and provenance relations in a descriptive path.
Keep any operation-result binding, result-episteme inception claim, evidence use, provenance, and current assurance claim separate, and apply the rule that defines or tests each relation.
A durable name, if needed, names one recovered evidence-use relation, status value, Work occurrence, result episteme, or provenance value. ModelFitEvidenceRole, a NameCard, row, or citation creates none of them and supplies no generic evidence-result relation. It is neither a local system-role kind nor a SystemRoleKindDescription label.
Relation Position Is Not a System-Role Name
In a relation signature, “provider role” may mean the provider argument position. Use E.10.ROLE and A.6.RSIR to recover the participant meaning; use A.6.5 to declare ProviderSlot, its ValueKind, and its reference mode. A provider system's classification under a local ProviderSystemRole kind is a separate C.3 claim. When assignment identity is irrelevant to naming that relation position, say only that any provider assignment remains independently governed by A.2.1; do not invent an occurrence. When it is relevant, recover the assignment occurrence and its declared species rather than asserting that the provider simply “has an assignment”.
Bias Annotation
- Semio-bias. A name, card, row, publication, or source label is mistaken for the named value or authority to use it.
- Role-bias. Evidence, status, access, source, requirement, participation, or argument-position wording is forced into
SystemRolemorphology. - Source-vocabulary capture. One source's term becomes the Tech designation without showing fit to the admitted value or exact local kind.
- Suffix formalism. Adding
SystemRole,KindDescription,Status,Record,Graph, orMapmakes a label look precise while the object remains unresolved.
The repair is object recovery first, designation second.
Conformance Checklist
Common Anti-Patterns and Repairs
Consequences
Good consequences:
- durable names become shorter because the ontology stays at the right object;
- local system-role-kind names stay usable without becoming assignment, capability, Method, or evidence claims;
- description names no longer collapse into the kinds they describe;
- U-kind names are easier to bridge because their comparison evidence remains explicit.
For an E.10 repair that uncovers a durable naming issue, use F.5 or F.18 instead of ad hoc word substitution.
Costs:
- authors recover the object and meaning source before naming;
- some familiar source labels cannot become FPF Tech designations;
- durable public names may need F.18 and F.17, while actual cross-local relations may need C.3.3 or F.9 even when a local label looks obvious; and
- source text that uses role for status, evidence, access, participation, or relation position needs ontological recovery, not suffix editing.
Reopen F.5 when U-kind neutrality, SystemRole or SystemRoleKindDescription morphology, the Tech-Plain relation, lineage, or durable cross-local naming boundaries change. Reopen a neighboring pattern when the dispute is about the named object itself.
Rationale
Recover the object before naming it. Durable names become references used in reasoning, search, publications, and pattern relations. A wrong name makes later readers inherit a false kind claim.
Names follow the source of the meaning. The spelling alone cannot settle which object to name. Bare role can point to many different objects or uses—for example, a local system-role kind, assignment, policy term, status, evidence use, relation position, representation position, or ordinary English. Do not decide by suffix. Use E.10.ROLE and the direct patterns to recover the object, then F.5 to name it.
F.5 remains narrower than F.18. Use F.18 for the full local-first protocol, NameCards, candidate comparison, lineage, and public naming. F.5 supplies the special discipline needed by U-kind names, concrete system-role-kind names, and SystemRoleKindDescription labels.
SoTA Decision for Precise, Readable Technical Names
The bounded question is: after the object is recovered, what is the smallest naming result that stays technically precise, readable to a project reader, and explicit about morphology and reuse?
Selected contribution. Keep the recovered meaning accessible through the designation and its direct explanation. A bare label is insufficient when the intended reader cannot recover that meaning; a full ontology lexicon may supply detail that one naming decision does not need. F.5 uses one already recovered value and a Tech designation, adding a short Plain explanation for a stated didactic need. Word form follows the kind of object, and the name's use remains bounded by the recovered meaning.
Source-use boundary: external labels, Concept-Set rows, and citations are evidence for local meaning or common practice, not automatic Tech designations, admission decisions, or Work, result, and provenance identities. A source term becomes selected only after the exact value is admitted and the naming comparison passes; naming changes none of those objects.
Relations
Builds on. A.2, C.3, F.4, F.7, F.18, E.10, E.10.ROLE, and E.10.ARCH.
Coordinates with. E.24.UK for U-kind admission; A.2.1 for system-role assignment; A.2.2 for capability; A.2.5 for assignment state; A.2.7 for relations among system-role kinds; A.6.5 and A.6.RSIR for relation positions; A.15 for system-role–Method–Work alignment and dated Work; C.16 for measurement results; C.2.1 for descriptions and result epistemes; G.6 and A.10 for provenance and ordinary evidence reliance; B.3 for assurance-bearing reliance; F.8 for mint or reuse; C.3.3 for relations between exact local kinds and F.9 for relations between distinct F.17 cells; F.10 for status; F.13 for lineage; F.14 for anti-explosion; F.15 for conformance; and F.17 for public term-sheet use.
Used by. Part F naming patterns, F.4 description authors, Concept-Set authors, E.10 repairs that uncover naming rather than phrase-use issues, and any pattern use that creates a durable local name for a U-kind, system-role kind, or SystemRoleKindDescription.
Does not replace. Direct evidence, status, requirement, source, publication, assurance, gate, decision, responsibility, relation-signature, Method, Work, or architecture patterns.
F.5:End
SystemRoleAssignment and Performed-Work Attribution Check
Type: Boundary and use pattern Status: Stable Normativity: Normative unless marked informative
Use This When
Plain name. Check whether this already admitted Work was performed under this exact system-role assignment.
Use F.6 only after A.15.1 has independently admitted a dated U.Work occurrence, and when you need to decide whether it was performed under a particular assignment occurrence from the U.SystemRoleAssignment family. When it was, the direct world-side performed-under-assignment relation obtains. A separate assertion or record can identify the two occurrences and state that relation.
Typical moments:
- a work record says “Alice reviewed”, “Robot-7 inspected”, or “the operations team approved”, but the assignment occurrence is missing;
- a MethodDescription names a system-role kind and the project must connect actual Work to the assigned performer;
- source wording says
RoleEnactment, “played the role”, orHolder#Role:Context@Window; - a stronger appointment has a commission, position, or locus participant and must retain that occurrence identity during attribution;
- an attribution claim names a report instead of the Work's already admitted performer System;
- a corresponding kind or assignment from another context is used to stand in for the actual assignment occurrence.
Primary EntityOfConcern. One obtaining performedUnderAssignment relation occurrence between a U.Work occurrence and an assignment occurrence whose species is declared under U.SystemRoleAssignment.
Primary working reader. An engineer, operator, Method author, manager or FPF author can use this check to decide whether an attribution is sufficiently grounded for its next use.
First useful move. Confirm that A.15.1 has already admitted the exact dated Work without using an F.6 conclusion. Name that Work and the assignment occurrence under which it is said to have been performed. Recover the assignment’s declared species and participant values. Confirm that the actual performer has the A.13 core for this action and any scope, working situation and window needed by the attribution use. Check that this is the same assignment occurrence that obtains under A.13. Evidence supports those core facts; a characteristic profile enters only for a consumed Grade, autonomy or profile result, criterion-dependent characteristic, or assurance use.
Ask what direct case fact links the exact pair. Confirm holder equality and interval coverage; those checks alone do not create the link. If the facts leave the pair undecided, leave it unresolved. If they rule this pair out, say so. Retain independently admitted Work while its admission grounds hold; apply section 4.2's premise-recheck rule when the same finding also defeats one of those grounds. When the case establishes the pair and the required grounds hold, say that the holder System performed the Work under that assignment.
What goes wrong if missed. Assignment is treated as proof of Work, a label replaces the assignment occurrence, a generic assignment duplicate erases a stronger appointment, or a log or report is made the performer. When several assignments overlap, interval coverage then attributes the same Work to all of them even though the exact pair was never established.
What this buys. Attribution is one relation between the Work and assignment occurrences. The holder System remains the actor, the assignment occurrence remains linked to its species and participant values, and Work, Method, capability, state, result, evidence, publication, and cross-context use remain separate.
Not this pattern when. Use A.2 for the system-role kind and classification, A.2.1 for assignment species and occurrence identity, A.2.5 for assignment state, A.2.2 for capability, and A.15.1 for the Work occurrence. Use the direct evidence, source-reliance, publication, access, authority, permission, responsibility, status, gate, or decision pattern when that relation is current. Use E.10.ROLE and A.6.RSIR when role denotes another object or relation position.
Problem Frame
U.SystemRoleAssignment and U.Work classify different world-side occurrences. An assignment occurrence belongs to a declared species, relates fixed participant values, and lasts for one maximal uninterrupted interval in which its predicate remains true. A U.Work occurrence is dated. Their existence does not establish the additional attribution relation.
Use F.6 to state that missing relation between the Work and assignment occurrences. Every assignment species declares a holder slot, and each assignment occurrence supplies its actual holder System. F.6 exposes that holder only so the attribution can compare it with the actual performer already recovered through A.13 and used by A.15.1 to admit the Work; it does not discover a performer. This preserves a commission-sensitive or otherwise stronger assignment instead of replacing it with a generic duplicate.
A roster can assert the assignment; a log can assert the Work and attribution; evidence can support either assertion. Those epistemes help a system know or use the claim. They do not become relation participants or make world-side attribution obtain merely by being stored.
This separation matters because assignment, classification, state, ability, performance, result, evidence, and acceptance vary independently. A system can hold an assignment and do no Work. It can perform poor Work under a valid assignment. A report can accurately describe the Work without performing it.
Problem
Without the direct attribution relation:
- Assignment becomes Work. Current assignment is treated as evidence that a system performed one occurrence.
- Performer comes from a label.
ReviewerorOperatoris used without a holder and assignment episode. - A stronger assignment is flattened. A commission-sensitive appointment is replaced by a weaker generic record.
- Episodes do not cover. Work is attributed outside the interval in which the exact assignment predicate obtains.
- Support becomes constitution. A log, report, standard, dashboard, or decision is treated as what makes attribution obtain.
- Enactment is duplicated.
RoleEnactmentorRoleEnactmentFactbecomes another object beside Work and attribution. - Locality is hidden. A context word replaces the exact local kind, assignment species, Work locus, scope, or selected model-use structure.
Forces
Solution
Treat performed-Work attribution as one direct relation species under U.Relation.
Direct Relation Declaration
WorkOccurrenceSlot names a dated Work already admitted under A.15.1 from independently grounded performance history, A.13-qualified actual performer facts, Method, extent, and containment. The typed slot consumes that completed membership result; F.6 does not establish W : U.Work; a finding that also invalidates a consumed admission premise invokes section 4.2's premise recheck.
The declaration-local SystemRoleAssignmentSlot names one occurrence of an admitted assignment species declared under U.SystemRoleAssignment. Its U.RelationRef names that occurrence and is limited to U.SystemRoleAssignment.
Filling the two slots, matching the holder, or finding temporal overlap does not establish that this Work was performed under this assignment; the case must independently establish that link.
For an obtaining attribution:
S is the admitted System already recovered as an actual performer through A.13 and used by A.15.1 to admit W; F.6 does not discover it. RA is the assignment under which that Work is attributed. The projection exposes the holder already carried by RA only to test equality with S; it creates neither performerhood, Work, attribution, classification, nor a generic assignment occurrence and discards none of RA's additional participants.
performedBy remains only a deprecated source relation name. Read it through the direct Work-assignment relation only after A.13 and A.15.1 have independently established the actual performer and admitted Work, and after holder equality is checked. New practitioner-facing claims say that the already recovered performer System performed the Work under the assignment, or name performedUnderAssignment when the relation name is needed; they never make the assignment the performer or use F.6 to discover one.
No evidence, log, status, MethodDescription, result, publication, context record, or assignment-state assertion is a generic attribution participant.
Obtaining and Occurrence Identity
The direct Work-assignment attribution is a world-side fact, separate from any assertion or evidence. A positive check requires all of the following:
Wis one exact datedU.Workoccurrence already admitted under A.15.1 from its independently grounded candidate-action history, A.13-qualified actual performer basis, Method actually followed, temporal extent, and containing-System relation; that admission neither assumes nor depends on this F.6 relation;- the actual performer
Shas the A.13 core for this action and any scope, working situation and window needed by the attribution use:Sis an admitted System, satisfies and is classified under one exact local agential system-role kind, and holds the same obtaining assignmentRA; evidence supports those core facts, while a characteristic profile is required only for a consumed Grade, autonomy or profile result, criterion-dependent characteristic, or assurance use; RAis one named assignment occurrence of a declaredU.SystemRoleAssignmentspecies, with all identity-bearing participants and its rule recovered;- the case establishes that
Wwas performed underRA, rather than deriving that link from a label, common holder, assignment existence, or temporal overlap; RA.HolderSystemSlot = S, the admitted System that actually performedW; and- RA's species predicate obtains throughout the attributed temporal extent of
W.
Conditions 1–3, 5, and 6 are five constraints on a valid attribution but do not establish it. F.6 reuses the obtaining A.13 assignment; it does not create the A.13 classification, assignment, evidence, optional profile, or Work. Where the independent A.13/A.15.1 admission grounds still hold, missing or failed attribution leaves W : U.Work intact and leaves only this exact assignment-bound attribution unasserted. If the same finding invalidates a premise used in that admission, recheck the premise and its dependent conclusions under A.15.1.
Preserve independently grounded history and relations; absence of attribution alone is not evidence of no performance.
Two overlapping assignments held by the same System may satisfy all five constraints while the Work was performed under only one. To recover the link, start from the contribution required by each assignment and follow the actual action's initiation and execution, including the return of a result when it matters to that contribution. A request or commission can identify which contribution was undertaken; execution facts and any result-return facts can show whether this Work carried it out. Use the identity-bearing participants and rule of the actual assignment species to distinguish competing occurrences.
These facts may be recoverable from an existing execution trace, correspondence, direct observation or another adequate source; no particular record format is required.
Reuse an already established pairing. When it is not established, inspect the point at which the histories would differ: for example, which commission's question the performer undertook and to which commission the resulting review answered. Shared task wording, holder and time alone do not decide that question. If the available facts still cannot distinguish the assignments, leave the attribution unresolved rather than asserting both.
Select the Work occurrence from the contribution the recipient wants attributed. If it is one constituent of a larger job, identify that constituent under A.15.1 and check its assignment. An adequate commission and execution record can link the whole job directly to its assignment. Where the answer depends on contributions to that job, establish which performances composed it and how they supplied its result (B.1.5.EW). A confirmed assignment for a constituent can answer a question about that constituent. Its bearing on the whole depends on that Work-part relation and the contribution the assignment covered.
When a receiver needs an explicit attribution occurrence:
This key identifies an already obtaining relation; it does not make one obtain. The attribution extent follows W. Extending an open Work interval or later recording its end does not create another attribution occurrence while both participants and the direct relation remain the same. Another Work occurrence, separately identified Work part, or assignment episode yields another possible pair whose relation must be checked independently.
An assertion can state the exact pair, and evidence can support reliance on that assertion. Neither the assertion nor its evidence constitutes the world-side relation. Missing evidence leaves reliance unresolved; missing pair grounding leaves the positive attribution unasserted. A demonstrated different performer, non-covering assignment, or false direct pair can support a stronger negative claim.
Preserve the Exact Assignment Species
Before checking or relying on attribution, recover RA's declared species and occurrence. This distinguishes the assignment even when the final practitioner sentence omits its full declaration. Every species declares:
- a
HolderSystemSlotwhoseValueKindisU.System; - a declaration-local
AssignedSystemRoleKindSlotwhoseValueKindis the exact local system-role-kind domain admitted for that species; - every additional participant meaning and its
ValueKind; - the rule, applicability, and maximal uninterrupted occurrence identity.
An assignment occurrence supplies one participant value for each slot. In particular, it supplies one local system-role-kind value from the AssignedSystemRoleKindSlot domain; the value does not replace or narrow that declared domain.
A simple assignment may have only holder and kind. A project appointment may also have ReviewCommissionSlot. F.6 accepts both through the family ValueKind and holder projection while retaining the declared species and all participants that distinguish the assignment occurrence. Those participants and the assignment rule still do not establish that the Work was performed under the assignment; the case must establish that link separately. F.6 never creates a two-participant generic assignment beside the appointment.
Taxonomy, scheme, KindSignature, assertion, and assignmentInterval can interpret or describe RA without becoming participants by default. Verify temporal coverage from whether the assignment rule actually holds, not merely from a recorded interval.
Do not replace the species with one Context value. Recover what the source token denotes and use its direct pattern. It can denote a system or Work locus, claim scope, or selected BoundedModelUseStructure; those objects are neither interchangeable nor optional participants of generic assignment or attribution signatures.
Attribution Check Sequence
- Start from the exact
U.Workoccurrence already admitted by A.15.1 without an F.6 premise. - Recover the assignment occurrence, including its declared species, identity-bearing participants, rule, applicability, and time span.
- Recover the link from the assigned contribution and actual initiation, execution or return of result as in section 4.2. Identify the fact that distinguishes this Work–assignment pair from any live competing pair; matching holder and interval alone is insufficient.
- Confirm that the assignment holder is the actual performer.
- Confirm that the assignment predicate obtains throughout the attributed Work interval.
- When all five checks pass, state the F.6 relation or say that the holder System performed the Work under that assignment. If the facts show that a required condition fails, rule out this pair. If the facts needed to decide the link or a condition are missing, leave the pair unresolved. Retain Work whose independent admission grounds still hold, and apply section 4.2's premise-recheck rule when the finding also affects those grounds; do not select another covering assignment.
- Keep assertions and evidence separate: they can support reliance on the attribution claim but do not make the relation obtain.
- Send classification, assignment state, capability, Method, evidence, source use, result, acceptance, publication, bridge, responsibility, and authority questions to their subject patterns.
This sequence is application guidance, not a new check record or workflow object. The first useful result is a finding that the Work was or was not performed under this assignment, an unresolved pair with the missing fact named, or a reference to the rule for the neighboring claim.
Method and Work Boundary
performedUnderAssignment has no Method participant. A separate claim may say that the Work enacts one exact semantic Method. The holder System performs the Work; the Work, not the performer or assignment, enacts the Method.
The assignment, system-role kind, capability, Method, and MethodDescription do not act or perform Work. Citing a description can identify, constrain, or support a receiving use of the Method, but it neither enacts the description nor establishes D : U.MethodDescription; use A.3.2 to test that membership separately.
Direct Neighboring Relations
Source Shorthand and RoleEnactment
Holder#Role:Context@Window is readable source notation only. Identify the system denoted by the shorthand, the local system-role kind, the assignment species and occurrence, and the object denoted by Context. The source spelling is not a signature.
When a source uses RoleEnactment or RoleEnactmentFact, recover what it claims. Its meaning in that source can answer a terminology question. For a claim that dated Work was performed under an assignment, identify the Work and test performedUnderAssignment. This direct relation suffices without a second enactment kind.
Lightweight Use
After the Work–assignment link and its necessary constraints are established, ordinary use can stop at:
Expose declarations and occurrence keys only when a dependent use must distinguish occurrences, cite one as a participant, compare assertions, or preserve provenance. If the assignment cannot be recovered for this attribution, retain the independently grounded Work and actual performer, leave the attribution unresolved, and apply section 4.2 to any affected admission premise. State any source-reliance or evidence limit under its direct predicate.
Another pattern may require a complete A.13/A.15.1/F.6 basis when its receiving use needs both admitted Work and precise assignment-bound performer attribution, and may point here instead of repeating this declaration and check sequence.
The combined basis has a fixed order: A.13 first supplies every performer's exact System, local agential kind and criterion, classification, obtaining assignment, needed scope, working situation, window, and adequate core evidence; A.15.1 independently admits the dated Work from its performance history, at least one obtaining enactsMethod relation, extent, and at least one obtaining locally declared Work-to-System containment relation; only then does F.6 test every required exact Work-assignment pair through the same obtaining A.13 assignment.
The combined basis is not an A.15.1 membership test. A missing F.6 relation alone preserves independently admitted W : U.Work and leaves the precise attribution unresolved; section 4.2 governs a finding that also invalidates an admission premise. A characteristic profile remains conditional, and another enactment or containing-system relation is named only when the receiving use relies on it.
Invariants
- Every positive performed-Work attribution links one dated
U.Workoccurrence to one assignment occurrence of a declaredU.SystemRoleAssignmentspecies. SystemRoleAssignmentSlotaccepts the family and preserves the assignment's declared species, all participants, rule, applicability, and occurrence identity.- The actual performer is the admitted System in
RA.HolderSystemSlot; the assignment and kind do not act. - RA's predicate obtains throughout the attributed Work interval; a declared window alone does not establish coverage.
- The species declaration, occurrence participant identity, holder match, and time coverage constrain but do not establish the Work–assignment link.
- Overlapping assignments are checked pair by pair; an unresolved basis never licenses attribution to every covering assignment.
- Every positive precise assignment-bound performer attribution starts from an already admitted Work whose actual performer has the A.13 core for the exact action and any scope, working situation and window needed by the attribution use, then adds its own F.6 link through the same covering assignment occurrence. A characteristic profile remains conditional on its receiving use. A lead, team, member, coordination, allocation, or responsibility claim substitutes for none of these.
- A passive assigned System is not thereby a performer. Any claimed passive participation needs a rule that defines it; otherwise A.6.RCD returns
missing-governor. - Assignment does not prove performance, and attribution proves neither classification, capability, state, Method validity, result quality, responsibility, authority, nor acceptance.
RoleEnactmentwording is repaired to Work plusperformedUnderAssignment; no duplicate object remains.- Assertions, logs, rosters, evidence, identifiers, and publications can support or designate an attribution but do not constitute it.
- Missing evidence leaves reliance unresolved; a missing case fact linking Work and assignment leaves the positive attribution unasserted.
- An episteme does not fill
HolderSystemSlotbecause it describes or supports Work. - Cross-context correspondence changes neither assignment identity nor Work attribution.
- Reduced prose may omit only an assignment identifier unused by the receiving claim, and only after the complete Work–assignment basis remains recoverable.
- For each obtaining Work–assignment pair, the holder System is the performer identified by that attribution. Which Method the Work enacts is a separate fact.
Reasoning Rules
- Accept the attribution only when the case establishes that
Wwas performed underRA,Wis a dated Work occurrence,RAis a named assignment occurrence with its participants and rule recovered, the assignment holder performed the Work, and the assignment rule covers the Work interval. A short account may then say that the holder System performedWunderRA. - If
WandRAexist, the holder performed the Work, and the assignment covers the interval, do not infer the Work–assignment link. Check the case fact that establishes the link or leave it unresolved. - If current support for an attribution statement is inadequate, reliance on that statement is unresolved. Do not infer that the Work was not performed under the assignment, and do not treat the statement or evidence as what makes the link true.
- If a source episteme merely names a performer or role, do not claim attribution until the Work and assignment are identified, the case establishes their link, the holder matches the performer, and coverage is checked.
Archetypal Grounding
Robot Inspection
MaintenanceInspectionAssignment is a declared species under U.SystemRoleAssignment. Its participants include a HolderSystemSlot for the assigned System and a local AssignedSystemRoleKindSlot whose value is an InspectorSystemRole. Its rule applies within the Plant A maintenance scheme and says that the fixed holder is assigned under that kind to supply the inspection contribution; one occurrence is the maximal uninterrupted interval for which that rule stays true for the same participants.
The case basis directly links that Work to that assignment; the matching holder and interval only confirm necessary conditions. Robot-7 is the actor.
Separately, the inspection Work enacts TurbineInspection@Maintenance-2026 as its Method. InspectorSystemRole, a sensor capability, algorithm-possession wording, the Method, and TurbineInspectionProcedure-v3 do not perform the inspection. Use A.3.2 to decide whether that last episteme is a MethodDescription. Calibration state, Method adequacy, report quality, and acceptance remain separate.
Two Review Commissions
ProjectReviewAppointmentAssignment is a declared species. It declares three participant positions: HolderSystemSlot, local assigned kind and ReviewCommissionSlot. ReviewAssignment-A and ReviewAssignment-B have Alice and ReviewerSystemRole in common but different commissions, and both cover the same interval.
Commission A’s request asks Alice to compare two pump alternatives; Commission B's request asks her to review a budget. The available correspondence identifies Alice's undertaking of A's request, and the execution and return facts show her comparison of those alternatives answering that same request. Following the required contribution through those facts establishes the link from ReviewWork-A to assignment A. The budget review has its own history and pairing; the shared holder and interval establish neither crossed pairing.
If the available facts say only that Alice performed review Work while both appointments covered the interval, the attribution remains unresolved. The same is true if both requests ask the same question and the history does not distinguish which appointment the action fulfilled. The readable projection “Alice is reviewer” selects neither assignment.
Reviewer and Review Report
CommissionReviewAssignment is a declared species. It declares three participant positions: holder, local reviewer kind, and commission. Its rule applies to admitted review commissions and says that the fixed holder is appointed under the identified commission to supply the review contribution; one occurrence is the maximal uninterrupted interval for which that rule stays true. ReviewAssignment-82 is its occurrence for Alice and Commission-82, and it covers ReviewWork-82. The case identifies this as the assignment under which Alice performed that Work.
ReviewReport-82 is a separate U.Episteme; it may state the attribution, and evidence may support reliance on that statement, but neither creates the Work–assignment fact. Use A.15.PROD only for a current report-inception claim. The report is neither the performer nor the attribution.
Standard Used during Safety Work
A safety MethodDescription cites a standard, and source prose says that the standard has a “normative role”. Do not create an assignment for the standard. The standard remains an episteme in the external-rule, source-use, specification-use, or evidence relation selected by the claim.
A safety engineer or tool System can separately hold a covering safety-analysis assignment and perform dated safety Work. Attribution names the assignment occurrence and the case fact linking it to the Work; it does not use the standard as performer.
Access Label and Approval Work
An access directory says Alice has DB-Admin. That entry describes an access or policy relation under its own scheme; the entry alone does not establish an ApproverSystemRole assignment.
ApprovalCommissionAssignment is a declared species. It declares three participant positions: holder, local approver kind, and ApprovalScopeSlot. Its rule applies to admitted release scopes and says that the fixed holder is commissioned to supply approval within the identified scope; one occurrence is the maximal uninterrupted interval for which that rule stays true. ApprovalAssignment-481 is its occurrence for Alice and the current release scope. If it covers ApprovalWork-481 and the case identifies it as the assignment under which Alice performed that Work, the attribution is grounded. The directory entry may support a separate authorization claim but cannot substitute for the assignment or its link to the Work.
Distributed Performers and Child Work
A.13 first recovers ReviewTeam-9 and Alice as the two exact actual performers through TeamReviewAssignment-9 and MemberReviewAssignment-A9, and A.15.1 independently admits JointReviewWork-9. Because this example expressly represents assignment-bound attribution for each performer, F.6 afterward establishes one relation for each Work-assignment pair through those same assignments. Neither assignment identifies or stands for the other performer.
If AliceFindingCheckWork-9 is separately admitted as child Work after its own A.13/A.15.1 basis passes, the child attribution uses its own covering assignment and F.6 link. To explain how the joint review reached its judgement, follow the finding into the team’s comparison and decision. If Alice instead sent the same finding to a different commission, the attribution to her assignment could still be sound. Evidence of what the team used distinguishes those histories. A request only for the name of the reviewer on the finding can finish with Alice.
Passive Test Article
TestArticle-7, admitted as a U.System, holds TestSubjectAssignment-7 throughout ValidationWork-7. ValidationRig-2, also admitted as a U.System, actually performs the Work under its own ValidationPerformerAssignment-7; only that Work-attribution link is established. The test article's assignment and presence during the interval do not make it a performer. If the project needs to say that the article participated passively in the validation, use the domain rule that defines that participation; while no such rule is current, return the A.6.RCD missing-governor result rather than treating the assignment as participation.
Bias Annotation
Conformance Checklist
WorkOccurrenceSlotnames one datedU.Workoccurrence already admitted by A.15.1 without relying on F.6; its actual performers already have the A.13 core for the exact action and any scope, working situation and window needed by the attribution use, and any characteristic profile is required only by its own Grade, autonomy, criterion-dependent, profile, or assurance use.SystemRoleAssignmentSlotnames one assignment occurrence of a declared species underU.SystemRoleAssignmentthroughU.RelationRef.- The assignment's declared species, all identity-bearing participants, rule, applicability, and uninterrupted occurrence identity remain recoverable. Each species keeps its SlotSpec
ValueKinddomains distinct from the participant values supplied by the occurrence;AssignedSystemRoleKindSlottakes one kind value from its declared local system-role-kind domain. - The case establishes that W was performed under RA; the assignment's existence, matching holder, and temporal overlap do not establish that link.
- The assignment holder is the System that actually performed W.
- The assignment predicate covers the selected Work interval; attribution to a Work part first identifies that part as
U.Work. - Checks 2, 3, 5, and 6 constrain a valid attribution but do not by themselves establish it.
- Overlapping assignments are distinguished by all their participants and by checking each Work–assignment link from the case; an unresolved case yields no blanket attribution.
- Every positive precise attribution for a top-level or child Work occurrence has its own covering assignment and F.6 link to that already admitted Work; lead, team, member, allocation, coordination, and responsibility claims do not substitute.
- A passive assigned System receives no performer attribution from assignment or overlap; any claimed passive participation uses the rule that defines it or returns the A.6.RCD
missing-governorresult. - F.6 uses
performedUnderAssignmentand introduces noRoleEnactmentFactor generic assignment duplicate. - Assertions and evidence may support reliance on the attribution claim but do not make it true.
- Classification, assignment state, capability, Method, result, evidence, source reliance, publication, responsibility, authority, gate, and decision claims use direct patterns.
- Any selected model-use structure is designated by the receiving assertion or use, not by an optional generic slot.
- Missing evidence leaves reliance unresolved rather than proving non-attribution; missing pair grounding leaves the positive relation unasserted.
- Source shorthand is unfolded before a receiver depends on hidden values.
- The Method enacted by W remains a separate fact, and no kind, assignment, capability, Method, or description is made the actor.
- A short practitioner sentence may omit declaration and occurrence detail only after the Work–assignment link and its constraints are established.
Common Anti-Patterns and Repairs
Consequences
Benefits. Assignment and Work remain independently identifiable, while attribution becomes a direct relation that can be cited, supported, corrected, or left unresolved. The rule applies to every admitted holder System because each assignment occurrence exposes that holder through its species-declared holder slot.
Costs. Reliance-bearing use must recover both the assignment occurrence and its declared species rather than stop at a familiar label. A compact sentence can split into assignment assertion, Work occurrence, attribution, Method enactment, change or production claim, result episteme, and evidence relation when the receiving use needs them.
Limits. F.6 determines neither classification, capability, readiness, Method validity, Work success, result acceptance, permission, authorization, responsibility, access, nor evidence sufficiency. It governs only attribution of one Work occurrence through one assignment occurrence.
Rationale
The direct relation is needed because assignment and Work have different occurrence identities. performedUnderAssignment is an additional world-side fact, not a field stored inside either participant. A separate assertion can say that the assignment obtains, the Work occurred, or their attribution relation obtains.
Using the family ValueKind in F.6 does not license a family-wide assignment signature. It lets F.6 project the actual holder from each occurrence through its species-declared holder slot while preserving any commission, position, locus, or other participant of that species. Creating a generic assignment for F.6 would duplicate the episode and weaken attribution identity.
Making a log, status, decision, or evidence item a participant would confuse attribution with knowledge of attribution. Creating RoleEnactmentFact would duplicate Work and the same relation. Treating a matching holder and temporal coverage as enough would instead attribute one Work to every overlapping assignment held by its performer. The two-participant relation avoids these errors: the case fact linking Work to assignment is checked separately, while assertions and evidence can change without rewriting the Work, assignment, or their link.
SoTA-Echoing and Source Use
Internal basis. A.2.1 and A.6.5 supply the declaration-local slot, domain, and participant-value discipline. A.2.5 keeps assignment state distinct from the assignment occurrence. A.6.REL supplies relation obtaining and occurrence identity. A.13 supplies the actual-performer basis; A.15.1 independently admits dated Work using that basis. F.6 uses these as its governing FPF neighbours; they do not replace comparison with external work.
gUFO is the current comparator because it directly addresses the classification–relational-aspect–occurrence separation at issue; PROV-O and OCEL answer narrower representation and logging questions and therefore serve as lineage and stress tests. A new edition number, publication status, or harmless wording change does not reopen the comparison. A material change to a distinction used above, or a competitor that offers a better direct Work–assignment attribution solution with at least the same exactness, readability, and use cost, reopens only the affected row and F.6 loci.
When to revisit the comparison. If A.2.1 or A.6.5 changes how an assignment species declares slot domains or how an occurrence supplies participant values, revisit §§4.3, 5, 7.1, and 9. If A.6.REL changes relation obtaining or occurrence identity, revisit §§4.1–4.2, 5, 7, and 9. If A.13 changes the actual-performer or obtaining-assignment basis, or A.15.1 changes Work admission, revisit the affected attribution conditions, examples, and checks. If a better direct Work–assignment solution changes the source decision, revisit §13 and only the solution or examples that depend on it. Wording or publication changes that leave these meanings intact require no refresh.
Relations
Builds on: A.6.REL for relation obtaining and occurrence identity; A.2 for system-role kinds; A.2.1 for direct assignment species; A.13 for the actual-performer basis; and A.15.1 for dated Work.
Uses when current: A.2.5 for assignment state; A.2.2 for capability; A.3 and A.15 for Method and Work alignment; A.10 for evidence; A.15.4 for encountered-material reliance; C.3.3, F.9, and A.6.9 for cross-context use; and A.1.1 only when a selected model-use structure changes the receiving interpretation.
Coordinates with: F.4 for system-role-kind descriptions; F.5 and F.18 for names; E.17 for publication; and E.10.ROLE for ambiguous source wording.
Completion Conditions
F.6 use is complete when the reader has one of these results:
- a grounded finding that the direct
performedUnderAssignmentrelation obtains or fails to obtain for the exact Work and assignment occurrences; - an unresolved attribution assertion naming the missing exact-pair fact, assignment species or participant, coverage, performer, or support claim; or
- a corrected route because the current claim concerns classification, assignment, state, capability, Method, evidence, source reliance, result, publication, permission, authority, responsibility, access, gate, or decision rather than performed-Work attribution.
F.6:End
Display Source-Local Comparisons in a Concept-Set Table
Put exact local meanings and established relations side by side. The table displays the argument. Each claim needs grounds independent of its place in the table.
Type: Architectural pattern.
Status: Stable
Depends on. E.10.D1 Recovering What “Context” Means in Use; F.0.1 Source-Local Meaning Recovery; F.1 for finding and selecting sources for the question; F.2 for source-local term harvesting and normalisation; F.3 Source-Local Sense Clustering; F.17 for optional exact cells; F.5 Naming Discipline; F.9 for established relations between distinct source-local meanings and the separate claims about their bounded uses.
Coordinates with. F.4 SystemRoleKindDescription; F.6 SystemRoleAssignment and Performed-Work Attribution Check; direct Part C patterns for the compared values; C.16 for characteristics; A.6.9 when umbrella sameness wording must be repaired before a relation is asserted.
Aliases (informative). Concept-Set table; comparison grid; Giants’ table.
Intent & applicability
Intent. Give a reader one compact surface for comparing exact source-local claims, optional F.17 cells, and any relations that already obtain between them. The table also shows the stated comparison or receiving use, direction, losses, evidence, and counterexamples. It makes a distributed argument readable without turning row membership into sameness or permission.
Use this when. Two or more selected sources must be compared for one named question, teaching contrast, designation choice, or receiving use, and prose alone scatters the relevant distinctions.
Do not use this when. One local claim is enough, or no receiving comparison is named. A table is optional. It creates no value, kind, relation, classification, assignment, evidence use, reliance, verdict, or authorisation.
Problem frame
Four common errors in cross-source comparison are:
- Silent equivalence: similar labels are treated as one meaning.
- Loss denial: a relation that holds is shown without direction or limitation.
- Name inflation: a new umbrella label is coined merely because several entries share a row.
- Cognitive scatter: source meanings, relations, evidence, and the receiving question are separated across documents.
Forces
Core idea (didactic)
A Concept-Set row is a didactic grouping for one stated comparison or use. It contains:
- the exact sources and editions;
- each source-local claim, or its F.17 SchemeSenseCell when durable addressability is useful;
- every already obtaining relation that matters here, with direction and declared loss;
- the separate conclusion about the named receiving comparison or use;
- the evidence or direct pattern that supports each substantive claim;
- a counterexample or boundary showing where the comparison stops.
The word set names the entries collected for display. It does not assert that they are one value. A row can show a grounded relation, such as identity, overlap, ordering, incompatibility or disjointness. When the comparison asserts no relation, say so. A claim that a particular relation does not hold needs its own grounds.
Use F.9 only for a relation between distinct local meanings. For every other relation, cite the pattern that defines, constrains, or tests it.
Minimal vocabulary
- Local entry — an exact LocalSenseClaim or optional F.17 SchemeSenseCell, with its source and edition and effective scheme.
- Obtaining relation — a relation already supported under its direct pattern; it is not inferred from co-placement.
- Direction — which participant is source and which is target when the relation is asymmetric.
- Loss or limit — what the relation or receiving use does not preserve.
- Receiving-use conclusion — the separate claim that judges whether and how the entries and relations may be used for one named purpose.
- Contrast row — a row that teaches a difference or an unresolved comparison and expressly asserts no sameness.
- Characteristic — a comparandum defined by its direct characteristic pattern; a table may display measured or target values but does not define the characteristic.
The table
Choose the fewest columns that make the current argument recoverable:
For a teaching contrast, the relation column may say none asserted and the conclusion may say keep distinct. For an F.9 relation, show its declared relation kind, direction, CL if the relation defines one, and loss. Do not compute a row-wide CL or replace the separate bounded-use claim with a table label.
Reading rules:
- Entries stay local. A cell cites the source’s expression and claim; it is not a translation supplied by the table.
- Relations stay direct. For each relation, cite the pattern that defines, constrains, or tests it and the evidence supporting the claim.
- Use stays separate. “These may be compared for this report” is a claim with its own basis, not a property of the row.
- Unknown stays unknown. A blank or unresolved relation is not an invitation to infer similarity.
- Loss stays visible. If the limit needs more than one line, link to the underlying relation or evidence rather than compressing it away.
Conceptual construction
- Start with what the reader must establish. Choose entries and columns for the distinctions that change the answer. If the question concerns what several contributions achieve together, identify the relation that would let the reader infer that result. Use F.0.2:4.1 to develop a missing cross-source argument. The finished row can show the argument itself or give the reader a direct route to it.
- Bring exact local meanings. Use F.0.1, F.2, F.3, and F.17 only as needed.
- Show established relations. Cite the direct pattern and evidence for each relation that matters.
- State the receiving-use conclusion separately. Say what the current comparison permits, with its basis and limits.
- Keep the row focused. Add an entry only when it changes the answer.
- State the outcome the evidence warrants. Show a contrast when a supported difference answers the question. Where the evidence leaves the proposed relation open, say which further conclusion depends on it and what could settle it.
Invariants
- Exact entries. Every filled source-local cell identifies the particular source, edition and claim, or the particular F.17 cell.
- No row-created relation. Co-placement, matching labels, or a shared FPF designation establishes nothing about the entries.
- Direct relation basis. Every stated relation cites its direct pattern and available evidence; F.9 is conditional, not universal.
- Separate receiving use. Any conclusion about comparison, substitution, reporting, or reuse is stated and supported separately.
- Direction, time stance, and loss. Asymmetric relations show direction; any design-time, run-time, or other temporal difference that changes the comparison is explicit; every material limitation remains visible.
- No automatic closure. Relations are not completed pairwise or transitively merely to fill a row.
- No universal row type. A
senseFamilylabel is not required and cannot substitute for an intensional account of what is being compared. - Parsimony. Keep only entries and columns that change the current answer.
- Didactic bound. Split a row when its combined distinctions prevent the intended reader from recovering the comparison. Keep related claims together when their joint reading is needed.
Micro-illustrations
The examples show table shapes. To use a row for a particular comparison, supply evidence for each positive relation.
(a) Class-order comparison
(b) Measurement comparison
(c) Contrast: process
Anti-patterns & remedies
Worked example and application sketches
Section 11.1 supplies exact source bases for one contrast row. Sections 11.2–11.5 show further applications; complete their source-local claims, relations and evidence before treating them as finished tables.
Actor wording across BPMN and PROV
Runtime occurrence comparison
Performed-Work attribution remains an A.15.1 and F.6 claim about actual Work and system-role assignment. The table supplies neither.
Measured value and target
Class inclusion and FCA order
A contrast row may show OWL class inclusion, FPF U.SubkindOf, and FCA concept order together. FCA orders formal concepts by inclusion of their extents, equivalently reverse inclusion of their intents, within a selected formal context (Rudolph, Săcărea and Troancă, 2015, §2.1). That order does not by itself establish an OWL class axiom or an FPF subkind fact. A positive correspondence requires an explicit interpretation and grounds for the chosen domain and membership criteria.
Role trigger word
Show NIST RBAC role as a permission grouping. Beside it, show the local claim that identifies a system-role kind whose members are Systems. Mark them distinct subjects. Use E.10.ROLE to recover other uses such as relation participation or signature position. Shared spelling alone gives no basis for assigning one senseFamily.
Reasoning with a comparison row
- Name the comparison. What question or receiving use makes the row worth having?
- Recover each entry. Cite its exact source, scheme, expression, and local claim.
- List only obtaining relations. For each, state direction when relevant, the pattern that defines, constrains, or tests it, the supporting evidence, the loss, and any temporal difference that changes the comparison.
- Judge the use separately. Explain what the named use may conclude and why.
- Keep open questions visible. A row can display a supported difference or identify a relation still to be tested. If the reader needs a joint conclusion, establish how the entries support it.
- Resist closure. Do not invent missing pairwise or transitive relations.
- Adding a source. Include it only when its claim or relation changes the answer, then re-evaluate the conclusion for this use.
- Keep evidence visible. A table cell never replaces the evidence-use or reliance relation.
Relations
Builds on:
- Use F.1 for the exact source cut and receiving question, and use F.2 and F.3 for exact expressions and source-local claims.
- Use F.17 only when durable local-meaning addresses are needed.
- Use F.5 for selected designations without making them the values being named.
- When the relation is between distinct source-local meanings, use F.9 to define and test it for those meanings and to state the separate bounded-use claim. A table creates none of these facts.
Constrains:
- F.4 may cite a table for reader navigation, but the direct kind description and C.3 membership criterion remain the basis for a local system-role-kind claim.
- F.6 may reuse a designation from the table; the row supplies neither classification, assignment, Work, nor performed-work attribution.
Used by. Part C patterns may use the table as a didactic comparison surface. B.3 may rely only on exact evidence and obtaining relations, never on row position or a computed row score.
Migration notes
- Relation changes. Update the exact relation cell and re-evaluate only the receiving-use conclusions that depended on it.
- New source. Do not auto-expand rows; add it only if it changes the stated comparison.
- Local claim splits. Replace the old entry with the relevant child claim or split the comparison.
- Use widens. Re-evaluate the new use directly; a former row label grants no promotion.
- Designation changes. Update F.5 wording without changing source-local entries or relations.
- Edition changes. Recover the successor claim and recheck affected relations; preserve the earlier source identity where historical claims remain relevant.
Acceptance tests
Static conformance
- SCR-F7-S01 (exact entries). Every local entry identifies an exact claim or F.17 cell and its source and edition.
- SCR-F7-S02 (no row-created fact). No relation or permission is inferred from co-placement, label similarity, or layout.
- SCR-F7-S03 (relation basis). Every positive relation cites the pattern that defines, constrains, or tests it, states direction where relevant, and cites its evidence.
- SCR-F7-S04 (receiving use). The row states the practical use conclusion separately from the relation between the entries and gives the grounds for that use.
- SCR-F7-S05 (loss disclosure). Material limitations and counterexamples remain visible.
- SCR-F7-S06 (parsimony). Every extra entry changes the current comparison or use.
Regression
- RSCR-F7-E01 (relation drift). A changed relation triggers re-evaluation of dependent use conclusions, not a global row score.
- RSCR-F7-E02 (sense split). A split local claim leaves no ambiguous cell reference.
- RSCR-F7-E03 (use integrity). No consumer treats a row label as licence outside the stated conclusion.
- RSCR-F7-E04 (no stealth growth). New entries create no silent relation, closure, or widened use.
Didactic distillation
“A Concept-Set table displays a comparison. Put exact source-local claims in it and show established relations with direction and loss. State separately what the named comparison or use may conclude and why. Row position, shared labels or a row-wide CL do not establish those relations or warrant their use. Use a row to display an established difference or a remaining question. For a conclusion that depends on several entries together, show the connection that warrants it.”
F.7:End
Mint-or-Reuse Decision for a Name
Type: Architectural pattern Status: Stable Normativity: Normative unless marked informative
Use This When
Plain name. Keep, reuse, or strengthen a name.
Use F.8 after the subject is known and a project must choose the least burdensome adequate naming treatment for one expression and one use. Start only when these four facts are available: the expression, the governed value or relation, its subject pattern, and the proposed naming use.
Typical triggers include:
- a familiar source word may be useful locally but would import the source ontology if promoted;
- a role-like word such as
ReviewerRole,AccessRole, orEvidenceRolemay name a system-role kind, another governed value or relation, or only ordinary wording; - an alias, subject-pattern name, or F.17 row may already serve the use, but only within its stated meaning and scope;
- a governed value may need a durable name, public row, or policy identifier; and
- pressure for a new U-kind appears. That last case stops before naming until E.24.UK has returned a stable admission disposition.
Primary working object. One F.8 disposition for the expression and proposed use. Ordinary naming needs no separate decision-occurrence or result-episteme identity. If a later claim must cite, replay or assign accountability to the decision itself, use §4.5.
Primary working reader. Use F.8 when you need to decide whether to keep an expression local, reuse a name or use a naming treatment that local wording or reuse cannot supply.
First useful move. Recover the four starting facts, then select the least burdensome treatment that supports the use: a local phrase, an existing designation, an alias, the subject pattern's name or an admitted F.17 row. Reuse known grounds for that choice; compare other treatments when their adequacy could change it. Open a cell, NameCard, public row or policy identifier only when the receiving use needs it.
What goes wrong if missed. A convenient expression is treated as the subject it merely names. Local or source wording becomes durable ontology; a row or alias gains uses it never admitted; a role-like word hides a kind, description, assignment, Work occurrence, or another governed relation; or a record is mistaken for the decision it describes.
What this buys. Teams get short usable names without creating duplicate kinds or naming records. When a stronger name is needed, readers can recover its governed subject and permitted use.
Not this pattern when.
- For one-off wording repair, use F.19; use E.10, E.10.ARCH, A.6.P, or the subject pattern when a meaning still needs recovery.
- If the governed subject or relation is not yet known, recover it first. For an unsettled U-kind proposal, use E.24.CD when the object is unclear and E.24.UK for admission.
- To constitute a
SystemRoleKindDescription, use F.4. To assign a system, use A.2.1. For precise performed Work, recover each exact actual performer through A.13 and let A.15.1 independently admit the dated occurrence; add F.6 only when the naming case or receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment. - For an obtaining relation between different local-sense projections, use F.9. Use F.17 when a public, Core-facing, durable, or cross-local row is needed.
- For a status, evidence use, policy, Method, Work, publication, or any other governed subject, use its subject pattern before naming it.
- After F.8 has selected a name family, use F.5 for its naming discipline and F.18 only for a durable naming settlement.
Problem Frame
Name pressure often reveals an unresolved subject. For example, one word may be proposed for a designation, local system-role kind, optional description of that kind, assignment occurrence, status value, policy identifier, or Work label. Shared spelling proves none of these identities.
F.8 therefore asks what the expression will designate and for which use before judging the wording. The decision determines whether the local expression needs a stronger naming treatment. It neither defines the governed value nor performs the later naming work.
Problem
When a project skips that decision, these errors can arise.
- Local or source wording becomes durable ontology. A temporary phrase or familiar standard term survives without a recovered subject and use.
- Role-like wording collapses different objects. A local system-role kind, its optional F.4 description, an A.2.1 assignment, performed Work, or another governed relation receives one misleading label.
- Reuse widens silently. An alias changes meaning, or an F.17 row admitted for naming is reused for equivalence, assignment, measurement, or structural inference.
- Naming is asked to admit ontology. A proposed U-kind enters F.8 before E.24.UK has settled whether the result is an admitted kind, reused kind, local kind, or recovered non-kind object.
- Identifiers and records act by proxy. A policy identifier lacks a resolvable specification, or a filled record is treated as the decision occurrence or result it describes.
- Locality labels become subjects. A review, team, project, or date label is used as if it created Work, assignment, evidence, status, or authority.
Forces
Solution
Treat mint-or-reuse as a decision about an already recovered subject, not a vote on wording. Start with four facts:
- the candidate expression;
- the governed value or relation;
- the subject pattern that defines or tests that value or relation; and
- the proposed naming use.
If a needed fact is missing, recover it through the subject pattern. Otherwise use the following dispositions to select a sufficient treatment. Reuse an already justified choice; examine another disposition when it could meet the use with less burden or preserve a distinction the proposed choice loses:
- keep a local phrase;
- reuse an existing designation;
- use an alias without changing the governed meaning;
- reuse the subject pattern's name;
- reuse an admitted F.17 row within its stated use;
- name a separately justified
SystemRoleKindDescriptionwhen that is the governed object; - open a durable naming settlement;
- propose a public row;
- introduce a policy identifier for an already recovered policy specification; or
- block or lower the naming use.
The ordinary result states the governed subject, proposed naming use, selected disposition and resulting name when any. Add a reopen condition when continued reuse needs it, and an explicit non-use boundary when F.19's grounded-contribution test warrants it.
For example, when a durable local designation is needed: “Under PatternReviewReferenceScheme-2026, use ReviewerRole as the Plain designation of ReviewerSystemRole for local review-method prose; select openDurableNamingSettlement and revisit the decision if the naming use becomes public or cross-local.”
The corresponding F.8 result labels are localPhraseOnly, reuseExistingDesignation, aliasOnly, reuseDirectPatternName, reuseAdmittedTermRow, nameSystemRoleKindDescription, openDurableNamingSettlement, proposePublicTermRow, introducePolicyIdentifier, and blockOrLowerUse. They are not new U.* kinds. A stronger result opens its subject pattern; it does not itself create a card, row, identifier, policy specification, or relation occurrence.
Decision Targets
Decision Sequence
Use the following questions to recover an adequate disposition without hiding a governed distinction. Skip questions already settled by the available basis.
- Recover the four starting facts. Name the expression, governed value or relation, its subject pattern, and the proposed use. If the value or relation is not available, stop and use its subject-recovery route; F.8 cannot establish it.
- Decide what needs naming. Several constituents may belong to one established whole, and several participants may enter one relation. If the recipient needs a name for that whole or relation, use its own identity and meaning. If the same expression would designate different objects or carry incompatible uses, distinguish the naming decisions the recipient needs. An ordinary heading can introduce several objects that the text keeps distinct.
- State the naming locality. Supply the naming reference scheme's designation and interpretation rules by value, and state the local-sense claim. Cite a
SchemeSenseCell, an obtainingLocalSenseBasisRelation, or a selected bounded-model-use Structure only when the naming use needs that object. - Apply F.14 when the needed naming treatment is unsettled. If ordinary local wording supports the use, choose
localPhraseOnlyand stop; use an already justified stronger treatment when the receiving use needs it. - Try an existing designation. Reuse it only when the value, kind, scope, occurrence identity, local sense, and proposed use match.
- Try an alias. Use
aliasOnlywhen the governed meaning is unchanged and lineage can expose the wording variation. An alias may not change kind, scope, occurrence identity, use, or authority. - Try the subject's existing name. Use the name supplied for the governed subject. A.2 and C.3 govern a local system-role kind and F.5 governs its designation; use F.18 only for a durable settlement and F.4 only for a separately needed
SystemRoleKindDescription. A.2.1 continues to govern any assignment. For precise performed Work, A.13 first recovers each exact actual performer and A.15.1 independently admits the occurrence; F.6 follows only when the naming case or receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment. - Try one admitted F.17 row. Reuse only the row's declared
AdmissibleUse. Local-sense reuse does not imply cross-local sameness; a row and equal spelling create no F.9 Bridge. - Create the next naming object only when its use justifies the cost. A stable local address may justify a cell; an enduring naming settlement may justify a NameCard; a public/Core/durable/cross-local need may justify an F.17 row. None implies the next object.
- Introduce a policy identifier only for a recovered policy specification. A local identifier can stop with that specification and its scope. If the mint history is cited, replayed, normative, cross-local, or accountable, recover its decision or choice occurrence through the subject pattern. If the occurrence cannot be established, give the applicable §4.5 outcome for the history claim. Keep any C.11 result, decision-making Work, result episteme, and record separate.
- Stop before naming an unsettled U-kind proposal. Select
blockOrLowerUse. If the governed object is still unclear, use E.24.CD; otherwise send the recovered proposal episteme or source construct to E.24.UK. F.8 does not test or admit the candidate. After E.24.UK returns a stableroot,same-individual-dependent,identity-dependent,reuse,local-kind, orrejectdisposition, re-enter F.8 only if the admitted or reused kind, local kind, or recovered non-kind object needs a designation. - Block or lower. If no disposition is justified, keep the expression local, quote it as source wording, or lower the claim.
Role Expression Boundary
A role expression is not enough to choose the object. For a system-role naming case, keep these four objects distinct:
Under the effective naming scheme, L designates K. Needing L does not create or require D; D may receive its own designation when a separate description is justified. Naming either object creates no A. The naming ReferenceScheme interprets the expression; it neither defines the kind nor assigns a system.
After A.2 and C.3 have recovered K, apply the naming ladder. Keep a one-off expression local when that is enough, and reuse an existing designation when it fits. If the kind needs a durable designation, select openDurableNamingSettlement, use F.5 to name K, and use F.18 for the durable settlement. Use nameSystemRoleKindDescription and F.4 only when the governed object is a separately justified description episteme D.
F.17 Row-Scope Consumption
F.8 consumes one named F.17 row and its declared use; it neither constitutes the row nor defines Bridge strength. F.17 keeps the row episteme, governed value, designations, cell, basis relation, any F.9 Bridge, edition relation, and publication package distinct. F.8 asks only whether AdmissibleUse covers the proposed naming use.
If the row does not admit the proposed use, look for another name or establish the grounds for extending the row. Narrow the naming use when that still meets the recipient’s need or they agree to the change. Wording alone establishes neither a stronger use nor cross-local sameness.
Accountable Decision Branch
Open this branch only when a receiving claim needs to cite, replay, or assign accountability to the mint-or-reuse decision occurrence itself. First recover that occurrence through the decision or choice pattern that admits it. The ordinary naming result remains valid without this branch.
Keep these objects distinct in the accountable branch:
- the governed value or relation and its subject pattern;
- the candidate expression, selected designation, and any alias;
- the effective naming reference scheme, local-sense claim, optional
SchemeSenseCell, and any obtaining two-participantLocalSenseBasisRelation; - the decision or choice occurrence and the pattern that admits it;
- any C.2.1 decision-result episteme and the record or carrier that designates it;
- any F.18 NameCard, F.17 row, policy specification, policy identifier, publication occurrence, form, or carrier; and
- a selected bounded-model-use Structure only when its organization changes interpretation for this naming use.
When a result episteme is needed, use the full projection below:
The block describes the result episteme. EntityOfConcernRef resolves to the decision or choice occurrence admitted through DecisionGovernorLocator; the predicate, participants, applicability, and identity basis show why that occurrence exists. GovernedValueSubjectPatternLocator identifies the pattern for the value being named. NonAdmissibleOverread is included only when admitted by F.19's grounded-contribution test. A C.11 ChoiceResult and dated decision-making Work keep their direct identities and relations.
To assert a decision occurrence, establish it under the rule for that occurrence. If no governing rule can state or test the claim, return missing-governor. If the rule exists but a deciding fact is missing, return missing-information. If sufficient facts fail the positive test, the positive claim is factually unsupported. Use A.6.RCD:0.1 for an inapplicable case or a negative claim with its own grounds. Leave an unestablished occurrence out of the block. When no separate result episteme is needed, state the warranted result in ordinary language.
Policy-Identifier Mint-or-Reuse Discipline
FPF treats policy identifiers such as Phi(CL), Phi_plane, Psi(CL^k), Aut-Guard, EmitterPolicyRef, insertion-policy identifiers, and acceptance-clause identifiers as versioned references whose meaning must be recoverable. The identifier resolves to a separate policy specification; it does not itself assert a gate outcome or permission.
PolicyIdentifier is the selected designator. PolicySpecificationRef resolves to the separate policy-definition episteme and pins an edition or equivalent digest when needed. A local non-accountable introduction can stop there with explicit local scope.
The conditional mint-occurrence fields are required when the use cites, replays, makes normative, reuses across the local boundary, or assigns accountability to the mint history; together they resolve one admitted decision or choice occurrence and the pattern, predicate, actual participants, applicability, and identity rule that establish it. If that use requires a mint history you cannot establish, apply §4.5 to that claim.
A C.11 ChoiceResult and any dated decision-making Work remain separate. MintDecisionResultEpistemeRef, when current, resolves to a C.2.1 episteme or accepted record describing the occurrence; the record does not perform the decision.
For FPF normative policy identifiers, the durable result episteme is usually an accepted E.9 decision record, but only after the decision or choice pattern has admitted the occurrence that record describes. A local non-exported and non-accountable identifier needs only its separately recoverable specification and explicit scope; it need not create a decision or result episteme. In every branch, the policy specification, identifier, any decision or choice occurrence, any C.11 result, any decision-making Work, and any record remain distinct.
Rules:
- No silent policy-identifier introduction. Every new identifier resolves the separate
PolicySpecificationRefand states its scope. A local non-accountable introduction stops there. A cited, replayed, normative, cross-local, or accountable mint history additionally resolves the decision or choice occurrence plus the pattern, predicate, participants, applicability, and identity rule that establish it; if the occurrence cannot be established, leave the history unasserted and apply §4.5. - Reuse is reference use. Reusing an existing identifier resolves the same identifier and policy specification. Resolve the original mint occurrence only when the current reuse consumes or asserts that history. Reusing the identifier does not restate policy semantics, turn a record into the occurrence, or silently create another decision.
- Gate checkability. A gate, crossing, Bridge, assurance, or publication claim that depends on a policy identifier includes
PolicyIdentifierReferenceor an equivalent resolvable structure admitted by its subject pattern. - Policy authority stays with the subject pattern. F.8 selects introduction or reuse of the identifier; it does not decide whether the policy permits Work, passes a gate, makes a relation obtain, or provides evidence.
- The identifier grants nothing by itself. Name, namespace, suffix, source prestige, specification publication, or decision record grants no permission, status, equivalence, or authority beyond the policy claim defined by its subject pattern.
Invariants
- Four facts, one use. Each disposition names the expression, governed value or relation, its subject pattern and the proposed use. Keep distinct designations identifiable when the reader needs them. A name for a whole can refer to it through its own criterion.
- Lightest sufficient result. Use adequate ordinary wording or an existing designation. Create a cell, NameCard, row or policy identifier only for the receiving need it serves. An unsettled U-kind proposal stops before naming.
- Reuse preserves meaning. Reuse or aliasing changes no kind, scope, occurrence identity, local-sense claim, admitted use, authority, or lineage. Shared spelling under another scheme establishes neither sameness nor an F.9 Bridge.
- Kind, description, assignment, and Work stay distinct. A designation may name a recovered local system-role kind
Kwithout creating the optional F.4 descriptionD. Neither name classifies a candidate, creates an A.2.1 assignmentA, or demonstrates Work. Other governed uses remain with their subject patterns. - Rows stay within admitted use. Reusing an F.17 row supplies only its
AdmissibleUseand no equivalence. - Ordinary and accountable decisions stay distinct. Ordinary F.8 use needs no decision occurrence or result episteme. When the use requires that history, test the occurrence under the rule for that decision or choice. Use the §4.5 result block when a separate result episteme is needed and its subject occurrence is established. If the occurrence cannot be established, give the applicable §4.5 outcome for the history claim.
- Naming objects imply none of one another. A designation, cell, basis relation, NameCard, row, identifier, publication occurrence, form, or carrier is created only for its receiving use. A selected Structure appears only when its organization changes interpretation of this naming use.
- Admission precedes naming. Before E.24.UK returns a stable disposition, F.8 can only block or lower the proposed naming use. Afterward it may name only the admitted, reused, local, or recovered non-kind object identified by that result.
- Policy identifiers remain references. Every identifier resolves a policy specification. Its mint occurrence and history are required only for the stronger uses stated in §8.1 and remain distinct from any C.11 result, decision-making Work, result episteme, or record.
- Labels grant no authority. A source title, suffix, row, record, or identifier creates no governed subject, obtaining relation, permission, evidence, equivalence, or publication authority.
Reasoning Checks
Use the table to check the decision. No particular notation or record is required.
Archetypal Grounding - worked cases
ReviewerRole Expression vs Review Report
The source label PatternReview_2026 is not a context object. Before using the label, identify what it denotes in the claim:
ReviewWork-82can be one datedU.Workoccurrence underA.15.1;ReviewPlan-2026-v3can be a separately constituted plan episteme or edition under its subject pattern;PatternReviewReferenceScheme-2026can designate the effective rules carried by value for a review use, provided those rules and their semantic dependencies are recoverable and interpret the review terminology; and- "used while deciding the label for the 2026 review method" can be claim content describing the decision-use setting without minting any context entity.
If the recovered ReviewerSystemRole kind needs a durable local designation, F.8 returns openDurableNamingSettlement: A.2 and C.3 keep governing the kind, F.5 governs its designation, and F.18 supplies the settlement. The recovered kind is the governed subject; F.4 enters only for a separately needed description, A.2.1 for an assignment, and A.15.1 for performed review Work.
The expression "review report has reviewer role" is a different case. ReviewReport-82 is an episteme. An evidence, source, or publication relation may later use it for an adequacy claim about a reviewed pattern; the report is not a U.System, is not classified by the review-system-role kind, and cannot enter its assignment relation. Its title establishes neither evidence use nor publication authority.
Actor Across BPMN and PROV
A manager wants one word, "actor", for a BPMN participant and a PROV agent in a diagram. First recover the two local senses under their ReferenceSchemes. If an obtaining F.9 Bridge relates the named cells and an F.17 row admits naming-only use, F.8 returns reuseAdmittedTermRow for prose and diagram labels only. This supports no governed-value identity, substitution, system-role assignment, or Work.
If the project later needs a local system-role kind under one scheme, it first recovers the kind through A.2 and C.3. F.5 then governs any new designation, with F.18 only for durable reuse; F.4 is added only if a separate description episteme is needed.
Access Role
An access-control source says ApproverRole. Under its naming ReferenceScheme, the expression may designate a permission grouping or policy relation. First recover the access, policy, status, or deontic claim and predicate. Only if A.2 and C.3 recover a local approval-system-role kind does F.8 consider a name for that kind. F.5 governs its designation, F.18 applies only for durability, and F.4 remains optional for a separately needed description.
Otherwise any needed durable designation belongs to the access, policy, status, or gate pattern. The Role suffix, a source card, or a selected model-use Structure creates no local system-role kind or assignment.
Policy Identifier
A gate profile proposes Aut-Guard-2026. F.8 treats this as a policy-identifier question only after the policy specification is recovered. Ordinary reuse resolves the identifier and specification.
Recover the mint decision or choice occurrence only when reuse relies on that history for citation, replay, accountability, supersession, or another named relation. If the case does not establish that history, use §4.5 for the outstanding claim. Any C.11 result, decision-making Work, result episteme, or record stays separate.
The identifier is not the specification, local system-role kind, Method, gate result, evidence value, permission, or source authority. It is a reference used by the pattern that defines or constrains the governed policy claim.
New U-kind Candidate
A team proposes U.InfluenceEdge because many documents use "influence". At F.8 entry there is no recovered governed value with a stable admission disposition, so F.8 returns blockOrLowerUse and stops naming.
If the expression still hides the governed object—for example, whether it is a causal relation or a characteristic—E.24.CD recovers that object or the unresolved proposal. A recovered governed object returns to its subject pattern; a surviving U-kind proposal goes to E.24.UK for root, same-individual-dependent, identity-dependent, reuse, local-kind, or reject. Only after that result is stable may F.8 reopen for a name of the admitted or reused kind, bounded local kind, or recovered non-kind object. F.8 creates neither the proposal object nor a public spelling and admits no kind.
Readable Disposition and Explicit Stops
The ReviewerRole case closes with one readable result. The recovered kind is a local U.Kind for U.System candidates, distinguished by its stable review contribution and tested by its KindSignature; any assignment remains separate. The result is:
Under
PatternReviewReferenceScheme-2026, useReviewerRoleas the Plain designation ofReviewerSystemRolefor local review-method prose. No existing designation or alias supports that use, so selectopenDurableNamingSettlement: A.2 and C.3 continue to govern the kind, F.5 governs its designation, and F.18 supplies the durable settlement. Reopen it if the proposed use becomes evidential, status-bearing, access-related, source-facing, published, or cross-local, or if another change of use or audience exceeds this local settlement's scope.
This naming result needs no decision occurrence or result episteme. If a later claim must cite, replay, or assign accountability to the decision, use §4.5. No naming-decision governor is available in this case, so that branch returns missing-governor rather than inventing ReviewerSystemRoleNamingDecision-2026-07-31.
C.11 applies only to a genuine local choice among available options. For any precise decision-making Work, A.13 first recovers the exact actual performer and A.15.1 independently admits the dated Work; F.6 follows only when the later claim expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment.
EvidenceRole stops earlier and does not enter F.8. The known subject is ReviewReport-82 : U.Episteme, proposed for evidence use concerning an adequacy claim. Still missing are the target claim and polarity, the evidence-use relation and relation kind, the provenance and any assurance or reliance use, including a validity window when that use depends on it, and one subject pattern that defines the relation. Apply that pattern and keep the wording local until those facts are recovered.
PatternReviewReferenceScheme-2026 may interpret the source wording, but the review label creates no evidence relation, system-role kind, description, assignment, authority, or publication. No SchemeSenseCell, LocalSenseBasisRelation, or selected Structure is needed merely to record this stop.
Re-enter F.8 only after one governed relation, its kind, its subject pattern, and the proposed naming use are available. If the target claim, polarity, provenance, assurance or reliance use, or a required validity window changes, reopen the subject claim rather than the name.
Bias-Annotation
F.8 counters two shortcuts: a familiar word is treated as proof that a stronger name is needed, or a record is treated as the subject or decision it describes. Recover the four starting facts, choose the lightest disposition, and add a Structure, decision result, NameCard, row, or publication object only when its own receiving use requires it.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Consequences
Good consequences:
- vocabulary grows only when a receiving use needs a stronger name;
- role-like, source, and record-like expressions return to their governed subjects before naming;
- aliases and F.17 rows retain their admitted meaning and scope;
- F.5 and F.18 receive a recovered subject and a selected naming path; and
- accountable decisions and policy identifiers remain inspectable and distinct from their records.
Costs:
- authors must recover the subject, pattern, use, scheme, and local sense before naming;
- mixed expressions need separate decisions, and some attractive words stay local or remain aliases;
- durable or public use may require its own NameCard, Bridge, row, reliance, decision result, or publication object; and
- a proposed U-kind receives no F.8 name before E.24.UK settles admission.
When to revisit a naming rule.
- If F.14, F.5, F.17, or F.18 changes the lightest-sufficient naming ladder, row-entry threshold,
AdmissibleUse, or escalation to a stronger naming object, revisit §§0 and 4.1–4.4, case 7.2, checks 03–09, and the compact corpus entry. - If A.2, C.3, F.4, A.2.1, A.13, A.15.1, or F.6 changes how a local system-role kind is recovered, how
L,K,D, andArelate, how an exact actual performer and Work are admitted, or when precise assignment-bound attribution enters, revisit the corresponding target rows, step 7, §4.3, cases 7.1, 7.3, and 7.6, invariant 4, and checks 06–08. - If A.6.RCD, C.11, C.2.1, or A.15.1 changes how a decision or choice occurrence, result, result episteme, decision-making Work, or missing governor is established, revisit §4.5, the accountable stop in 7.6, invariant 6, and check 10.
- If E.24.CD, E.24.UK, A.8, or A.11 changes object recovery, admission dispositions, or the admission-before-naming order, revisit the entry and ordinary conditions for use, stopping, or returning, the pre-admission target and step 11, case 7.5, invariant 8, and check 12.
- If the policy subject pattern, E.9, A.6.RCD, C.11, or C.2.1 changes the policy specification–identifier distinction or the support required for mint history, revisit the policy target, step 10, case 7.4, §8.1, invariant 9, and checks 10 and 13.
- If a source used in §12 changes a distinction that F.8 adopted, or a better current source preserves the needed precision and readability at lower use cost, revisit that source row and only the F.8 loci named by it.
A new edition number, publication status, link, harmless wording repair, or added example does not reopen F.8 when these meanings stay unchanged. A changed disposition, governed subject, kind or relation, admitted use, ordering dependency, or better solution reopens only the dependent loci named above.
Rationale
A naming mistake is often a subject or use mistake. Asking “what name should we use?” too early lets wording decide ontology, scope, or authority. F.8 therefore recovers the subject and proposed use before selecting a naming treatment.
F.8 is narrower than F.18. It decides whether the expression stays local, reuses something available, or opens one stronger naming path. F.18 runs a durable settlement, candidate comparison, NameCard, lineage, and any later public-row gate. F.8 creates neither the governed value nor the objects produced by the stronger path.
The system-role branch shows why the ordering matters. A designation L, local system-role kind K, optional F.4 description D, and any assignment A are different objects. The word role can also point to another governed subject. Recovering that subject first keeps naming from creating a false kind or assignment.
The accountable branch follows the same rule. A decision occurrence, C.2.1 result episteme about it, C.11 result, decision-making Work, and rendered record answer different questions. Section 4.5 opens only when a receiving claim needs that stronger traceability; ordinary F.8 use stops earlier.
SoTA-Echoing - Source-Use
Source selection. ISO 704 and SKOS are current standards or durable lineage, not claimed as research frontiers. The gUFO–OBO row combines a current formal comparator with an operational ontology-maintenance practice. The Cedar row is a current research-and-practice comparator. A source belongs here because it changes an F.8 disposition or boundary, not because it is official, popular, or easy to find.
Internal FPF basis.
- F.14, F.5, F.17, and F.18 supply the local-phrase, designation, alias, row-use, and durable-naming ladder.
- A.2, C.3, and F.4 keep designation
L, local system-role kindK, and optional descriptionDdistinct; A.2.1 governs assignmentA. For precise performed Work, A.13 recovers each exact actual performer and A.15.1 independently admits the dated occurrence; F.6 is a later separate relation only when precise assignment-bound attribution is expressly consumed. - E.24.CD, E.24.UK, A.8, and A.11 recover an unclear object and decide kind admission before F.8 names the result.
- A.6.RCD, C.11, C.2.1, and E.9 govern any accountable decision occurrence, separate result, result episteme, and policy-history record.
- F.1–F.3 and F.9 govern local-sense discovery and an obtaining Bridge; A.1.1 and A.22 govern any selected bounded-model-use Structure. F.8 cites those objects only when the naming use needs them.
Source-use boundary. External sources can supply candidate expressions, a comparison pressure, or a narrow representation test. They do not select the F.8 disposition, establish the governed subject, make a relation obtain, admit a kind, or grant authority. Those results remain with the named FPF pattern and the recovered facts.
Relations
Builds on. A.7, E.24.UK, A.8, A.11, E.10, E.10.ARCH, F.19, F.1, F.2, F.3, F.5, F.9, F.14, F.17, and F.18.
Coordinates with. A.2, A.2.1, A.2.5, A.2.7, A.6.5, A.6.RCD, A.15, A.15.1, C.11, F.4, F.6, F.10, F.13, F.15, C.2.1, C.3, E.9, E.24.CD, and E.24.PUB, plus the subject pattern for any other governed value.
Constrains.
F.5names only after F.8 has selected the naming case.- Use
F.4only for a separately neededSystemRoleKindDescription; naming the local kind itself does not require that episteme. F.9governs an obtaining Bridge between F.17 cells;F.17governs admitted public-row use before F.8 reuses it.F.18supplies a durable naming settlement when the receiving use needs one.F.14supplies the anti-explosion stop before every stronger F.8 disposition.F.15may check the resulting distinctions; it neither chooses the disposition nor creates a naming object.
Does not replace. The subject pattern for any governed value or relation. For example, it does not replace the rules for a system-role kind, assignment, performed Work, decision occurrence, status, evidence use, policy, relation slot, selected Structure, or description episteme.
Didactic Memory
Name the subject and use before judging the word. Use the lightest treatment justified for the use. Compare alternatives when their adequacy could change that choice. An unsettled U-kind goes to E.24.UK before naming; an accountable decision opens §4.5 only when the decision occurrence itself must be used. A name, card, row, identifier, publication, or record creates neither the subject nor any relation or authority it mentions.
F.8:End
Relate Local Meanings across Contexts (Alignment and Bridge)
Type: Pattern Status: Stable
What relation between these two local meanings would answer the question? What can the recipient infer from it?
Type: Architectural pattern.
Normativity: Normative.
Builds on: F.17 for exact scheme-based SchemeSenseCell identity and SenseCellAddressRef; F.18 for designation selection; C.2.1 for assertion and description-episteme identity; F.0.1 for recovering each source-local meaning before proposing a relation between the two; F.7 and F.8 for downstream naming and reuse decisions.
Coordinates with: A.6.REL for demand-driven occurrence individuation; C.2.1 for assertion, occurrence-description, and Card identity; E.24.PUB for publication occurrence, form, and carrier; A.10 for descriptive evidence-provenance paths citing independently established direct relations and for local reliance dispositions; B.3 for actual named assurance claims and their bounded AssuranceResult values; E.10.ROLE for claim-bearing source wording with role; A.2, C.3, F.4, F.5, and A.2.1 for local system-role kinds and assignments; A.13 and A.15.1 for exact actual performers and independently admitted Work; F.6 only for a Bridge use that explicitly relies on precise assignment-bound attribution; A.6.5 for relation-slot discipline; C.29 for mathematical-lens use; A.6.3.CSC for controlled coarsening; C.26.1 and C.26.2 for quantum-like export boundaries.
Plain entry cues (informative). Context-to-context translator; sense bridge.
Intent and applicability
Intent. Govern one actual semantic Bridge relation between two exact F.17 SchemeSenseCell values from different semantic contexts. Keep that occurrence separate from every assertion, Bridge description episteme, Bridge Card, registry record, publication occurrence, publication form, presentation carrier, bounded-use claim, evidence or assurance relation, and object created when a proposed use is performed.
Applicability. Start with the claim the recipient needs. Comparing two standards can call for a relation between their terms, an account of how their rules apply together, or a decision about one job. Use F.9 for a correspondence or difference between local senses. F.0.2:4.1 helps construct the cross-source account if that is what the recipient needs.
Primary EntityOfConcern in plain terms. One actual correspondence or difference between two exact local senses. This pattern concerns the direct Bridge occurrence, not a card, context, transport chain, work process, local system-role kind, assignment occurrence, evidence item, or global meaning layer.
First move. What difference between the source and receiving interpretations could change the requested answer or its grounds? Use that to choose the senses and the correspondence to test. Establish the relation under its own predicate. If the question concerns a proposed use of that relation, judge the use separately, with its direction, correspondence rule and tolerated loss.
For that bounded-use claim, check ordinary reliance under A.10; use B.3 only when an actual named assurance claim is current. If the use happened, recover its Work, assertion, publication, relation, operation application, or other object under its subject pattern. Add a Bridge Card only when a reusable package is worth maintaining.
Primary working reader. An author, checker, or practitioner deciding first whether a cross-local semantic relation obtains and then whether it supports one named use.
Use this when. Use F.9 when a receiving claim needs an exact semantic relation between two local senses whose <ReferenceScheme, LocalSenseClaim> interpretation bases differ. Different schemes, identical spelling, a mapping implementation, or a request for comparison does not establish that relation.
What goes wrong if missed. Teams turn shared labels and convenient mappings into silent equivalence, substitution, structural inference, status transfer, classification under a local system-role kind, or assignment to it. They also mistake evidence about a proposed use, or a card, for the relation itself.
What this buys. A reader can see which relation is true, which proposed use is being judged, what evidence supports reliance on that judgement, and whether any downstream act happened. Those facts can change independently without silently merging local meanings.
Not this pattern when. Skip F.9 within one semantic context or when the live question is a local system-role kind, assignment occurrence, performed-work attribution, evidence use, status use, source use, publication, assurance, authorization, a gate, a decision, or a mathematical-lens operation. Use the subject pattern for that object; cite F.9 only when cross-context semantic correspondence is also needed.
Recognition versus assurance note. Resolving the endpoint senses and testing the direct Bridge predicate recognizes the semantic relation. A separate C.2.1 claim judges one bounded use. A.10 states whether ordinary evidence reliance passes. When an actual named assurance claim is current, B.3 supplies its bounded result for the same use. None of those steps supplies legal, policy, or deontic authorization.
Problem frame
Cross-context work fails in predictable ways:
- String-equals fallacy. An identical spelling, such as "process", is taken as identical meaning.
- Relation-to-use jump. A true semantic correspondence is treated as sufficient for whatever comparison, substitution, translation, or publication is wanted next.
- Design-run jumping. Design artefacts are substituted for run-time occurrences, or run-time occurrences are treated as design definitions.
- Direction amnesia. A symmetric relation is read as two use licences, or narrower and broader senses are used in the unsafe direction.
- Loss blindness. The proposed use does not name which differences it tolerates.
- Evidence and permission collapse. A score, card, or assurance record is treated as the semantic relation, authorization, or proof that the use occurred.
F.9 answers these failures by separating the direct relation, the bounded-use proposition, reliance on that proposition, optional packaging, authorization, and any actual receiving object.
Problem
A shared label across contexts can look like identity or permission before any semantic relation is tested. Even after a Bridge obtains, its truth does not answer whether one particular comparison or substitution is suitable. The problem is to preserve useful cross-context work while keeping the relation, the proposed use, its evidence, and the downstream act individually testable.
Forces
Solution
Identify the conclusion for which the meanings matter. For a question about their semantic relation, choose the two senses whose correspondence would settle it. For a question about the practice they describe, follow each contribution to the result the recipient needs. One may supply information to the other, or independently produced findings may both have to apply to the same case. Which distinction would change what the recipient can conclude? A semantic relation can supply a premise. Establish the connection in the practice when the conclusion depends on it. If the inquiry needs a Bridge, resolve its two local senses as exact F.17 SchemeSenseCell coordinates:
For F.9, semantic bounded context is a Plain practice name for the local interpretation basis recovered from one exact cell's <ReferenceScheme, LocalSenseClaim> projection. It names this interpretation basis, not a separate FPF entity, relation participant, or selected model-use Structure.
Two expressions under the same projection remain with ordinary designation and scope operations. Different projections make a Bridge question possible but do not make a Bridge obtain.
When the two cells are from different semantic contexts, declare one relation-semantic BridgePredicateProfile and test it against their current meanings. Shared spelling, different schemes, a mapping implementation, a card, a registry entry, evidence, an assessment score, or publication establishes none of those facts by itself.
Direct Bridge relation
Bridge is a direct species of U.Relation. Its reusable RelationSignature has exactly two participant meanings:
Only the two endpoint meanings are RelationSignature participants. CL, Loss Notes, U.ClaimScope, an admitted-use qualifier, evidence, counterexamples, policy, time or as-of values, BoundedModelUseStructure, description, Card, publication, registry identifier, form, and carrier are qualifiers or neighboring objects. No proposed-use field, use direction, use-specific rule, permitted-loss tolerance, assertion, or reliance result is a third participant.
The reusable Bridge declaration is one independently constituted C.2.1 episteme whose exact EntityOfConcern is the direct Bridge relation kind. The same declaration episteme is used relation-facing as the compatible RelationSignature; its two SlotSpecs declare participant meanings but create neither endpoint nor occurrence.
The relation kind, declaration episteme, RelationSignature use, SlotSpecs, actual cells, obtaining occurrence, assertion, occurrence-description episteme, Card, and publication remain distinct.
An F.9-local BridgePredicateProfile is a by-value predicate declaration, not a U-kind, participant, card, claim, or evaluation result. Direction is stated in the Bridge kind and endpoint orientation when the predicate is asymmetric. Its identity-bearing content is only:
- the
BridgeKindand its kind-defined symmetry or endpoint orientation; - the exact source and receiving endpoint-sense readings, including their
senseFamilyreadings where material; - the relation-kind-specific congruence, difference, or loss condition, distinct from observed Loss Notes and a proposed use's permitted-loss tolerance;
- the applicability and as-of basis for testing that condition;
- the Boolean truth condition; and
- every stop dependency whose absence prevents a truthful result.
The profile contains no proposed-use field, use direction, use-specific correspondence rule, permitted-loss tolerance, bounded-use proposition, assertion polarity, evidence-reliance classification, assurance claim, authorization, or receiving object.
Bridge(SourceSenseCell, ReceivingSenseCell; BridgePredicateProfile) obtains exactly when:
- both endpoint references resolve to exact F.17
SchemeSenseCellvalues; - their semantic-context projections differ;
- the profile applies to those endpoint readings at its stated as-of basis;
- the current endpoint meanings satisfy its kind-specific correspondence or difference condition and Boolean truth condition; and
- every required dependency is present.
If an endpoint is unresolved, the projections are the same, a dependency is missing, or the predicate is false or unresolved, assert no positive occurrence and state the exact exit: ordinary designation, unresolved SenseCell endpoint, same semantic context, missing Bridge dependency, Bridge predicate false, or Bridge predicate unresolved.
Admitted-use qualifier. The Bridge declaration admits this relation only as the semantic-correspondence or semantic-difference premise for a comparison, explanation, translation, naming, or other bounded-use claim. The nearest non-use is substitution on the strength of the Bridge alone. The Bridge creates no scope result, model-use crossing, local system-role kind, assignment occurrence, Work, evidence authority, status transfer, U-kind admission, publication, or other subject relation. This use boundary is a declaration or description qualifier; it is neither a participant nor profile identity and grants no specific use.
Non-optional occurrence identity and recurrence rule. BridgeOccurrenceIdentityRule identifies the occurrence by the exact endpoint cells together with the exact profile. For an asymmetric kind, the ordered source-to-receiving tuple is identity-bearing and an inverse relation requires another profile and directed occurrence. For a symmetric kind, swapping only the readable presentation of the same canonical endpoint pair does not create another occurrence. A changed endpoint or changed relation-semantic profile identifies another candidate.
A Bridge is non-recurrent for one fixed canonical endpoint tuple and exact profile: at most one occurrence has that identity. Repeated tests, assertions, descriptions, Cards, registry rows, or publications neither split nor repeat it.
A later applicability or as-of basis changes the profile and therefore opens another occurrence candidate. If a claimed lapse and resumption cannot be represented by an endpoint or profile change, stop at missing Bridge recurrence basis rather than inventing two occurrences with one identity. Changed proposed use, direction, rule, tolerance, evidence path, reliance disposition, assurance claim, Card, registry entry, publication, form, or carrier never reidentifies or recurs the fixed Bridge.
Judge a bounded use separately
When the question is whether an established Bridge b suits a particular use, work back from the recipient’s intended inference or action. Which differences in the source can the receiving rule merge while still giving a warranted answer? Try source cases that the receiving expression treats alike. If they call for different answers or different grounds that the recipient needs, preserve the distinction, restrict the proposed use to cases where it is immaterial, or reject that use. State the proposed use in ordinary language. Give the following particulars in the bounded-use claim.
u: what the reader proposes to compare, substitute, translate, publish, or otherwise do;d: the exact source-to-receiving direction for that use;r: the use-specific correspondence rule;t: the semantic-loss tolerance for that use; and- whether the claim is affirmative or negative.
The resulting C.2.1 claim asks whether b is suitable for <u,d,r,t>. Its exact EntityOfConcern is b; its ClaimGraph designates u, d, r, t, and polarity; its effective ReferenceScheme makes those designations interpretable. That C.2.1 triple identifies the claim episteme. Changing u, d, r, or t changes the claim, not the Bridge.
An affirmative claim is one premise for the proposed use. It is not a permission, authorization, evidence-provenance relation, reliance classification, assurance claim, decision, or occurrence of that use. A negative claim says that the Bridge is not suitable for the named use; it does not make the Bridge cease to obtain.
For ordinary evidence reliance, recover the descriptive A.10 evidence-provenance path, the independently established direct relations it cites, and local RelianceDisposition for the same bounded use. Only pass supports reliance on the affirmative claim for that use; degrade supports only its named narrower use, while abstain, reopen, evidence-needed, assurance-needed, or blocked-current-use supplies no passing classification.
Use B.3 only when an actual named assurance claim about the proposed use is current. Require its result for the same bounded assurance use; a non-positive disposition stops or narrows that use. A direct domain rule may require the claim, but the Bridge, display, consequence, or A.10 disposition does not create it.
Neither an A.10 passing disposition nor a B.3 AssuranceResult with disposition=supported-for-use is legal, policy, or deontic authorization. If authorization is needed, recover it under its direct pattern.
If a later claim says the use happened, recover the actual Work, assertion episteme, publication occurrence, direct relation, operation application, or other object under its own pattern; the u designation in the ClaimGraph names the proposed use and is not that occurrence.
Minimal vocabulary
- Semantic context - Plain shorthand for the interpretation basis
<ReferenceScheme, LocalSenseClaim>recovered from an exact F.17 cell; it is not a separate entity or participant. - SchemeSenseCell - the exact F.17 local composite value
<ReferenceScheme by value, LocalExpression, LocalSenseClaim>. - SenseCellAddressRef - an address that resolves one exact
SchemeSenseCell; the address is not the cell. - Bridge - an obtaining direct semantic relation between two exact
SchemeSenseCellvalues from different semantic contexts under one exact profile. - BridgePredicateProfile - the F.9-local by-value declaration of the direct relation's kind, symmetry or orientation, endpoint readings, correspondence or difference condition, applicability and as-of basis, Boolean truth condition, and stop dependencies.
- Bounded-use claim - an ordinary C.2.1 claim that says whether one exact obtaining Bridge is suitable for one named use, direction, use-specific rule, and loss tolerance. This phrase is descriptive, not a new public kind name.
- Relation orientation - how the Bridge kind orders or symmetrically relates its endpoint slots; it is not a use licence.
- Use direction - the ordered
<UseSourceSenseCell, UseReceivingSenseCell>designated inside one bounded-use claim. - Observed semantic loss - a difference or counterexample found in evidence. It can bear on a bounded-use claim but is not the use's permitted-loss tolerance.
- Permitted-loss tolerance - the maximum named loss accepted by one proposed use; it is content of that use's C.2.1 claim.
- Bridge occurrence description - an independently constituted C.2.1 episteme whose exact EntityOfConcern is one already individuated Bridge occurrence. It describes; it neither makes the predicate true nor supplies occurrence identity.
- Bridge Card - optional claim-bearing packaging. A filled Card may itself be a Bridge description episteme when its C.2.1 triple concerns an actual occurrence; a candidate Card instead modally describes the admitted relation kind and proposed endpoints. The reusable Card layout, registry row, publication form, and carrier remain separate.
- CL (Congruence Level) - optional F.9-local shorthand for the strength of evidence about a stated correspondence. It is neither a participant nor a use threshold and never grants a use.
- senseFamily - the local meaning family used by Part F. A
senseFamilylabel is not a durable U-kind.
Bridge kinds
A Bridge kind classifies the direct semantic relation tested by a profile. It says what correspondence or difference obtains; it does not settle any proposed use.
Same-family relation kinds
- Equivalence - the endpoint senses have the same extension and relevant intension under the stated relation condition. The relation is symmetric. A later use still names its direction, rule, and tolerance.
- Narrower-than - the source sense is properly included in the receiving sense. The relation is asymmetric.
- Broader-than - the source sense properly includes the receiving sense. The relation is asymmetric.
- Partial-overlap - the senses have a non-empty intersection, while each has cases excluded by the other. The relation is symmetric.
- Disjoint - the senses have no common admissible case under the stated readings. The relation is symmetric.
For inclusion, a narrower-to-broader proposed use is usually easier to justify than the reverse, but neither direction follows from the relation alone. State and warrant the bounded-use claim for a proposed broader-to-narrower use separately. If it needs narrower endpoint meanings to succeed, refine the cells and test the resulting Bridge. A reverse use can also be ruled out on the existing Bridge.
Cross-family relation kinds
These kinds state semantic correspondence across different senseFamily readings. They explain a connection; they do not create substitution, evidence authority, policy force, or a receiving occurrence.
- Design-spec-to-run-occurrence - a design sense corresponds to a run-time occurrence sense while remaining different in temporal and realization status.
- Measurement-evidence-for - a measurement sense corresponds to the measured aspect of another sense. The kind is semantic; actual evidential support remains with A.10 or B.3.
- Policy-constraint-on - a policy or deontic sense corresponds to a constrained behavioral sense. Actual obligation, permission, or authority remains with the policy or deontic governor.
- Viewpoint-correspondence - a sense used in one view corresponds to a sense used in another view over an EntityOfConcern. View, description, publication, and source-use claims keep their subject patterns.
Evidence about relation and use
Evidence must answer the question it bears on:
CL may summarize evidence strength for a stated correspondence: 0 contradicted, 1 weakly comparable, 2 bounded support with explicit counterexamples, and 3 matched stated invariants with no current material counterexample. It is optional and never serves as a use threshold.
A CL=3 label does not make a type-structure use suitable; the separate claim must still name the rule and tolerance, and reliance must still pass under A.10 or B.3.
Observed losses, unit differences, counterexamples, and invariant checks can supply grounds for a relation or bounded-use claim. Use an A.10 path when provenance or reliance must remain replayable, including the ordinary reliance required by §4.2; use a Card when reusable packaging is warranted under §8.
The proposed use's permitted-loss tolerance belongs in its ClaimGraph. A loss observation can change without reidentifying either the fixed Bridge or the bounded-use claim; it may instead reopen the claim's polarity or the current reliance disposition.
Bridge occurrence, description, Card, and publication
Recover and, when needed, individuate the direct relation before describing it. A Bridge may obtain without any assertion, description, Card, registry row, or publication.
A Bridge occurrence description is constituted independently under C.2.1 from its complete interpreted claim content, the already individuated occurrence as EntityOfConcern, and its effective interpretation rules (ReferenceScheme). Its content and rules satisfy the C.2.1:4.1.2 ranges; each endpoint cell retains its own scheme and exact local sense.
A proposal may instead be a modal C.2.1 episteme whose EntityOfConcern is the admitted direct Bridge relation kind and whose ClaimGraph designates proposed endpoints and profile; it supplies no positive occurrence reference and makes no relation obtain.
Use a Bridge Card only when durable reuse, delayed handoff, evidence review, audit, publication, or costly reversal makes reusable packaging worthwhile. A particular filled Card's claim-bearing account can be the description episteme when its C.2.1 identity supports that exact use. Identify its particular expression separately from the reusable layout rules. PublicationExpressionRelation states which claim content can be recovered from the expression for the bounded use, and PublicationFormConformanceRelation states whether it follows the selected Card form. That expression witnesses the form-mediated PublicationFormExpressionRelation; use the same expression for any carrier-bearing claim relied on for availability.
When availability matters, publish one selected description/Card edition through E.24.PUB: its EpistemePublicationRelation occurrence, particular expression, reusable form, and independently identified physical or digital carrier remain distinct from the selected episteme and from the Bridge.
For ClaimMode: actual, the description/Card episteme's exact EntityOfConcern is the already individuated Bridge occurrence.
In actual mode, the description/Card may package the Bridge assertion, one or more bounded-use propositions, their evidence and polarity, the descriptive A.10 evidence-provenance path, the independently established direct relations it cites, and local disposition, or the exact B.3 AssuranceResult when an actual named assurance claim is current, plus currentness and nearest non-use. Its C.2.1 identity is not the occurrence identity.
For ClaimMode: candidate or negative, no positive occurrence reference exists. The modal description/Card episteme's EntityOfConcern is the admitted F.9 direct Bridge relation kind; its ClaimGraph designates the proposed endpoints and profile. candidate says the proposed Bridge may obtain; negative says its predicate does not obtain. Any bounded-use proposition in the same graph keeps its own polarity. Completing, approving, registering, or publishing the description/Card creates no Bridge.
The exact <ClaimGraph, EntityOfConcern, effective ReferenceScheme> triple identifies each description/Card episteme. A changed description or Card edition, evidence path, reliance disposition, B.3 AssuranceResult, registry record, E.24.PUB publication occurrence, publication form, carrier, or layout does not reidentify a fixed Bridge. Publish only the selected description/Card edition needed by the named audience and bounded use; publication changes availability, not relation truth.
Boundary to coarsening and quantum-like export
Open F.9 when a receiving use needs an actual semantic relation between exact local senses. A lossy or approximate export is not thereby a Bridge, and an actual Bridge is not thereby a quantum-like state transition.
For a question that includes both a Bridge and its use, the use determines which correspondence to investigate. Establish the independent grounds in this order.
- resolve the exact F.17 cells, state the relation-semantic profile, and test whether the Bridge obtains;
- state the proposed use separately as
<u,d,r,t>and give the C.2.1 claim its polarity; - recover the descriptive A.10 evidence-provenance path, the independently established direct relations it cites, and local disposition or, when an actual named assurance claim is current, its B.3
AssuranceResultfor the same use; - if the use happened, identify the actual governed object and apply its subject pattern;
- add a Bridge Card only if durable packaging pays;
- open A.6.3.CSC, C.26.1, or C.26.2 only when coarsening, probe effects, or failure of any faithful-enough report is the live question.
When a state, metric, option, causal reading, or viability claim crosses the semantic boundary, its subject pattern states what survives and what is lost. The Bridge supplies only the semantic-correspondence premise; the bounded-use claim supplies only the named suitability proposition.
Invariants
- Exact endpoints first. A Bridge has exactly two F.17
SchemeSenseCellparticipants. - No context object. Semantic context is recovered from endpoint content and is not a relation participant.
- Different context is not enough. Different projections trigger the question but do not establish the relation.
- Profile contains relation semantics only. The proposed use, its direction, its correspondence rule, loss tolerance, polarity, reliance, authorization, and receiving objects are absent from profile identity.
- Obtaining before occurrence reference. A positive Bridge reference appears only after the fixed predicate is true and its dependencies are present.
- Use claim is separate. Every proposed use names
u,d,r,t, and polarity in a C.2.1 claim about the exact Bridge. - Reliance is separate. A.10 says whether ordinary evidence supports relying on the bounded-use claim. When an actual named assurance claim is current, B.3 supplies its bounded
AssuranceResult. Neither answer comes from F.9 or the Card. - Proposed use is not an occurrence. The
udesignation in the ClaimGraph names the proposed use; any actual Work, assertion, publication, relation, or operation application keeps its own identity; apply the relevant pattern to each claim about it. - Card separation. Card identity, completion, approval, registration, and publication neither make the relation obtain nor make the use happen.
- Loss separation. Observed semantic loss is evidence; permitted loss is tolerance inside the bounded-use claim.
- No authorization by implication. Semantic suitability, evidence reliance, and assurance are not legal, policy, or deontic permission.
- No silent inverse or composition. An inverse asymmetric relation and any direct A-to-C relation are tested independently.
- Two-SlotSpec declaration. The reusable RelationSignature declares only source and receiving SenseCell participant meanings;
CL, Loss Notes, scope/admitted use, evidence, counterexamples, policy, time, model-use structure, description, publication, and registry values remain qualifiers or neighbors. - Recurrence and identity. The non-optional identity rule uses the canonical exact endpoints and exact profile; one fixed tuple/profile is non-recurrent, and a changed applicability/as-of basis changes the profile before another candidate is admitted.
- Description and publication separation. A Bridge description/Card is independently constituted under C.2.1, and E.24.PUB independently governs any selected edition's publication occurrence, form, and carrier. None establishes relation truth or identity.
Micro-examples
The labels below are readable aliases. An actual case resolves exact F.17 cells and tests one profile before stating a proposed use.
- Participant versus Agent. A
Partial-overlapBridge may obtain between the exact BPMN and PROV senses. A separate claim may affirm use of the label "actor" in one orientation table under a rule that preserves the stated participation distinction. That claim creates no local system-role kind or assignment occurrence. - Process design versus Activity occurrence. A
Design-spec-to-run-occurrenceBridge may explain the semantic connection. A separate claim can bound an explanatory use; it does not identify a run occurrence from a design artefact. - Observation versus SLO fulfilment. A
Measurement-evidence-forBridge can relate the exact senses. A separate claim asks whether the observation sense is suitable for interpreting one named SLO comparison. A.10 handles ordinary evidence reliance; when an actual named assurance claim is current, B.3 supplies its bounded result for that use. - Subtype across OWL and a curated taxonomy. An
EquivalenceBridge obtains only under a profile whose relation condition includes the required class-level invariants. A separate claim asks whether one exact type-structure row may use that Bridge under its stated rule and zero material-loss tolerance. - Accuracy in metrology versus data quality. A
Partial-overlapBridge can make the shared word intelligible. A bounded-use claim may affirm that the label is suitable in one explanatory table while rejecting transfer of measurement methods or values.
Worked example and application sketches
Two display glossaries
In this constructed case, display scheme A has the closed value domain {0, 1, 2} and uses high exactly for value 2. Display scheme B has the same closed domain and uses high exactly for values 1 and 2. These sentences are the complete glossary definitions for this case; no claim about a real measurement scale or acceptance status is made.
Cell A is <scheme A by value, high, value = 2>; cell B is <scheme B by value, high, value = 1 or value = 2>. The different sense claims give different semantic-context projections. Each glossary definition directly supports its cell's exact claim under that scheme, so its F.17 LocalSenseBasisRelation can be cited separately.
Profile P declares Narrower-than, oriented A → B. Its relation condition is proper inclusion of the admitted values; it applies only to the two complete definitions printed above. The Boolean test is {2} ⊂ {1, 2}: 0 belongs to neither set, 1 only to B's set, and 2 to both. The closed domain and both membership rules are its required dependencies; an absent or unresolved rule stops the test. All dependencies are present and the test is true, so the exact Bridge b obtains for these cells and P.
Now propose two uses. For the first, u is to explain that every A-high value is B-high; d is A → B; r is membership implication for each value in the closed domain; t permits no counterexample. The C.2.1 claim about b is affirmative.
For the second use, u is to replace B-high by A-high in the same explanation; d is B → A; r again requires membership implication for every value; t again permits no counterexample. Value 1 defeats that rule, so this claim is negative while b remains unchanged.
For reliance on this bounded explanation, the A.10 descriptive account cites the two exact glossary definitions, their independently established cell-basis relations, and the complete three-value comparison. Those inputs supply the evidence required by these membership rules; pass is limited to this explanation of the printed definitions. No actual named assurance claim is current.
The result for this example is one obtaining Bridge and two separately warranted use claims, not a published glossary, an operational classification or performed Work. No Card is needed for this one-case explanation.
Sections 12.1–12.4 are application sketches. Their external source labels and profile placeholders must be replaced by exact cells, rules, dependencies and evidence before an actual positive Bridge or passing reliance is claimed.
Service target and monitoring observation
A service team resolves two exact cells: the ITIL sense of an availability target and the SOSA sense of an availability observation. Profile P-SLO-OBS-v2 states a Measurement-evidence-for semantic relation: the observation sense concerns a measured availability quantity relevant to the target sense, while observation and target remain different kinds of claim. The profile names the endpoint readings, direction of the semantic relation, applicability to the cited editions, Boolean condition, and required quantity-definition dependency. Current meanings satisfy it, so Bridge b-slo-obs obtains.
The team next proposes use u-slo-check: compare one observation result with the target. Direction d-slo is observation-to-target; rule r-slo requires the same quantity kind, aligned windows, and the stated unit conversion; tolerance t-slo permits the named rounding loss but no quantity-kind change. A C.2.1 claim with EntityOfConcern b-slo-obs states affirmative polarity for <u-slo-check,d-slo,r-slo,t-slo>.
Because this is an ordinary bounded evidence use and no assurance claim is made, the team recovers the descriptive A.10 evidence-provenance path for the observation record, checks the independently established direct relations it cites, and states RelianceDisposition=pass only for u-slo-check. The team may therefore rely on the comparison claim within that boundary. The disposition does not establish SLO fulfilment, authorize acceptance, or prove that comparison Work occurred.
Behavioral participant and access role
An exact Partial-overlap Bridge obtains between a BPMN participant sense and a named RBAC role sense when the profile's overlap and difference conditions are satisfied. A separate bounded-use claim proposes the label "actor" for one glossary row, in the stated direction, under a rule that preserves assignment moment, enforcement locus, multiplicity, and accountability differences, with zero tolerance for reading the label as a local system-role kind or assignment occurrence. Current evidence can support that label use under A.10.
When a later claim uses the RBAC source word role, apply E.10.ROLE and first say whether it concerns access, permission, authority, a work-facing classification, an assignment, or performed Work. For access, permission, or authority, use the direct pattern for that relation. Use A.2.8.PER for granted permission while keeping actual access separate. If access wording still hides the subject or relation, use A.6.P:4.11a; if the participants and predicate are clear but no direct pattern defines the relation, return A.6.RCD missing-governor[direct access relation].
A work-facing classification requires an admitted System, one exact local system-role kind with its KindSignature, and the C.3.2 classification judgment under A.2 and C.3. Use F.4 only when the receiving use needs a SystemRoleKindDescription episteme, and F.5 only when it needs a durable designation.
An assignment claim separately identifies an occurrence of a directly declared species under U.SystemRoleAssignment through A.2.1.
If performed Work is also claimed, recover every exact actual performer through A.13 and use A.15.1 to identify the dated Work, exact Method, time, and containing System independently.
Use F.6 only when the Bridge account or receiving use explicitly represents precise assignment-bound attribution through the same obtaining assignment already established by A.13, and the direct case fact links that exact Work-assignment pair. Missing or failed F.6 alone leaves independently admitted Work intact. If the same finding invalidates a premise consumed by A.13/A.15.1 admission, recheck that premise and its dependent conclusions under F.6:4.2. The Bridge, bounded-use claim, and reliance result establish none of these facts.
Subtype notions in one structural row
The endpoint senses are OWL2:SubClassOf under a cited OWL profile and curated-taxonomy is-a under one named taxonomy edition. To test Equivalence, align the domain and class interpretation, compare which inclusions each sense admits, and test the relevant intensional criteria in §6.1. If the profile also requires acyclicity and antisymmetry, check both. When the full equivalence test passes and its dependencies are current, the Bridge obtains.
A second premise is still required. The C.2.1 claim names the proposed type-structure row, its source-to-receiving direction, the rule that preserves the three invariants, and zero material-loss tolerance. Only an affirmative current claim with passing A.10 reliance, or, when an actual named assurance claim is current, a B.3 AssuranceResult for the same use with disposition=supported-for-use, supports relying on the row.
A contradicted relation invariant makes the Bridge predicate false; a use-specific tolerance failure can instead make the bounded-use claim negative while the Bridge remains unchanged.
Setpoint versus service target
CTRL:setpoint and ITIL:target can both be described as a "target", but usually have only Partial-overlap or are Disjoint under the exact readings. A proposed substitution in a control calculation receives a negative bounded-use claim because its rule and tolerance cannot preserve the physical-reference meaning. A didactic comparison may receive a different affirmative claim. Neither claim changes which Bridge obtains.
Common Anti-Patterns and How to Avoid Them
Reasoning primitives
These conceptual judgements do not supply work-enactment, card-completion, registry, publication, permission, or authorization rules.
Direct Bridge occurrence
The proposed use, its direction, its rule and tolerated loss, the claim's polarity, the reliance result and the Card are outside P. An asymmetric relation retains its endpoint orientation in P.
Bounded-use proposition
Changing u, d, r, or t changes C; it does not change b. Affirmative polarity is not evidence reliance, assurance, authorization, or occurrence.
Ordinary A.10 reliance
A non-passing or narrower disposition supplies no support for the attempted use. The disposition is a local A.10 classification statement, not a new result kind.
B.3 assurance branch
A narrowed disposition supports only its stated narrower use. abstain, evidence-needed, reopen, or blocked stops the attempted use. If no assurance claim is current, do not open B.3. A consequence, display, or local threshold creates no assurance claim.
Receiving occurrence stays separate
A claim about an actual receiving object needs the participants or arguments, obtaining or performance facts, and identity supplied by its subject pattern.
Direction guard
Relation symmetry or orientation does not select d. Each proposed direction receives its own bounded-use claim. For inclusion, a broader-to-narrower use needs its own grounds. If it requires refined meanings, test their Bridge separately. A reverse use can be rejected while the original Bridge stands.
Chained-use guard
For a proposed use through both correspondences, follow the result of the first into the rule for the second. Judge the end result against the receiving tolerance over the admissible cases. Local losses may reinforce or offset one another. Where that distinction matters, preserve their dependence or use a justified bound for the combined loss. Cite each Bridge used and establish current reliance on this composite claim. If a direct A-to-C correspondence is needed, test it independently.
Candidate-card guard
The card concerns the admitted direct Bridge relation kind and places proposed endpoints, profile, ClaimMode, and polarity in its ClaimGraph. It creates neither relation nor receiving occurrence.
Relations
Builds on: F.17, F.18, C.2.1, F.0.1, F.7, and F.8.
Coordinates with:
- A.10. Use it for the descriptive evidence-provenance path, the independently established direct relations it cites, and local
RelianceDispositionfor ordinary bounded evidence use. - B.3. Use B.3 only after an actual named assurance claim is current; it states the bounded
AssuranceResultor non-positive disposition and does not create the claim, authorization, or use. - E.10.ROLE, A.2, C.3, F.4, F.5, A.13, A.15.1, A.2.1, and F.6. When source wording leaves role ambiguous, use E.10.ROLE first. Use A.2 and C.3 for the local system-role kind and any separate System-classification judgment. Use F.4 only when a description of that kind is current, and F.5 only when its durable naming is current. Recover each exact actual performer through A.13 and admit dated Work independently through A.15.1. A.2.1 governs the assignment species and obtaining occurrence already required by the A.13 core. Add F.6 only when the receiving Bridge use explicitly relies on precise assignment-bound attribution through that same assignment. A Bridge establishes none of these facts.
- F.8. A mint-or-reuse decision may consume an obtaining Bridge plus a separately warranted bounded-use claim; it does not strengthen either.
- A.2.6. Scope translation may use an obtaining Bridge only together with an affirmative claim naming the exact direction, scope-correspondence rule, and loss tolerance. Use A.2.6 for the translated scope and membership.
- A.6.1. Use it to identify any actual operation application. The
udesignation in a Bridge claim names a proposed use and is not an application binding. - A.6.5. Relation-position labels and SlotSpec claims remain governed by slot discipline.
- C.29. Mathematical-lens use may cite a Bridge and bounded-use claim; C.29 still governs its mathematical object, preserved and lost structure, and actual lens use.
- C.34. Structural correspondence or morphism adequacy may cite an obtaining Bridge and a bounded-use claim but states its own preserved and lost architecture structure.
- A.6.REL. Applies the F.9 recurrence and occurrence-identity rule only when a receiver must distinguish or reference the occurrence.
- C.2.1. Independently constitutes assertions, modal proposals, occurrence-description epistemes, and filled Cards; none supplies the Bridge predicate or occurrence identity.
- E.24.PUB. Use it for any
EpistemePublicationRelationoccurrence, particular expression, any selected reusable publication form, and actual presentation carrier for a selected description/Card edition. Publishing creates neither Bridge nor receiving use. - A.6.3.CSC, C.26.1, and C.26.2. Govern coarsening, probe effects, and no-faithful-enough-report cases when those questions are live.
Revision law
- Endpoint change. A changed by-value scheme, local expression, or local-sense claim identifies another F.17 cell and requires another Bridge test.
- Profile change. A changed kind, symmetry or orientation, endpoint reading, relation-specific correspondence or difference condition, applicability or as-of basis, Boolean truth condition, or stop dependency identifies another profile and occurrence candidate.
- Use-content change. A changed proposed use
u, direction, use-specific rule, or permitted-loss tolerance identifies another C.2.1 claim while the fixed Bridge remains unchanged. - Polarity change. Affirmative versus negative is changed claim content; it is not a changed reliance disposition.
- Evidence or reliance change. A changed evidence item, path, currentness window, independently established direct relation cited by the A.10 path, local
RelianceDisposition, or B.3AssuranceResultreopens reliance without reidentifying the fixed Bridge or fixed C.2.1 claim. - Obtaining change. New endpoint facts may establish, refute, or leave unresolved the predicate for a fixed occurrence candidate without silently changing its identity.
- Description, Card, registry, or publication change. Apply C.2.1 to description/Card identity and E.24.PUB to publication occurrence, form, and carrier; none creates, removes, reidentifies, or recurs the Bridge.
- Receiving occurrence change. Reidentify or revise the Work, assertion, publication, relation, application, or other receiving object under its subject pattern.
Acceptance tests
Static conformance
- SCR-F9-S01 (Well-typed direct relation). Each actual Bridge has exactly two resolved F.17
SchemeSenseCellendpoints and one exact relation-semantic profile. - SCR-F9-S02 (Different semantic contexts). The endpoint
<ReferenceScheme, LocalSenseClaim>projections differ; same-context aliases stay with designation resolution. - SCR-F9-S03 (Profile boundary). The profile contains only kind, symmetry or orientation, endpoint readings, relation condition, applicability and as-of basis, Boolean truth condition, and stop dependencies.
- SCR-F9-S04 (Obtaining). Current endpoint facts satisfy the exact profile and all required dependencies are present. Scheme difference, spelling, implementation, evidence score, card, registry, or publication alone fails this test.
- SCR-F9-S05 (Separate bounded use). Every use claim identifies exact Bridge
b, namesu,d,r,t, polarity, and an effective ReferenceScheme under C.2.1. - SCR-F9-S06 (Reliance branch). The same bounded use has a descriptive A.10 evidence-provenance path citing independently established direct relations plus a passing local disposition or, when an actual named assurance claim is current, its exact B.3
AssuranceResult; onlysupported-for-usesupports the attempted assurance use, whilenarrowedsupports only its stated narrower use. - SCR-F9-S07 (No authorization overread). Semantic fit, A.10 reliance, and B.3 assurance are not described as legal, policy, or deontic permission.
- SCR-F9-S08 (Receiving-object boundary). A named proposed use is never treated as performed Work, assertion, publication, relation, or operation application.
- SCR-F9-S09 (Card truthfulness). An actual card concerns an already individuated occurrence; a candidate or negative card concerns the admitted relation kind and has no positive occurrence ref.
- SCR-F9-S10 (Plain action). A practitioner can tell what relation to test, what use is proposed, what would stop reliance, and which downstream claim still needs an applicable pattern.
- SCR-F9-S11 (Non-optional identity and recurrence). The declaration states
BridgeOccurrenceIdentityRule, asymmetric ordering or symmetric canonicalization, and the non-recurrence of one fixed endpoint/profile tuple; a later basis changes the profile before another candidate is admitted. - SCR-F9-S12 (Description and publication boundary). Every actual description/Card concerns an already individuated occurrence under C.2.1; every modal proposal has no positive occurrence ref; E.24.PUB publication, form, carrier, and registry identity establish neither.
- SCR-F9-S13 (No adjacent fact by Bridge). No Bridge creates a local system-role kind or assignment, Work, evidence authority, status transfer, U-kind admission, publication, model-use crossing, or another subject relation.
Regression checks
- RSCR-F9-E01 (Same Bridge, changed use). Reversing direction, changing the use rule, or changing tolerance reidentifies the C.2.1 claim, not the Bridge.
- RSCR-F9-E02 (Same claim, changed evidence). Stale or stronger evidence changes the descriptive A.10 evidence-provenance path or local disposition, or the B.3 branch, without reidentifying the fixed claim.
- RSCR-F9-E03 (Required but missing assurance claim). If a direct domain rule requires an assurance claim and none is current, return
RelianceDisposition=assurance-neededor block the use. Do not manufacture a positive claim or a generic safety-case record. - RSCR-F9-E04 (Profile change). A changed relation condition or endpoint reading identifies another profile and occurrence candidate.
- RSCR-F9-E05 (Packaging change). A changed card, registry entry, publication, form, or carrier leaves the Bridge and fixed bounded-use claim unchanged unless their own discriminators changed.
- RSCR-F9-E06 (Positive proposal versus occurrence). An affirmative claim with passing reliance proves no comparison Work, assertion, publication, direct relation, or operation application.
- RSCR-F9-E07 (Polarity versus reliance). Negative claim polarity and a non-passing reliance disposition remain different facts.
- RSCR-F9-E08 (Reliance versus authorization). A.10
passor B.3supported-for-usedoes not imply permission. - RSCR-F9-E09 (No inverse or composition). Neither an asymmetric inverse nor a direct A-to-C Bridge follows without its own profile and obtaining test.
Didactic distillation
Choose the inquiry from the recipient’s question, then use the steps that it requires.
- What conclusion depends on a correspondence between meanings? Select the two exact senses that bear on it.
- State the proposed correspondence or difference and test it. Report whether the Bridge obtains, is ruled out or remains unresolved.
- If the recipient proposes using the Bridge, state the action, direction, rule and tolerated loss that still allow their conclusion. Give the judgement its supported polarity.
- Check whether current evidence or assurance supports relying on that claim; recover authorization separately if needed.
- If the use happened, identify the actual object and apply the relevant pattern to the claim about it. Make a card only when reuse is worth the maintenance.
A question about two meanings can finish with their relation. For a proposed use, follow what the correspondence preserves to the recipient’s conclusion. An account of what actually happened requires evidence about that occurrence.
Archetypal Grounding
Tell
A Bridge is an actual semantic relation, not a synonym claim or enactment edge. Its profile says what correspondence holds. A separate claim says whether that relation suits one proposed use. Evidence, authorization, packaging, and actual performance remain separate.
Show: service lane
The observation sense can bear a semantic relation to the target sense without being the target status. The team then states and warrants one comparison use; status and acceptance remain separate claims.
Show: access-control lane
A process team and an access-control team both use operator. An obtaining overlap Bridge plus an affirmative, warranted label-use claim can support one glossary row. It establishes no access, permission, authority, local system-role kind, assignment, performer, or Work. Recover whichever fact is claimed under its direct pattern; preserve the source term access-control role only when referring to the external scheme.
Show: episteme lane
An actual card can package the relation claim, bounded-use claim, evidence path, and reliance disposition. It remains an episteme about the Bridge and creates neither the relation nor the receiving act.
Bias-Annotation
- Governance bias. A separate use claim and reliance check add work. Keep them in ordinary sentences when durable packaging has no payoff.
- Architecture bias. Typed relation profiles can look heavier than synonym prose. The five-part script begins with the practical action and introduces exact terms only where they stop a real overread.
- Ontology and episteme bias. F.9 resists global meaning claims and keeps claims separate from their subject. Explicit Bridges permit practical cross-context comparison.
- Pragmatic bias. Conservative separation can feel slower than reusing a mapping table. A fixed Bridge can support many independently stated uses without being reidentified.
- Didactic bias. An explanation using only internal nouns can obscure the action. Every use starts by saying what a person will do, what rule they will follow, and what result would make them stop.
Conformance Checklist
A positive Bridge claim conforms if and only if it meets the applicable conditions below. Apply the further conditions for a use, assurance, Card or publication when it is proposed. When no positive Bridge can be asserted, give the applicable outcome in §4.1.
- both endpoints resolve to exact F.17
SchemeSenseCellvalues; - their semantic-context projections differ;
- the Bridge has exactly two participants and one relation-semantic profile;
- the profile contains no receiving-use or reliance content;
- the profile applies, its Boolean predicate is true, and its dependencies are present before a positive occurrence is cited;
- every proposed use is a separate C.2.1 claim naming
u,d,r,t, polarity, and effective scheme; - observed loss stays in evidence while permitted loss stays in the bounded-use claim;
- current ordinary reliance uses the A.10 branch for the same bounded use; when an actual named assurance claim is current, use its B.3
AssuranceResultfor that use; - no reliance or assurance statement is read as authorization;
- any actual receiving object is recovered under its subject pattern;
- description episteme, Card, registry record, E.24.PUB publication occurrence, form, and carrier remain distinct from Bridge occurrence and receiving-use occurrence;
- inverse and composed relations are tested independently;
- the reusable RelationSignature declares only two endpoint SlotSpecs, while every
CL, Loss Note, scope/admitted-use, evidence, counterexample, policy, time, model-use, description, publication, or registry value remains a qualifier or neighbor; - the non-optional occurrence identity and non-recurrence rule is stated and applied before a Bridge occurrence is referenced; and
- same-context designation remains outside F.9; claim-bearing source wording with role first uses E.10.ROLE, and each recovered system-role kind, assignment, access, permission, authority, Work, evidence-authority, status, U-kind, publication, structure-crossing, or other subject claim returns to its direct pattern.
Consequences
Benefits. F.9 permits comparison, translation, and bounded reuse without collapsing local senses. One stable Bridge can support several differently directed or differently tolerant use claims, and evidence can change without silently changing relation identity.
Costs. A reader must state two premises instead of one: the semantic relation and the bounded-use proposition. Ordinary evidence reliance may require A.10 work. An actual named assurance claim additionally requires a B.3 result. This cost is paid only when a real cross-context use is proposed; a Card remains optional.
Failure mode avoided. A Bridge, score, or card can no longer act as a quiet substitute for a local system-role kind or assignment, status transfer, evidence authority, authorization, publication, or performed-work attribution.
Rationale
Cross-context comparison is unavoidable, but the truth of a semantic relation and the suitability of one action are different claims. Putting a proposed use's direction, rule and tolerated loss into BridgePredicateProfile would reidentify the relation whenever the proposed use changed. Putting them in a separate C.2.1 claim lets one Bridge remain fixed while several uses are affirmed, rejected, narrowed, or reopened independently.
A.10 or B.3 can reopen reliance without erasing the relation. A card can travel without becoming the relation. A proposed use can be warranted without being authorized or performed. These boundaries preserve practical reuse and make each failure local and repairable.
SoTA-Echoing
Bridge Card publication discipline
Minimal truthful card
A reusable description/Card states its mode and exact C.2.1 identity. An actual one names its already individuated Bridge; a candidate or negative one names the admitted direct Bridge relation kind and modally designates proposed endpoints, profile, and polarity in its ClaimGraph.
When the description or Card packages a proposed use, it states u, d, r, t, polarity, observed loss, evidence, currentness, nearest non-use, and the exact A.10 disposition or, when an actual named assurance claim is current, its B.3 AssuranceResult. Missing relation facts are never repaired by filling more fields.
If availability matters, E.24.PUB publishes the selected episteme edition for one declared audience and bounded use through its independently governed publication occurrence, form, and carrier.
One occurrence, several claims and descriptions
Several bounded-use claims and description/Card epistemes may concern the same actual Bridge. Their C.2.1 identities differ when their ClaimGraph, EntityOfConcern, or effective scheme differs; the Bridge identity does not. Any review Work or publication occurrence is identified separately under its subject pattern. Prefer one primary current card only when it reduces navigation cost.
Revision without silent ontology change
If evidence, observed loss, reliance, assurance, wording, or publication changes while endpoints and profile remain fixed, revise the corresponding claim, evidence relation, disposition, card, or publication. Test another Bridge only when an endpoint or relation-semantic profile component changes.
Bundle and endpoint interaction
Descriptions and displays may cite Bridges and bounded-use claims; their governed objects and rules of use remain separate.
When a quality-family claim crosses contexts, observed loss may bear on its bounded-use claim and on B.3 assurance, but neither fact retypes the quality family. An F.9.1 stance note may help readers interpret the claim; it remains a separate episteme and cannot widen the relation, proposed use, reliance, authorization, or occurrence.
C.29 mathematical-lens use relation
When a mathematical-lens use relies on cross-local meaning, first recover an actual F.9 Bridge. Then state the exact bounded-use claim for the lens direction, correspondence rule, and tolerated loss, and recover current reliance. Use C.29 for the mathematical object, LensMappingMode, preserved and lost structure, lens-use judgement, and actual lens use. A Bridge can make a lens interpretable without making any lens use occur.
Review matrix
A reader can test bridge integrity with eight questions:
- Do both endpoint refs resolve exact F.17
SchemeSenseCellvalues from different semantic contexts? - Does the profile say only which semantic relation holds, with its endpoint readings, condition, applicability, truth rule, and stop dependencies?
- Is the Bridge claimed only after that fixed predicate is true?
- Does each proposed use separately name the action, direction, correspondence rule, tolerated loss, and polarity?
- Does the same use have a current descriptive A.10 evidence-provenance path citing independently established direct relations and a local disposition or, when an actual named assurance claim is current, its B.3
AssuranceResult? - Are semantic suitability, reliance, assurance, and authorization kept distinct?
- If someone says the use happened, is the actual Work, assertion, publication, relation, operation application, or other object recovered under its own pattern?
- Does any card remain optional packaging rather than the source of relation truth, permission, or occurrence?
For wording such as same, equivalent, align or map, begin with what the speaker says matches and for what purpose. When the claim concerns meanings, recover the senses and test the proposed Bridge. When it concerns compatible designs, a performed action or another subject relation, establish that relation under its own conditions. Check a bounded use, its reliance and any required authorization when the recipient depends on them.
F.9:End
Bridge Stance Note — Clarify a Bounded Use
Type: Architectural (A) Status: Stable Normativity: Normative unless marked informative
Plain name. A short note that helps a reader interpret one use of a Bridge.
One-line summary. A bridge stance note is a separate claim-bearing episteme about one exact F.9 bounded-use claim. It may say that the use is best read as a local rename, operationalization, partial analogy, projection, or warning against equivalence. It changes neither the Bridge nor the claim it explains.
Use this when. Use F.9.1 after an F.9 Bridge has been shown to obtain and a separate bounded-use claim says what someone proposes to do with that Bridge, in which direction, under which rule, and with what tolerated loss. Add a stance note only when one short interpretive cue will help a reader without replacing those facts.
Start here when. You already have the Bridge and its bounded-use claim, but a reader may still overread a phrase such as “operationalizes”, “projection”, or “roughly analogous”.
First useful move. Point to the exact bounded-use claim and write one ordinary sentence: “For this named use, read the relation as ___.” Add a rejected overread only when it has an independent local ground and changes the intended reader's interpretation, as F.19 requires. Choose a stance word only if it makes the sentence shorter and clearer.
What goes wrong if missed. An interpretive gloss is treated as proof of equivalence, permission to substitute, or evidence that a use happened. Authors can also build a second bridge taxonomy where one warranted use needs a gloss.
What this buys. Readers get a compact caution or reading aid while the Bridge, proposed use, evidence, reliance, authorization, actual downstream act, and optional publication package remain separately checkable.
Not this pattern when. Use F.9 when the Bridge or suitability of a proposed use is still unsettled. Use A.6.3.CSC when a shortened or coarsened rendering needs a source tether, a narrower admissible use, blocked downstream uses, or a return trigger. Do not create a stance note merely because a comparison sounds informal.
Problem
People often need one short gloss after they have established a cross-local semantic relation and judged one bounded use of it. “Operationalizes”, “partial analogy”, and “projection” can be helpful, but each can also sound stronger than the underlying claim. The task is to make the intended reading explicit without making the gloss a Bridge kind, a use licence, or a substitute for evidence.
Object and identity
A bridge stance note is an ordinary C.2.1 episteme. It is neither a U-kind nor part of the Bridge occurrence.
Three values determine its C.2.1 identity:
- its
EntityOfConcernis one exact F.9 bounded-use claim; - its
ClaimGraphstates the selected stance, the ordinary-language reading that stance abbreviates, and any grounded overread it needs to reject; - its effective
ReferenceSchemesupplies the local meanings of the stance wording.
Changing the bounded-use claim, stance, material reading boundary, or effective scheme identifies another stance episteme. Rewording that preserves those values may be a new edition of the same claim-bearing account under the applicable C.2.1 edition rule. Changing how the note is published or found—for example, its Card, registry entry, file, or visual placement—does not reidentify it.
The underlying bounded-use claim remains primary. It names the obtaining Bridge, proposed use, direction, correspondence rule, tolerated loss, and polarity. The stance note describes how to read that claim; it supplies none of those values and cannot change them.
Forces
Solution
Recover the claim before writing the gloss
Start with the reader’s difficulty. What could they infer from the wording that would change their use of the claim? If the existing explanation already serves them, finish there. When a separate interpretive note would help, confirm the following.
- the two exact F.17 local senses are known;
- an F.9 Bridge between them obtains;
- one separate bounded-use claim identifies the proposed use, direction, rule, tolerated loss, and polarity; and
- the stance note has a named reader benefit that the claim does not already express plainly enough.
If an endpoint or the Bridge is missing, use F.9 when the original question needs that correspondence. If the question needs a judgement about its use, establish the bounded-use claim before writing a gloss for it.
Write one bounded reading
Write the note in this order:
- cite the bounded-use claim it explains;
- state one plain reading of that claim;
- name one primary stance when a reusable label helps;
- state the nearest overread only when F.19's grounded-contribution test requires the correction; and
- emphasize a material loss only when that emphasis changes how the reader should understand the claim.
The stance note may cite evidence or a current A.10 or B.3 reliance result, but it does not create either. If the proposed direction, rule, tolerated loss, polarity, or reliance result changes, repair that object under F.9, A.10, or B.3 rather than editing the stance as a proxy.
Starter stance vocabulary
These labels are optional local designations inside the stance episteme. They are not Bridge kinds, levels, scores, permissions, or a second taxonomy. If a plain sentence is clearer, use the sentence without minting a label.
A stance word carries no direction by itself. Direction belongs to the bounded-use claim it explains: a projection reading in one direction never establishes the reverse reading or use.
Keep Bridge, use, reliance, and packaging separate
- A practitioner uses F.9 to test and record whether the Bridge obtains.
- The bounded-use claim says whether that Bridge suits one named use.
- A practitioner uses A.10 or B.3 to judge whether current evidence or assurance supports relying on that claim.
- The stance note says how to read that one claim.
- A claim that a comparison, translation, publication, Work occurrence or other downstream act occurred uses that object's subject pattern and evidence. When authorization is needed, recover it under the pattern that defines the permission or authority claim. The stance note establishes neither.
A Bridge Card is optional claim-bearing packaging. A Card may package the Bridge description, bounded-use claim, evidence references, reliance result, and stance note together, but the Card is not a prerequisite for any of them.
The Card's layout and edition, and any publication occurrence or file carrying it, remain separate from the objects it brings together.
CL is optional F.9 shorthand for evidence strength about a stated correspondence. It is not a substitution threshold and does not rank the stance. A stance note establishes neither suitability nor permission from CL, a loss note, or the gloss itself.
Preserve source return and coarsening boundaries
A projection or nonEquivalent stance can highlight loss, but it does not provide the source tether, narrower admissible use, blocked downstream use, or return trigger required for controlled coarsening. When those duties are live, use A.6.3.CSC and keep the stance note only if it still helps explain an independently established bounded-use claim.
When checking the Bridge or bounded-use claim requires the source episteme or source publication, return to that source first. Publication cannot substitute for the missing source.
Archetypal Grounding
Local rename inside one bounded use
The Bridge obtains between the two local senses. The separate affirmative claim says that one operational label can serve as a near-renaming of the other for this glossary. A localRename stance can make that reading quick to recognize. It establishes neither cross-local identity nor unrestricted replacement outside the stated use.
Operational reading of a control cue
An F.9 Partial-overlap Bridge obtains between one exact operator-alarm sense and one broader control-cue sense. A separate affirmative bounded-use claim says that this Bridge is suitable for using the alarm term in one training explanation, in the alarm-to-control direction, under a rule that preserves the response trigger but not the complete control model, with zero tolerance for treating the term as implementation authority.
For this training use, a current A.10 reliance judgment says the available evidence supports relying on that narrow claim.
A stance note may now say: “For this training explanation, read the alarm term as operationalizing the response-trigger part of the control cue; do not read it as the complete control model or as permission to change the controller.” The stance is operationalizes.
Projection into a report
The Bridge relates the two local senses. The bounded-use claim says that this relation suits a weekly report showing one aspect of the operations concept. The claim names the report use, direction, correspondence rule, and tolerated omission. A projection stance note may foreground that the report keeps only queue age and drops causal and capacity distinctions.
If readers must return to the source account before making a staffing or release decision, follow A.6.3.CSC: return to the source and do not use the report for those decisions. The stance note alone cannot set that boundary.
Partial analogy without substitution
A teacher wants a reader to recognize a recurrence. In this constructed comparison, a speaker’s felt-sense phrase in a TAE session, “back where we started”, describes returning to the same unresolved question. A formal account describes a path that returns to its starting state. Suppose the two local readings and their correspondence have been established for this teaching use. The teacher can explain the analogy in one sentence. “The speaker comes back to the unresolved question just as the path returns to its starting state.” That feature supports the comparison. To infer that every permitted continuation repeats, the learner would also have to rule out a continuation that escapes the recurrence. The learner can use the analogy without borrowing either school’s method.
The nonEquivalent warning
An affirmative bounded-use claim says that the Bridge is suitable for placing two senses in the same comparison table. A separate negative claim rejects the proposed substitution use. A nonEquivalent note on the negative substitution claim can explain why the comparison heading gives no basis for replacement. The label does not make the Bridge Disjoint, set CL=0, negate another use, or prohibit a future use with different direction, rule, and tolerance.
Anti-case: the gloss comes first
“The dashboard metric is basically a projection of operational resilience.” First ask what the reader needs from the sentence. To choose staffing for the next shift, establish which demand the staff must cover and what capacity will be available. A gloss about the metric could leave both open. To explain what the displayed value preserves from a source account, establish the correspondence and what it permits the reader to infer. Apply A.6.3.CSC if a controlled reduction is needed. Add the stance only where a remaining ambiguity affects the reader.
Conformance checklist
- One exact F.9 Bridge obtains before the stance note is cited.
- One separately constituted bounded-use claim names the proposed use, direction, correspondence rule, tolerated loss, polarity, and effective scheme.
- The stance note is a separate C.2.1 episteme whose EntityOfConcern is that exact bounded-use claim.
- The note states a plain reading; it includes a rejected overread only when F.19's grounded-contribution test requires it. A stance token is optional.
- The stance changes neither Bridge kind nor Bridge identity, claim identity or polarity, reliance, authorization, nor any downstream occurrence.
- A Card is optional packaging, not a prerequisite or source of truth.
CL, observed loss, permitted loss, reliance, and stance remain distinct.- One primary stance is normally enough; additional cautions belong in plain boundary or loss wording.
- Reusing a stance label for another claim requires a new local justification; equal labels do not identify the claims or their losses.
- A coarsened rendering that needs source return, narrowed use, or blocked downstream use applies A.6.3.CSC.
Common Anti-Patterns and How to Avoid Them
Consequences
Benefits. A reader receives one concise interpretive cue without losing the exact relation, use, evidence, or action boundary. Cards remain optional, and the same Bridge may support several independently judged uses and stance notes.
Costs. Authors must identify the bounded-use claim before adding a stance. A card-first record may need a reference to the bounded-use claim and removal of a stance that supplies no reader benefit.
Limits. F.9.1 neither establishes a Bridge nor judges suitability, evidence, reliance, permission, performance, publication, or controlled coarsening. It explains one already constituted bounded-use claim.
Rationale
The stance vocabulary is useful because practitioners already write short interpretive glosses. The ontology stays small by treating each gloss as ordinary claim-bearing content rather than a new relation kind or universal classification. Making the bounded-use claim the EntityOfConcern also keeps a Card as optional publication packaging rather than the subject or source of the relation and use claims.
One primary stance per claim is a readability default, not a new cardinality law. The criterion is whether the note makes the claim easier to understand without hiding a materially different reading; when one ordinary sentence does that better, use the sentence.
Bias-Annotation
Favored reading. Prefer disciplined, claim-local comparative reading to sweeping synonym claims.
Do not turn harmless comparison into a Bridge, bounded-use claim, reliance judgment, and stance note when that machinery changes no action. Apply the four checks in 4.1; when item 4 has no reader benefit, stop.
SoTA-Echoing
Relations
- Builds on and coordinates with: F.9 for the obtaining Bridge and bounded-use claim; C.2.1 for the stance episteme and its identity; A.10 and B.3 for reliance; A.6.3.CSC for controlled coarsening and source return; E.17.ID.CR for bounded comparative review; C.16.Q for quality-term repair; E.24.PUB for publication; F.17 for local sense; and F.18 for naming when that question is live.
- Does not replace: the pattern that defines or constrains an actual comparison, translation, publication, Work, decision, permission, or other downstream object.
Migration rule
For each legacy stance passage, start with the work its reader needs to do.
- What would a short note let this reader interpret that the claim itself leaves unclear?
- Which two exact local senses does that reading depend on?
- Does the F.9 Bridge between those senses obtain?
- Which bounded-use claim would the note explain, and what does that claim affirm or rule out?
Keep a stance note when you can identify both senses, establish their Bridge, cite the bounded-use claim and explain what the note adds for its reader. Preserve useful legacy wording as the plain reading, stance, loss emphasis, example or grounded non-use boundary. If the note adds nothing for that reader, omit it. Retain any independently useful legacy explanation with the claim or example it explains. A Card reference may remain as optional publication access, never as the stance's identity or prerequisite.
F.9.1:End
Apply and Interpret Evidence, Standard and Requirement Statuses
Type: Boundary and relation-use pattern Status: Stable Normativity: Normative
Problem frame
Use this when. Use F.10 when a receiving use depends on a status word and its exact value, governed target, scope, window, source, rule, or use is still implicit.
F.10 is especially useful when evidence, standard and requirement statuses are mixed: a dashboard says a service is ready, a standard says a method is approved, a measurement is cited as requirement satisfaction, a model card says a model is validated, or a requirement register says a clause is waived.
Primary EntityOfConcern. One status-use relation connects an admitted bearer and an admitted target, which may be the same governed object, with one local status value, one ClaimScope/use scope, one validity window and one intended receiving use. F.10 does not define or create the target and does not turn a display, source, list membership, approval act, evaluation rule, result, or evidence item into the status-use relation.
First useful move. Recover the exact target and any direct domain result on which this status depends. Then name the status-value SchemeSenseCell and family under the effective ReferenceScheme, status scope/window, exact source and provenance/currentness constraints, intended use, and stronger use not carried.
Recover the rule application and result when they are used. Identify dated evaluation Work separately only when that history is current and independently admitted under A.13/A.15.1.
What goes wrong if missed. One compact word does the work of domain result, evidence standing, standard approval, requirement satisfaction, gate passage, release readiness, permission, and assurance at once. A dashboard list or traffic-light cell is treated as actual status use. An F.9 Bridge or family edge is treated as the explanation or evaluation rule. Design approval becomes runtime satisfaction.
What this buys. Status words remain local, typed, comparable, and usable without hiding the target or the work that justified the status. Evidence status says only what evidential standing is being asserted for a claim; standard status says only what a named governing source sanctions; requirement status says what its rule warrants about an exact clause, including applicability, a completed evaluation, a waiver or a pending evaluation. Cross-local vocabulary and cross-modality interpretation remain explicit and loss-aware.
Not this pattern when. Use the subject's direct pattern for its target and domain result; A.2.4 for first evidence/status-use classification; A.10/G.6 for source recovery, provenance, and bounded reliance; G.11 for currentness; B.3 for assurance; C.28 for causal use; A.21 for a gate; the direct permission, commitment, requirement, standard, acceptance, release, or decision pattern for those results; E.17/E.24.PUB for publication; and A.15.1/A.6.1 for performed evaluation work and actual bindings.
Problem
A compact status label is useful and easy to misread. The same label often hides different objects and claims:
- Modality collapse. Validated is read as evidence standing, standard approval, requirement satisfaction, and release permission at once.
- Target collapse. The status does not say whether it concerns a claim, quantity, method description, standard edition, clause, system-role assignment, work result, publication, gate record, or another exact target.
- Result collapse. A measurement, proof, conformance verdict, requirement-evaluation result, or assurance result is renamed as a generic status instead of retained under its direct governor.
- Window and scheme loss. Status is asserted without the effective ReferenceScheme, ClaimScope, conditions, edition, or relevance window that makes contradiction and freshness checkable.
- Source and display collapse. A badge, list row, dashboard tile, screenshot, certificate view, or generated summary becomes the status source or status use by visibility.
- Design-run substitution. Standard approval is read as runtime satisfaction, or runtime evidence as approval, without an exact interpretation relation and evaluation rule.
- Bridge overread. Shared spelling, a common family label, an F.17 row, an F.18 NameCard, or an F.9 Bridge is treated as the direct explanation, status application, or target result.
- Episteme use drift. A report, standard, model card, dashboard cell, or requirement document is said to hold an “evidence role”, “status role”, or “standard role” rather than participate in an evidence-use, status-use, source-use, standard-use, or requirement-use relation.
Forces
Solution
Recover the governed target and direct result before applying a local status. Treat status value, status-use occurrence, status assertion, source, evaluation, display, and receiving use as distinct.
Three status families
F.10 supplies three status families—EvidenceStatus, StandardStatus, and RequirementStatus—for common project use. A family classifies local status values; it is not a universal result kind and does not create its targets.
A project may define local sublevels or labels, but each label resolves under one effective ReferenceScheme to one exact local sense. Each label maps to one of the three F.10 families—EvidenceStatus, StandardStatus, or RequirementStatus—or to another status family defined by its own pattern. Adding a family row creates neither a system-role kind nor a global synonym.
Status value, use occurrence, assertion, and display
A local status value is designated through an exact F.17 SchemeSenseCell:
An F.18 NameCard may govern its selected public designation. An F.17 row may collect one or more cells for a named unification use; one-cell rows are valid. Neither the cell, card, row, spelling, nor family membership applies the value to a target.
One StatusUseRelation candidate names:
For an F.10-family status, StatusUseRelation(B,T,V,G,W,U) obtains only when: B and T resolve to admitted governed objects; exact cell V has the required F.10 family/local sense under its effective ReferenceScheme; the family-specific source and any direct result/evaluation basis support applying V to T; G and W bound that application; and U is the named intended use without a stronger inference.
If the deciding facts are unavailable, withhold the positive status assertion. State Pending, Inconclusive or another unresolved disposition only when its own rule is satisfied. Absence of evidence is never target falsity.
One F.10 occurrence is identified by the exact ordered tuple <B,T,V,G,W,U>. Repeated evaluations, assertions, displays, rows, records, or citations create no duplicates. A changed bearer, target, value cell, scope, window, or intended use identifies another candidate.
A changed source, evidence path, evaluation, or currentness fact can change whether the fixed candidate is warranted or obtains; it does not enter the relation identity. A status under another exact predicate keeps its own subject assertion and defining or constraining ClaimGraph instead of inheriting this predicate by family resemblance.
A distinct C.2.1 status-assertion episteme states affirmative or negative polarity for the exact StatusUseRelation. A separate display or publication form may render that assertion. The assertion does not perform evaluation, and the display does not become the assertion, source, or actual receiving use.
Recover the target and result first
Use this order:
- name the receiving question and exact target;
- recover the target's identity and direct governor;
- recover any measurement, formal, causal, conformance, diagnostic, comparison, acceptance, requirement-evaluation, gate, assurance, permission, or decision result under its own pattern;
- identify the C.2.1 episteme that states that result;
- resolve the local status expression to its exact F.17 cell and F.10 family;
- recover the source, edition, scheme, scope, conditions, window, provenance, and currentness required by this status use;
- recover the exact rule and any direct/A.6.1 application and evaluation-result claim needed by the use; identify dated evaluation Work and its enacted Method only when that history is current and independently admitted under A.13/A.15.1;
- test the status-use relation under §4.2 and state the corresponding C.2.1 status assertion; then separately recover publication/display and any actual later premise, decision-use, status-use, gate-use, or operation-argument relation.
Status never defines or constitutes the target. A changed status may change a receiving disposition without changing target identity or the earlier domain result. Conversely, a changed target or direct result requires the status application to be re-evaluated; copying the old value is not continuation proof.
A.2.4 status-use positions
When an A.2.4 first-use classification is current, retain its positions by value:
These are relation positions, not system-role-kind qualifier slots, a record schema that applies status, or a new generic status ontic.
Family value sets
EvidenceStatus local values:
Observed— seen or recorded once under declared observation conditions.Measured— supported by a declared measurement method, model, calibration basis, value, and uncertainty.Corroborated— supported by more than one independent source, procedure, or observation line.Replicated— repeated by independent work or under varied declared conditions.Refuted— counter-evidence defeats positive evidential standing inside the same scope and window.Inconclusive— available input results and evidence-use relations are insufficient or mixed for the target claim.
These values classify evidential standing; they do not replace the observation, measurement, proof, causal, or other direct result, and Inconclusive is not target falsity.
StandardStatus local values:
Candidate— proposed and not yet normative for the named scheme/use.Draft— worked text or profile, not yet the governing edition.Approved— sanctioned by the exact governing source for the named scheme, edition, scope, window, and use.Deprecated— discouraged, conditionally allowed, or being phased out.Superseded— replaced by another named edition, profile, or governing source.
Approved alone does not establish an approval act. That claim requires separate recovery of its direct speech-act/decision relation. The status grants no permission and proves no runtime satisfaction.
RequirementStatus local values:
Applicable— the exact clause binds under its governed scope, conditions, and window.Inapplicable— the clause does not bind under those conditions.Satisfied— a direct requirement/acceptance evaluation result says the clause is met for the exact target, scope, conditions, and window.Violated— the direct evaluation result says it is not met there.Waived— binding is suspended or excepted by an exact authorized source/relation and window.Pending— the status application awaits a needed source, input result, evaluation, decision, or currentness repair.
Satisfied, Violated, Waived, and Pending do not replace the clause, evaluation work/result, waiver act or permission, gate decision, assurance result, or action.
Bridge and interpretation discipline
Status meanings do not travel by label. When a comparison needs semantic correspondence between different <ReferenceScheme, LocalSenseClaim> projections, resolve the exact F.17 cells and test the F.9 predicate. Cite an obtaining Bridge as the semantic premise alongside the separate bounded-use claim and current A.10 or B.3 reliance needed by that use.
A Bridge Card or description is separate and optional. F.9 CL is optional evidence-strength shorthand, not a use threshold. The Bridge makes no status-use occurrence obtain and produces no target result.
When one status-use occurrence is used to explain or evaluate a status question of another family, scheme, or modality, recover an exact StatusInterpretationRelation:
It obtains only when the named interpretation rule admits that source occurrence for the exact target question, direction, scope, window, and use. Its occurrence identity is the exact ordered <SourceStatusUseOccurrenceRef, TargetStatusQuestionRef, Direction, InterpretationRuleRef, ClaimScopeAndWindow, IntendedUse> tuple; a Bridge ref is a separate qualifying premise when the use needs an obtaining F.9 semantic relation between different local-sense projections.
A family edge, shared word, Bridge, table row, or source order is not this relation. The rule application and its result claim remain separate. Identify dated evaluation Work only when its history is current and independently admitted under A.13/A.15.1. A positive interpretation relation alone does not produce RequirementStatus=Satisfied, StandardStatus=Approved, a gate result, permission, assurance, or actual later reliance.
Design-run discipline
Keep three questions separate:
- What do exact observation, measurement, proof, causal, or other input results warrant as evidence standing for this target claim and window?
- What does an exact governing source sanction for this method description, profile, standard edition, or configuration and use?
- What does the direct requirement-evaluation result say about this exact clause, target, scope, conditions, and runtime/design window?
A standard-approved method description may be admissible for selection under that profile. It does not show that the method was enacted or that a runtime clause was satisfied.
Runtime evidence may become an admitted input to requirement evaluation through an exact evidence-use and status-interpretation relation. It does not approve the method, standard, gate, or release.
Archetypal grounding
Service acceptance from runtime evidence
July uptime is first recovered as an exact C.16 measurement result, stated by a distinct C.2.1 episteme. A.2.4 classifies that episteme for the uptime claim, and F.10 may assert EvidenceStatus=Measured for that exact claim, scheme, scope, and July window.
The SLO clause and service target are independently recovered. Dated evaluation work applies the SLO rule to the measurement result through exact bindings and produces a requirement-evaluation result claim. A separate RequirementStatus=Satisfied claim for this clause and window needs a result that meets the clause’s evaluation rule. Here the dated evaluation Work supplies that result.
If the comparison relies on a semantic correspondence between the monitoring and service-management senses, test the F.9 Bridge and the bounded use. If the evaluation draws on the Measured status itself, use StatusInterpretationRelation to check whether the rule admits that status for the clause question. The measurement, evidence status, bridge, interpretation, evaluation work/result, requirement status, dashboard display, gate, assurance, and release decision remain distinct.
Approved method description
One exact safety-controller MethodDescription is StandardStatus=Approved only under the named standard/profile edition, source relation, scheme, scope, window, and selection use. That status neither creates the MethodDescription nor proves an approval speech act, permission, method enactment, or response-time satisfaction.
For a controller run independently admitted as U.Work under A.15.1, keep that occurrence separate from its response-time measurement result. That result and its evidence-use relation can enter direct clause evaluation. A separate requirement status may follow from the evaluation; it does not inherit Approved by label or family edge.
Model card and fairness requirement
A model card reports high cross-validation AUC. Recover the exact predictive-performance result and claim episteme first; distinguish that claim from the card's publication or display. F.10 may assert an EvidenceStatus for that predictive claim under its validation scheme and window.
The AUC result leaves the clause “demographic parity delta ≤ 0.1” open. Apply the clause's selection-rate definition to the eligible groups and production window. Measurement can supply the rates. If both groups are nonempty and the evidence establishes the same constant decision for every eligible member throughout the window, their rate difference is zero. A promised measurement procedure or trace has its own completion condition. For a status assertion, retain the policy clause, evidence used, rule application and result. Recover dated evaluation Work when its history is claimed or consumed.
Status display cue
A release dashboard cell shows Ready. The cell is only a cue until exact source assertion, target, value cell, scheme, scope, window, provenance/currentness, and intended use are recoverable. Display or list membership does not establish a status-use occurrence or actual reliance. If the status is consumed for a gate, release, assurance, admission, permission, or decision, the subject pattern must admit the separate use and result.
Bias-Annotation
F.10 addresses five recurring biases:
- label-authority bias: familiar wording is treated as source authority;
- target-by-status bias: assigning a value is treated as defining or creating its target;
- display/list bias: visibility, row membership, or dashboard aggregation is treated as application or actual use;
- family/bridge explanation bias: a family edge, shared spelling, row, Card, or Bridge replaces the exact interpretation relation and rule; and
- system-role drift: an evidence, status, standard, or requirement use is treated as proof that its bearer is a
U.System, has a local system-role classification, or holds a system-role assignment.
The repair is to recover target and direct result first, then the exact local value, relation occurrence, assertion, evaluation basis, display, and receiving use.
None of those use facts establishes System admission, a local system-role classification, or an assignment. Recover System admission and any local system-role classification separately. If an assignment is claimed, identify its obtaining occurrence and declared species, with the independently admitted System as holder.
Conformance checklist
Common Anti-Patterns and How to Avoid Them
Consequences
F.10 requires relation recovery before status can be relied on. The user names exact target/result, local value, scheme, scope, window, source, rule, and use so the status label does not stand in for any of them.
Teams can compare statuses across disciplines, explain why a status was asserted or rejected, locate bridge/interpretation loss, and stop a display from becoming target truth, permission, assurance, gate passage, or work evidence.
The extra work is obtaining each consumed domain result from its subject pattern. F.10 supplies only the status-use and interpretation judgments.
Rationale
Status words sit at the meeting point of evidence, norms, and action, so they are tempting shortcuts. The shortcut remains safe only when target, direct result, local sense, scope/window, source, evaluation rule, and intended/actual use stay visible.
The three status families—EvidenceStatus, StandardStatus, and RequirementStatus—support quick recognition without constituting a common result algebra.
F.17/F.18 govern local sense and designation; F.9 governs cross-local sense bridges; F.10 governs the status-use and interpretation questions; direct subject, evidence, work, result, assurance, gate, permission, and decision patterns retain their own objects.
SoTA-Echoing
Relations
Builds on: F.17 for exact SchemeSenseCells and local-sense rows; F.18 for designation NameCards; F.9 for actual cross-local Bridge occurrences; A.2.4 for first status-use positions; C.2.1 for target-result and status-assertion epistemes; A.15.1/A.6.1 for evaluation work and applications; and the exact subject pattern of every target/result used.
Coordinates with: A.10/G.6 for provenance and bounded reliance; G.11 for currentness; B.3 for assurance-result claims; C.28 for causal use; E.17/E.24.PUB and C.29 for publication/representation; and the direct standard, requirement, acceptance, gate, permission, commitment, release, and decision patterns for their own results and uses.
Precision-restoration exit. When wording such as status role, approved role, validated means compliant, green means ready, or a family arrow hides target, result, value, scheme, window, source, interpretation rule, or actual use, recover those exact objects here and apply the pattern that defines each neighboring claim. Do not repair the phrase by minting a generic status, evidence, or result relation.
F.10:End
Distinguish Method, MethodDescription, Work and Outputs
“Ask separately about the way, its description, the Work that occurred, and any control output produced during that Work.”
Type: Architectural pattern.
Status: Stable
Builds on: E.10.D1 Recovering What “Context” Means in Use; A.3, A.3.1, and A.3.2 for U.Method and U.MethodDescription; A.13 for actual-performer admission, then A.15 and A.15.1 for dated U.Work; C.2.1 for the identity of each claim-bearing episteme or report.
Coordinates with. F.0.1 and F.17 for exact source-local meaning and an optional durable cell; F.4 for local system-role-kind descriptions; F.5 for naming; A.2.1 for system-role assignment and F.6 for precise performed-Work attribution; F.9 only for an actual relation between distinct local meanings; F.10 for status and windows; B.1.5 for Method composition and Work enactment.
Aliases (informative). Method, description, Work, and control-output split; design and run distinction.
Intent & applicability
Intent. Give the reader four practical questions that prevent common category errors:
- What
U.Method, the way of doing, is meant? - What
U.MethodDescription, the episteme whose one exactEntityOfConcernis that Method, is being used? - What dated
U.Workoccurred? - Did that Work produce a domain-specific output that matters to this claim?
These questions are a reading aid, not a universal four-kind ontology. The fourth answer is whatever value and relation its direct control or transformation pattern defines; F.11 does not mint a universal U.Actuation kind.
Use this when. A sentence risks mixing a way of doing with its specification, a design with an occurrence, an approved description with evidence of results, or a Work occurrence with one of its outputs.
Do not use this when. The statement already names one exact value and direct relation unambiguously. F.11 does not prescribe files, tools, workflows, or a generic fact-transfer relation.
Problem frame
- Design-time and run-time blur. A BPMN process description is cited as though it happened.
- Description mistaken for result. Approval of an SOP is treated as proof that later Work met a target.
- Work mistaken for output. A log of setpoints is treated as the whole Work occurrence.
- Word drift. Activity, task, execution, process, and command change meaning across sources.
- Agency blur. A Method or description is said to act, or a vague “System-in-Role” replaces the actual System, local system-role kind, assignment, Work, and performed-Work attribution.
Forces
Core idea (didactic)
Use four questions, then state only the relations that obtain:
- Method — the way. A.3 and A.3.1 govern the way of doing, such as a welding technique.
- MethodDescription — the description episteme. An episteme is a
U.MethodDescriptiononly when A.3.2 finds one admitted Method as its exactEntityOfConcernand at least one substantive claim about that Method as a way of doing. - Work — the occurrence. A dated occurrence is
U.Workonly after admission under A.15 and A.15.1. Work is the occurrence itself, not a record of the occurrence; a record or report is a separate episteme or carrier. - Control or transformation output — if present. Identify the output under its direct pattern and relate it to the Work only when that relation obtains.
F.11 allows the plain sentence “this MethodDescription describes the Method” as shorthand for that A.3.2 constitution and membership judgement. It does not add a binary description relation.
Several epistemes may each have the same Method as their exact EntityOfConcern; one episteme may concern one admitted composite Method; and one document or publication may present several separately identified epistemes. One MethodDescription episteme cannot have several Methods as its EntityOfConcern.
Every admitted Work enacts at least one Method. To claim that it enacted the Method in question, identify that Method and the performance history supporting the claim. Recover the actual performer through A.13 and admit the Work independently under A.15.1; add F.6 only when precise attribution to the same obtaining assignment is needed.
A claim that a System or Work used, followed, deviated from, conformed to, or relied on a particular MethodDescription edition is separate. Cite the pattern that defines or tests that claim; use A.10 or B.3 for evidence reliance, and return A.6.RCD missing-governor when no current pattern supplies the needed relation after its participants and sentence are explicit.
Minimal vocabulary
- Method — a
U.Method, the way of doing. - MethodDescription — a
U.MethodDescription, the already identified episteme that A.3.2 recognizes as having one admitted Method as its exactEntityOfConcern; describes is plain shorthand for that judgement. - Work — a dated
U.Workoccurrence. - Control or transformation output — the exact signal, command, value, output, or changed entity defined by the direct domain pattern when the case contains one.
- Description use — the exact claim that a System or Work used, followed, interpreted, or departed from a MethodDescription; do not assume one universal relation.
- Enactment — the exact relation between Work and Method under B.1.5 and A.15, when supported.
- Performed-Work attribution — F.6
performedUnderAssignmentbetween already admitted Work and the same obtaining assignment recovered through A.13; the actual performer is the assignment's holder, not a third relation participant. - Window — the time or condition envelope used by an F.10 status or evaluation claim.
Solution — four questions
Which way of doing?
Name the Method and its relevant boundary. Do not identify it merely by a file name, notation, or local expression. If a stable method kind or composition is claimed, use A.3, B.1.5, and any required kind pattern.
Which description?
Name the already identified episteme and the one admitted Method that is its exact EntityOfConcern. State edition or version when it matters.
Several epistemes may each concern the same Method. One episteme may concern one admitted composite Method, while a document or publication may present several separately identified epistemes; one MethodDescription episteme never has several Methods as its EntityOfConcern.
Saying that the episteme describes the Method is A.3.2's plain shorthand for this membership judgement. Approval remains a separate claim and is not evidence that Work occurred or succeeded.
Which Work occurred?
For every claimed actual performer, recover the A.13 core: the admitted System, local agential system-role kind and classification, obtaining assignment, and the scope, situation, window and evidence needed by the claim. A.15.1 then independently admits the dated Work.
Use F.6 only when this account needs precise assignment-bound attribution through that same assignment; a missing attribution alone leaves independently admitted Work intact. Recheck a consumed admission premise and its dependent conclusions if the same finding invalidates it, as F.6:4.2 requires. Do not replace these objects with a behavioural “mask” or a System-in-Role pseudo-object.
Which output matters, if any?
Name the signal, command, setpoint, transformation output, or resulting value the Work produced. State how it relates to the Work under the direct control or transformation pattern. Some Work has no control output; the fourth question does not require a universal output kind.
Which evidence and status claim?
Keep approval or validity claims about a MethodDescription distinct from observations of Work and verdicts about Work outcomes in a window. Keep actual use, following, deviation, and conformance claims separate and cite the pattern that defines or tests each one; return A.6.RCD missing-governor when such a pattern is absent. Use C.16 for observations and measurements, A.10 and B.3 for evidence use and reliance, and F.10 and F.12 for status or promise evaluation.
Source-local harmonisation map
Use the following examples as prompts for recovering local meanings. They do not declare identities.
- SPEM and ISO 24744: inspect whether a selected task definition, activity definition, or process description denotes a Method, a MethodDescription, or another value in that exact passage.
- BPMN 2.0: a process diagram is ordinarily a design description; do not call it the Work that later occurred.
- PROV-O: an Activity is a time-bounded occurrence under the PROV scheme. Do not identify it with
U.Workby label; when a report uses that source-local claim for a Work occurrence, state the exact semantic or representation relation established for the case. - IEC 61131-3: distinguish the runtime task execution, the program or program description, and output commands or setpoints.
- SOSA/SSN: an Observation and its Result provide measurement structure; reporting on the Work does not make either one the Work.
Use F.17 only when these local meanings need stable addresses. Use F.9 only if an actual semantic relation between two exact local meanings must be stated. A relation between a description and Work, or between Work and an output, belongs to its direct pattern rather than to a generic Bridge.
Invariants
- Method and description distinction. A MethodDescription is the same episteme recognized by A.3.2, with one admitted Method as its exact
EntityOfConcern; it is not the Method, and describes adds no binary relation. - Occurrence distinction. Work is an actual dated occurrence, not its plan, report, record, or output.
- No universal actuation kind. A control or transformation output is typed and related under its direct pattern.
- Supported enactment. Assert that Work enacts a Method only when the exact relation and its basis are established.
- Explicit description use. MethodDescription use, following, conformance, deviation, interpretation, and reliance are separate claims under their defining or testing patterns; absent such a rule, return A.6.RCD
missing-governor. - Exact agency. Each claimed actual performer has the A.13 core, including the obtaining assignment, before independent A.15.1 Work admission. F.6 is required only for precise assignment-bound attribution; no vague
System-in-Rolesubstitute. - Evidence separation. Approval of a description does not establish Work occurrence or outcome.
- Source-local wording. Ambiguous expressions are recovered with F.0.1; F.9 is conditional on a real relation between local meanings.
Micro-examples
For each dated Work named here and in §11, take independent A.13/A.15.1 admission as a case premise. Its actual performers and obtaining assignments are recovered through that basis. Add F.6 only when a case also needs precise assignment-bound attribution.
- Data pipeline deployment. Method: delta-load transformation. MethodDescription:
etl_delta.py@v3plus its documented rules. Work: the nightly run on 2025-07-14. No control output is material. Approval of the description and measured rows processed are separate claims. - Valve control. The PLC runs the PID control method using gains recorded on its tuning sheet. The cycles from 18:00 to 18:30 produce PWM duty commands from a setpoint and measured temperature. A higher loop may produce the setpoint supplied to this loop. To establish tuning, identify the work in which someone selected or adjusted the gains. Judging settling time requires observations of the controlled temperature.
- Clinical assay. Method: ELISA. MethodDescription: kit IFU v7. Work: the assay run on batch B217. Robot commands are outputs during the Work; absorbance observations support the batch evaluation. IFU approval does not settle the batch verdict.
Anti-patterns & remedies
Worked examples
ML service rollout
- Method: canary deployment strategy.
- MethodDescription: the versioned canary plan with traffic slices and rollback rules.
- Work: two dated canary deployment occurrences.
- Outputs: traffic-shifting commands, if material to the claim.
- Agency: the deploying System and its A.13 core, including the exact local agential kind, classification and obtaining assignment, are part of the independent Work-admission premise. Add F.6 only for a precise attribution through that same assignment. Any responsibility claim keeps its own governor.
- Evidence: latency and error-rate observations about the Work; the plan’s approval is separate.
The plan does not establish SLO satisfaction. F.12 evaluates the promise from Work outcomes in the stated window.
Industrial furnace control
- Method: PID with feed-forward.
- MethodDescription: controller tuning sheet and program description.
- Work: the actual PLC task cycles in the stated interval.
- Outputs: valve-duty commands from this loop. Its input setpoint can be an output of an identified higher loop. Attribute a tuning result to the work that selected the gains.
- Evidence: temperature observations and their scale and unit basis.
If the IEC task expression and a PROV Activity expression are related for reporting, state that exact F.9 relation and loss. It does not create the Work-to-output or evidence relations.
Clinical assay
The Method is ELISA, its MethodDescription is kit IFU v7, and the Work is the assay run on batch B217. Robot commands are optional output detail. Absorbance observations support the quality verdict. Any deviation from the IFU is an explicit description-use or conformance claim, not a property inferred from the four-question layout.
Incident response
The Method is triage-first incident handling; the MethodDescription is the playbook and diagram; the Work is the handling of INC-3421 from 09:10 to 10:02. The handling duration, 52 minutes, is computed from observations of that Work. Command invocations are included only if a direct control or transformation claim needs them.
Reasoning with the four questions
- Classify the subject. Is the sentence about a Method, MethodDescription, Work, or a particular output?
- Keep design and occurrence apart. A design claim does not establish a Work outcome.
- Check membership. Identify the episteme’s one admitted Method as its exact
EntityOfConcernand at least one substantive claim it makes about that Method as a way of doing. For this A.3.2 judgement, say that the episteme describes the Method. - State enactment only when supported. Name the Work, Method, and basis for the enactment claim.
- State description use separately. Say whether and how the Work or performing System used, followed, deviated from, or conformed to the versioned description, and cite the rule that defines or tests that claim; otherwise return the bounded missing-governor result.
- Locate outputs. Relate a signal or changed value to the Work through its direct pattern.
- Bind agency exactly. Recover the A.13 core for each performer before independent A.15.1 Work admission; add F.6 only for precise assignment-bound attribution through the same obtaining assignment.
- Use outcome evidence. Observations about the Work support evaluation; commands and approvals alone do not.
- Preserve history. A new description does not alter past Work or its evidence.
- Recover words locally. Use F.9 only when a relation between local meanings is part of the question.
Relations
Builds on:
- Use A.3 and A.3.1 for
U.Method, A.3.2 forU.MethodDescription, A.13 for the actual-performer basis, and A.15 and A.15.1 for datedU.Work. - Use B.1.5 for Method composition and Work enactment, C.2.1 for the identity of every claim-bearing episteme and report, and E.10.D1 when vague context wording hides the actual source, scheme, scope, use, situation, or evidence basis.
- Use the direct transformation, observation, or control pattern when an output claim is made; F.11 creates no universal actuation kind.
Constrains:
- F.4 and F.6: a local system-role kind, its description, an assignment, and performed-Work attribution remain distinct from Method, MethodDescription, Work, and output.
- F.5: use distinct designations when one source expression hides different subjects. Add a Plain counterpart only when it helps the intended reader and retains the same subject.
- F.7 and F.9: compare exact local claims or F.17 cells and cite F.9 only when a semantic relation obtains. A shared heading creates no identity, relation, or use licence.
Used by. Part C examples and the A.3, A.15, and B.1.5 method-and-work stack. Every MethodDescription membership judgement, enactment, performed-Work attribution, description use, observation, output, evidence use, or publication claim still uses the pattern that defines or tests it.
Migration notes
- Split conflated process. Separate MethodDescription from actual Work; add only the exact relations the case supports.
- Repair statuses. Keep approval and validity claims about descriptions distinct from Work-outcome verdicts and their windows.
- Expose actual outputs. Replace a universal Actuation box with the precise signal, command, value, or transformation output and direct relation.
- Repair agency. Replace
System-in-Roleor behavioural-mask language with the actual System and its A.13 core before independent Work admission; add F.6 only when precise attribution through the same obtaining assignment is needed. - Version fences. Preserve the description version actually used or referenced by past Work.
- Repair hidden transfer. Replace generic Bridge language with MethodDescription membership or the direct enactment, description-use, Work, output, observation, evidence, or source-local semantic relation. Return A.6.RCD
missing-governorinstead of inventing a relation when no defining or testing rule exists.
Acceptance tests
Static conformance
- SCR-F11-S01 (four questions). Every relevant statement identifies the Method, the MethodDescription with that one Method as exact
EntityOfConcern, the Work, or the exact output it concerns. - SCR-F11-S02 (Work actuality).
U.Workis an occurrence, not a record, plan, or output. - SCR-F11-S03 (no universal actuation). Outputs are typed and related by their direct patterns.
- SCR-F11-S04 (agency). Every claimed actual performer has the A.13 core before independent A.15.1 Work admission; any precise assignment-bound attribution has its later F.6 basis through the same obtaining assignment.
- SCR-F11-S05 (separate claims). MethodDescription membership, enactment, description use, output, observation, and evidence claims use their defining or testing patterns and are not replaced by a generic Bridge or invented description relation.
- SCR-F11-S06 (evidence). No approval or command alone is used as proof of Work outcome.
Regression
- RSCR-F11-E01 (description update). Earlier Work and its actual description-use claims remain unchanged.
- RSCR-F11-E02 (source drift). Changed source-local wording reopens only the affected F.9 relation or citation.
- RSCR-F11-E03 (status drift). New statuses do not migrate between description and Work outcome without a new direct claim.
- RSCR-F11-E04 (output growth). Added output detail does not erase or replace Work.
Didactic distillation
“Ask four questions. What is the Method, the way of doing? Which MethodDescription has that one Method as its exact
EntityOfConcern—plainly, describes it? What dated Work occurred? Which particular control or transformation output matters, if any? These are not four universal boxes. Work is the occurrence, not its record. MethodDescription membership adds no binary relation. Every admitted Work enacts at least one Method. Identify the Method or Methods supported by the performance history. Recover each performer's A.13 core before independent Work admission; add F.6 only for precise assignment-bound attribution. Use observations for outcome claims, and use F.9 only for a real relation between source-local meanings.”
F.11:End
F.12 — Evaluate Promise Fulfilment from Work Evidence (Service Acceptance)
“Judge a promise from what happened, in a stated window, with evidence that bears on the promised outcome.”
Type: Architectural pattern.
Status: Stable
Builds on: F.1 Question-Relative Source Selection; F.0.1, F.2, F.3, and F.17 for exact source-local meaning and optional addresses; F.5 for naming; F.9 only for actual relations between local meanings; F.10 for status families and windows; F.11 for the Method, MethodDescription, and Work distinctions; A.2.3 for U.PromiseContent; A.15.1 for evaluation Work; and A.6.1 for the applied evaluation operation and its result binding.
Coordinates with. C.2 and C.16 for observation, characteristic, scale, unit, and measured-value claims; A.3.2 when a selected evaluation MethodDescription edition changes the result; A.10 for evidence use; B.3 only when an actual named assurance claim is current; C.16.P and A.6.RCD when indicator or proxy wording hides an unsupported relation; E.13 only when a proxy is optimized or used as a target, incentive, gate, release argument, reputation signal, repair target, or decision driver; and direct transformation or control patterns when outputs are material.
Scope. F.12 explains the claims needed to justify a promise-fulfilment judgement.
Intent & applicability
Intent. Relate an exact promise-content claim to the delivery Work and outcome being judged, the observations and measured values used as evidence, the relevant window and population, and application of the declared acceptance rule to those inputs. Return a result on its declared scale.
Recover a separate dated evaluation Work when the claim or receiving use consumes that performance history. Map the result to F.10 status when needed, and identify a durable verdict episteme when another use needs that assertion. Keep these objects and relations distinct so a reader can see what would change the result.
Use this when. An SLO, SLA clause, safety margin, response-time target, quality gate, or other promise must be judged from actual occurrences.
Do not use this when. The question is only what the promise says, how the Method is described, or how a measurement is made. Use the direct A.2.3, A.3, A.15, or C.16 pattern. F.12 does not turn a lexical cell or comparison row into a verdict subject.
Problem frame
- Plan ≠ proof. A diagram or playbook is treated as evidence that its promise was met.
- Output ≠ outcome. Commands or setpoints are mistaken for the consumer-relevant result.
- Word ≠ measure. Availability, latency, or incident is used without recovering its exact local meaning and characteristic.
- Proxy by assumption. Monitor output is called a proxy before asking whether the observation and measurement model already concern the promised characteristic directly; when they do not, the needed indicator relation has no defining or testing rule.
- Evaluation disappears. A verdict appears without the selected facts, acceptance rule, calculation or other application that supports its result. A claim about who performed the evaluation additionally needs its actual performance basis.
- Status families collapse.
Satisfied,Violated, andInconclusiveare treated as one universal scale even though the first two are RequirementStatus values and the third is an EvidenceStatus value unless an exact local result scale says otherwise.
Forces
Core idea (didactic)
For an ordinary evaluation, the delivery facts, adequate observations, acceptance rule and supported result may suffice. Recover the evaluation’s performance history when it is claimed or used. Recover the following content as needed to justify the result:
- the exact
U.PromiseContentclaim being evaluated; - the actual Work occurrence or defined Work population whose delivery is in question;
- the promised outcome or characteristic and its scope;
- the relevant observations and measured values, including scale and unit;
- the explicit window and population boundary;
- the evaluation Method and, when its performance history is claimed or consumed, the independently admitted dated evaluation Work that enacts it;
- the selected evaluation inputs, applied rule and result; recover the exact A.6.1 application and result binding when that identity is claimed or required by the governing rule;
- the acceptance rule and the result scale it declares; examples include Boolean, trichotomous, graded,
N/A, andInconclusive-including scales; and - the PromiseContentUse, delivery, fulfilment, measurement, evidence-use, any separately defined indicator or proxy, reliance, and status relations that connect these claims.
The operation returns a value on its declared scale. A separate F.10 rule may warrant a RequirementStatus or EvidenceStatus for that result and use. If the evidence leaves the threshold open, say so. Inconclusive can itself be a local acceptance result when the scale provides for it.
When the receiving decision concerns acceptance of a commissioned result, ask what effect the arrangement gives this evaluation. A passing value settles acceptance if the agreement makes it decisive and the other required conditions hold. A customer empowered to accept a disclosed deviation can accept an item whose measured value still misses the target. The terms determine what the acceptance covers and whether correction is still due. For several linked undertakings, use A.6.C:4.1.
A plain summary may say met, not met, or cannot judge while distinguishing the result on its declared scale, the standing of the evidence and the status of the requirement.
A SchemeSenseCell addresses a local meaning; the evaluation still concerns the exact promise, Work, observations, values, result and status-use relation. A comparison table may display the argument; display alone does not establish those claims.
Minimal vocabulary
- Promise-content claim — the exact
U.PromiseContentunder A.2.3, including its subject, scope, target, and conditions. - Work — the actual dated
U.Workoccurrence, or an explicitly defined population of Work occurrences, being judged. - Observation — the actual observation occurrence and its result under C.2 and C.16.
- Measured value — a value for a named characteristic on an explicit scale and unit.
- Window — the time, batch, episode, phase, or other bounded evaluation interval under F.10.
- Evaluation Work — the dated Work in which a System enacts the evaluation Method over the selected facts and states; identify it when the evaluation account or receiving use consumes that performance history.
- Evaluation application and result — the application of the rule to its selected inputs and the result on the acceptance specification's declared scale. An exact application identity and argument/result binding use A.6.1 when that formal claim is current.
- Evaluation rule and result scale — the stated comparison or aggregation and its declared admissible results. Boolean, trichotomous, graded,
N/A, andInconclusive-including scales are examples, not defaults. - Status use — a separate F.10 application of an exact EvidenceStatus or RequirementStatus value to its exact target, scope, window, and use after the direct result is recovered.
- Verdict episteme — an optional C.2.1 episteme that states the evaluation result or status when another use needs a durable assertion; it is not the operation result or the fulfilment relation.
- Indicator or proxy relation — only a separately defined or tested relation in which one observed characteristic or result stands in for another subject or outcome for this use, with exact participants, coverage, and loss. The word proxy does not create it.
- Evidence use and reliance — the independently established direct relations supporting the acceptance claim, represented in A.10's descriptive evidence-provenance path, with a local reliance disposition for the bounded use. B.3 enters only for an actual named assurance claim. Neither the path nor the disposition creates an indicator relation or the acceptance result.
The binding, as eight practical rules
R1 — Match the promise to delivery.
Use A.2.3 to keep the exact promise content, PromiseContentUse, delivered outcome, and fulfilment relations distinct for the Work occurrence or population being judged. An abstract service label or lexical cell is not enough, and the later evaluation does not make those delivery-side relations obtain.
R2 — Test for direct measurement first. Ask whether the observation and measurement model directly concern the promised characteristic of the relevant Work outcome inside the window. If they do, use C.16 for measurement and A.10 for evidence use; add no proxy relation. Commands, approvals, and MethodDescriptions are not outcome evidence by themselves.
R3 — Recover an indicator relation when one is needed.
If a distinct observed indicator stands in for the promised characteristic or outcome, name both participants and cite the pattern that defines or tests that exact relation. Use C.16.P to recover the construction and distortion risk. If no current rule supplies the relation, stop with A.6.RCD missing-governor; do not treat the word proxy, A.10 evidence use, or B.3 reliance as its substitute.
Use E.13 only when the indicator is optimized or used as a target, incentive, gate, release argument, reputation signal, repair target, or decision driver.
R4 — Perform the evaluation. Select the delivery facts, affected-state references and measured values required by the acceptance rule. Apply its comparison or aggregation with the specified population, exclusions, uncertainty treatment and result scale, then state the supported result. For example, an incident rule can bind observed start and restoration times and compare their difference with the allowed duration. If the required observations or rule are unavailable, return the precise missing basis rather than a fulfilment verdict.
When the claim or receiving use consumes a dated evaluation occurrence, recover each evaluator's A.13 core and let A.15.1 independently admit that Work before asserting that it enacts the evaluation Method, as A.2.3:4.3 requires. Add F.6 only for precise assignment-bound attribution.
Recover an exact A.6.1 application and its argument/result binding when claimed or required by the rule. Cite a particular MethodDescription edition when its content changes the evaluation or replay.
R5 — Use the declared result scale.
Name the characteristic, scale, unit, aggregation, comparison, and admissible result values through the acceptance specification's verdictScaleDescriptionRef. Typical calculation shapes include:
- a value at or above, or at or below, a threshold;
- a stated percentile at or below a target;
- a share such as good time divided by total time;
- an event count within a limit;
- all relevant values remaining inside a band.
The calculation shape does not select a verdict scale. Use the exact scale declared by the acceptance specification.
R6 — State every needed relation directly. Use A.2.3 for promise use, delivery, and fulfilment; C.16 for observation and measurement; the defining or testing pattern for any indicator relation; A.10 for the descriptive evidence-provenance path and bounded reliance on independently established direct relations; B.3 only for an actual named assurance claim; and F.9 only when distinct local meanings themselves require a semantic relation.
One generic Bridge cannot establish clause–Work fit, measurement, indicator validity, evidence, evaluation, status, or fulfilment. A passing reliance disposition requires the evidence demanded by the acceptance claim and its direct rule; a descriptive path alone is insufficient.
R7 — Keep the window and population explicit. A monthly verdict, a batch verdict, and an incident verdict are different claims. A new promise, monitor, or window does not rewrite an earlier verdict.
R8 — Preserve result, status, and assertion boundaries.
Keep the result on its declared acceptance scale. When the evidence leaves more than one fulfilment outcome open, state the unresolved decision and its grounds. Assert a RequirementStatus such as Satisfied, Violated or Pending, or an EvidenceStatus such as Inconclusive, only under its applicable F.10 rule. Create a verdict episteme when another use needs that durable assertion.
Evaluation shapes
The calculation shapes and examples below preserve the delivery Work or population and illustrate an explicit evaluation-Work account. Use that account when its history is current. For an ordinary result, apply the same population, measurement and acceptance conditions through R4 without separately identifying the evaluation occurrence.
Availability share
Promise: availability is at least 99.9% for a calendar month. Delivery Work: the defined service-delivery occurrences or population during that month. Evidence: observations and measurement results for the promised availability characteristic. A System performs evaluation Work; its A.6.1 application binds the in-scope values and returns the declared result.
If the observation model already concerns the promised characteristic, there is no proxy relation. If synthetic probes measure a different characteristic, identify the rule relating it to user experience. State which regions or degradations the probes leave uncovered and how that limits use of their evidence. Without a defining or testing rule return missing-governor. If the rule exists, apply it and limit the verdict to what the probe evidence supports.
Latency percentile
Promise: p95 response latency is at most 120 ms for a stated request population and window. Evidence: response-time observations for that population. Evaluation Work applies the declared sampling, exclusion, and percentile rule and binds its result on the declared acceptance scale. Sampling bias or missing paths can leave the percentile threshold open. Apply any F.10 status rule or declared local result scale separately.
Safety or quality band
Promise: temperature remains within [L,U] during the batch phase. Evidence: calibrated temperature observations for the relevant EntityOfConcern and interval. Evaluation Work applies the stated sampling, uncertainty, and band rule; the exact application binds the values and returns a result on the declared scale. A RequirementStatus follows only through its own rule.
Incident duration
Promise: restoration occurs within 60 minutes for each in-scope incident. Delivery Work: each handling occurrence. Evidence: observations of the defined start and restoration events. Evaluation Work applies the elapsed-time rule to those bindings and returns its declared result. A BPMN design may be a MethodDescription under A.3.2, but it is not either Work occurrence or evidence of the result.
Invariants
- Exact promise. The evaluation identifies the
U.PromiseContentclaim, not merely an SLO label or cell. - Delivery-side relations. The judged delivery Work or population and the applicable A.2.3 promise-use, delivered-outcome, and fulfilment relations remain distinct from evaluation.
- Outcome evidence. Observations and values concern the promised characteristic of that Work; outputs and approvals alone are insufficient.
- Direct-before-proxy. When observation and measurement directly concern the promised characteristic, use C.16 and A.10 and add no proxy. For a distinct indicator, identify its participants and the rule that defines or tests their relation. Return
missing-governorif the rule is missing. If it is known, check its conditions for this case. - Window and population. Both are explicit and match the promise.
- Evaluation basis. The selected inputs, applied acceptance rule and result remain recoverable. A claimed or rule-required dated evaluation Work and exact A.6.1 application satisfy their separate admission and binding conditions.
- Declared result scale. Characteristic, scale, unit, aggregation, threshold, exclusions, and admissible result values are stated as applicable. Boolean, trichotomous, graded,
N/A, andInconclusive-including scales are examples, not defaults. - Status separation. Distinguish the acceptance result, any EvidenceStatus and any RequirementStatus. Each status requires its applicable F.10 rule. When the evidence leaves the fulfilment decision open, report that limit even if the status rules supply no value.
- Optional verdict episteme. A durable C.2.1 assertion is created only for a named later use and never replaces the application result, status-use occurrence, evidence relation, or fulfilment relation.
- Bounded reliance. A.10 supplies the descriptive evidence-provenance path and local bounded-reliance disposition; B.3 is used only for an actual named assurance claim; E.13 is used only for an optimized or decision-driving proxy.
- Non-retroactivity. Later promise, monitor, MethodDescription edition, or interpretation changes do not silently alter past evaluations or assertions.
- Cells are addresses only. An F.17 cell may identify local meaning but establishes none of the substantive claims above.
Micro-examples
SaaS uptime
- Promise content: availability ≥ 99.9% for the named service scope in June.
- Delivery Work population: the in-scope service-delivery occurrences during June, connected to the promise through the applicable A.2.3 relations.
- Evidence: synthetic-probe observations, with regions and outage-detection coverage stated. If their measurement model directly concerns the promised availability characteristic, no proxy is added; otherwise the distinct probe-to-user indicator relation needs a defining or testing pattern.
- Evaluation: the evaluating System applies the rule to the June window. Verified good time gives a lower bound. Include any unclassified time that could still be good when finding the upper bound. Check whether either bound already settles the threshold, using the indicator relation and regional coverage above.
- Status or summary: report whether the threshold was met, missed or remains open. Where a formal status is needed, apply the rule for that status and retain its basis.
Furnace temperature band
The promise content states [720,740] °C during the soak phase. The delivery Work is the actual batch soak occurrence. Calibrated thermocouple observations either measure the product characteristic directly or use a separately defined sensor-location indicator relation.
For this batch, evaluation Work applies the band rule and binds its result. An out-of-band result can support RequirementStatus=Violated under its F.10 rule. If the spatial evidence permits both outcomes, report the limit. Use an evidence status or a local acceptance result when its own rule applies.
Incident restoration duration
The promise content states restoration within 60 minutes per in-scope incident. Each incident-handling Work has observed start and restoration events. A separate evaluation Work occurrence applies the declared event and subtraction rule; its application binds those timestamps and returns the result. A playbook may be the selected evaluation MethodDescription when its edition changes that rule, but it is not the Work or proof of the duration.
Anti-patterns & remedies
Extended worked examples
CDN latency by region
Promise content: p95 end-user latency ≤ 200 ms per region per month. Delivery Work population: delivery occurrences per region in that month. Evidence: response-time observations tagged by region and path. If probes measure the promised characteristic directly, use C.16 and A.10 without a proxy. If they indicate a distinct user-experience characteristic, name the exact probe-to-user relation, its defining or testing pattern, and last-mile loss.
Evaluation Work applies the regional rule to its observations. A numerical p95 needs a nonempty population with the required measurement basis. If the population was empty, use the contract’s specified empty-region treatment. Missing observations alone do not establish an empty region. A global all-regions statement is a separate logical aggregation of those results or statuses, not a property of a table row.
Stroke care door-to-needle
Promise content: at least 90% of in-scope ischemic-stroke episodes achieve door-to-needle ≤ 30 minutes in the quarter. Delivery Work population: patient-episode care occurrences. Evidence: observations of defined door and needle events. Evaluation Work binds those events and divides the qualifying count by the eligible count when the denominator is positive. For an established empty eligible population, use the stipulated empty-population treatment or report that the percentage is undefined.
If missing triage tags or ambiguous events leave both threshold outcomes possible, identify what remains unresolved. A formal evidence status, requirement status or local acceptance result needs its own applicable rule.
Cold-chain warehouse
Promise content: product temperature remains in [2,8] °C for at least 99.5% of each day. Delivery Work: the daily storage occurrence or defined population. Evidence: calibrated thermistor observations. First ask whether the measurement model directly concerns product exposure. If the measurement concerns another characteristic, such as air temperature at the sensor, establish its indicator relation to product temperature and account for stratification loss. Return missing-governor if no rule defines or tests that relation. Otherwise apply the rule with the available coverage.
Apply the promise to the entire in-scope day. For equally weighted minutes, the known in-band count is a lower bound. Adding all unclassified minutes gives an upper bound. Of 1,440 minutes, 1,433 in band suffice for 99.5%. Eight confirmed out of band already establish failure. With 1,430 confirmed in band and ten unclassified, two further in-band minutes would give 1,432 and four would give 1,434. If both completions are compatible with the evidence and the applicable rule, both outcomes remain possible. An independent bound establishing that at least three of the ten minutes were in band already settles this threshold. Use the contract's own treatment of missing readings and assess a separately promised trace on its own terms. Retain relevant indicator limits in the result, status and reliance claim.
SaaS incident restoration duration
Promise content: restoration within 60 minutes for each in-scope incident. Delivery Work: each incident-handling occurrence. Use the start event named by the contract and the observed restoration event. If notification at 10:00 starts the clock and service returns at 11:15, the interval is 75 minutes. Starting repair at 10:40 gives a different, 35-minute interval. Apply the declared duration rule per incident. Quarterly reporting aggregates those results or their separately warranted statuses.
Reasoning about promise fulfilment
- Match scope. Confirm that the promise content covers the exact delivery Work or population and keep A.2.3 promise use, delivery, and fulfilment distinct.
- Name the window. Make time, batch, phase, and exclusions explicit.
- Test direct measurement first. Confirm whether each observation and its measurement model directly concern the promised characteristic; if so, use C.16 and A.10 and add no proxy.
- Recover an indicator only when needed. When another characteristic stands in, name both participants, the defining or testing pattern, coverage, and loss. Use C.16.P to recover the indicator. Return A.6.RCD
missing-governorwhen its defining or testing rule is missing. A known rule can also show that the indicator fails for this use. - Check values. Name characteristic, scale, unit, aggregation, and uncertainty.
- Perform the evaluation. Apply the declared rule to its selected inputs and recover the result on its scale. Identify the performing System, dated Work and exact A.6.1 application when that performance account is claimed or required; keep the actual admission and binding grounds. Cite a MethodDescription edition when it changes the result or replay.
- Use evidence directly. Recover the direct evidence-use claims, the descriptive A.10 path and its bounded-reliance disposition. Enter B.3 only for an actual named assurance claim, and E.13 only when a proxy is optimized or drives a decision, gate, incentive, release argument, reputation signal, or repair.
- Keep the result on its declared scale. Boolean, trichotomous, graded,
N/A, andInconclusive-including scales are examples, not defaults. - Map status separately. Report the result that the acceptance rule and the evidence establish. Apply the corresponding F.10 rule before assigning a requirement or evidence status.
- Create a verdict episteme only on demand. Use C.2.1 only when another use needs a durable assertion about the result or status.
- Aggregate explicitly. Population-level results and statuses follow the promise's stated quantifier; they are not inferred from a few green cases.
- Preserve history. New promises, monitors, evaluation methods, or scales create new evaluations rather than changing old ones silently.
Relations
Builds on:
- Use F.1 and F.0.1 to recover exact sources and local claims, and F.2, F.3, and F.17 only when expressions or durable addresses are needed.
- Use F.5 for clear designations, F.9 only for an actual relation between distinct local meanings, F.10 for separate EvidenceStatus and RequirementStatus uses and windows, and F.11 to keep Method, MethodDescription, Work, and output distinct.
- Use A.2.3 for exact promise content, PromiseContentUse, delivered outcome and fulfilment; A.15.1 for admitted delivery Work and any claimed evaluation Work; and A.6.1 when the exact evaluation-operation application and result binding are current.
Uses direct subject patterns. Use C.2 and C.16 for observations, characteristics, scales, units, and measured values. When the measurement does not directly concern the promised characteristic, use C.16.P to recover the distinct indicator relation and cite the pattern that defines or tests it; use A.6.RCD missing-governor when no such rule exists.
Use A.10 for descriptive provenance and bounded reliance, B.3 only for an actual named assurance claim, E.13 only for optimized or decision-driving proxies, and the appropriate direct pattern for kind, control, or transformation claims.
Constrains: Reporting and assurance preserve the promise content, delivery Work, observations, measured values, relevant window, evaluation rule and result scale. Any claimed evaluation Work, exact application, verdict episteme, EvidenceStatus, RequirementStatus, evidence use, material reliance, indicator relation and F.9 relation retains its own meaning and grounds. A relation-specific CL or loss is reported with that relation rather than folded into the result or status.
Migration notes
- Promise revision. Keep the old promise-content identity, evaluation results, and status assertions; evaluate the new claim separately.
- Monitor change. State whether the new observation model directly measures the promised characteristic or needs a separately defined indicator relation; preserve past evidence identity.
- Scope correction. Retire a result or status assertion about the wrong Work or population and issue a corrected evaluation rather than redefining the promise.
- Scale and unit change. Apply the direct conversion and measurement relations; use F.9 only when local meanings also differ.
- Population refinement. Treat per-region, per-zone, or per-episode changes as explicit promise or evaluation changes.
- Indicator retirement. Prefer direct measurement when available; keep prior indicator-dependent results, status assertions, and evidence uses with their original limits.
Acceptance tests
Static conformance
- SCR-F12-S01 (actual subjects). Every evaluation names exact promise content, delivery Work or population, observations and measured values, window, and rule; cells are optional addresses only.
- SCR-F12-S02 (scope match). Promise, Work, evidence, population, and window align.
- SCR-F12-S03 (evidence). Observations concern the promised outcome of the judged Work.
- SCR-F12-S04 (evaluation explicit). The selected inputs, applied rule, supported result, characteristic, scale, unit, aggregation, threshold and exclusions are recoverable as required by the acceptance claim. Any claimed or rule-required performing System, evaluation Work, enacted Method and exact A.6.1 application satisfy their direct admission and binding conditions.
- SCR-F12-S05 (indicator boundary). Direct measurement adds no proxy. For a distinct indicator, identify both participants and apply the defining or testing rule. Return A.6.RCD
missing-governorif no such rule is available. Limit the acceptance conclusion to the evidence the rule supports. - SCR-F12-S06 (direct relations). Promise use, delivery, fulfilment, measurement, indicator, evidence use, assurance or material reliance, evaluation, status, and any verdict assertion use their defining or testing patterns.
- SCR-F12-S07 (result and status). The operation result stays on its declared scale; RequirementStatus and EvidenceStatus are mapped separately, and evidence insufficiency is never implicit target falsity.
- SCR-F12-S08 (optional episteme). Identify a durable C.2.1 verdict episteme when a receiving use needs that assertion, keeping it distinct from result and status.
- SCR-F12-S09 (no generic Bridge). F.9 is used only for a real local-meaning relation and establishes none of the other relations.
- SCR-F12-S10 (temporal identity). No timeless or retroactively rewritten result or status assertion appears.
Regression
- RSCR-F12-E01 (relation update). A changed indicator, evidence, or semantic relation affects only evaluations that depended on it.
- RSCR-F12-E02 (edition change). Source-local meaning remains tied to the edition used by each evaluation.
- RSCR-F12-E03 (population drift). New population definitions create explicit new evaluations.
- RSCR-F12-E04 (window partition). Weekly and monthly results and statuses remain distinct; any roll-up states its aggregation.
- RSCR-F12-E05 (indicator retirement). Direct measurement changes future evaluations without silently rewriting prior indicator-dependent results or assertions.
Didactic distillation
“Name the exact promise, the delivery Work it covers, the promised characteristic, the observations and measured values, and the relevant window and population. First ask whether the measurement is direct; if another indicator stands in, establish its relation or leave that basis unresolved. Apply the acceptance rule to its inputs and return the supported result on its declared scale. Recover an evaluation-Work or exact application identity when the claim or rule needs it. Use the acceptance result to assign RequirementStatus when its rule applies. Assess EvidenceStatus from the evidence for the stated use. Identify a durable verdict assertion when another use needs one. Plainly: met, not met, or cannot judge.”
F.12:End
Lexical Continuity & Deprecation
“Change names without changing history.” Type: Architectural pattern.
Status: Stable
Builds on: F.1 Question-Relative Source Selection; F.2 source-local term harvesting; F.3 source-local sense clustering; F.5 naming discipline; F.7 Concept-Set comparison displays; F.8 mint-or-reuse for a name; F.9 sense relations and bounded uses; F.17 exact cell and row identity; F.18 naming settlement and lineage. Coordinates with. Part C CALs when canon editions change (Sys/KD/Type/Method/LCA). Boundary. Shared names establish no Cross-context equivalence. Preserve the wording of earlier texts.
Intent & applicability
Intent. Preserve historical reading when source-local expressions, Concept-Set row labels, or names of governed values and their descriptions change:
- new names clarify without erasing what earlier texts meant;
- aliases remain local to Contexts;
- a changed meaning is recovered under its defining rule and requires a stated decision to revise, replace, split, merge or retire it. A rename cannot supply that decision.
Applicability. Use this when changing a source-local expression, an external label or address of a SchemeSenseCell, a Concept-Set row label, or the name of a governed value or description. First say which of these is changing; their identity rules differ.
Problem frame
Readers lose the connection between earlier and later claims when names or senses change without an explicit account.
- Silent relabeling. A new label is introduced as if nothing changed; readers cannot connect past to present.
- Alias bloat. Synonyms accumulate without discipline; reading becomes guesswork.
- Cross‑context aliasing. A single alias is made to stand for different Contexts (“global slang”), defeating locality.
- Retroactive edits. Old texts are silently rewritten to today’s names, corrupting provenance.
Forces
Core idea (didactic)
Keep the named value, its meaning and its expression distinct. Recover the old and proposed expression, effective ReferenceScheme, local-sense claim, intended use and any independently governed value being named. When only wording changes, record the rename or alias and retain a path for interpreting the old text.
An F.17 SchemeSenseCell is the exact coordinate <ReferenceScheme by value, LocalExpression, LocalSenseClaim>. Changing LocalExpression creates another cell even when the <ReferenceScheme, LocalSenseClaim> projection stays the same. Retain both coordinates in the naming lineage as distinct cells.
An external display label or address can change while its unchanged target cell remains the same.
A Concept-Set row under F.7 displays established comparison content; the row does not constitute a Cross-context relation.
A SystemRoleKindDescription under F.4 describes a particular system-role kind identified independently. Recover any continuity of the named value, description episteme, or F.17 UnifiedTermRow under its own rule. Changed identity-bearing claims require the corresponding later object or episteme.
If meaning changes, first apply that subject rule and settle the later naming use under F.18. Split or merge only for one-to-many or many-to-one changes in senses or rows. A one-to-one revision needs its own revision or replacement decision under the subject rule. Preserve the earlier meaning and read-path; when that decision is missing, stop the dependent continuity claim.
Keep aliases local. Compare the exact
<ReferenceScheme, LocalSenseClaim>projections. A different projection opens F.9 only when a current semantic correspondence is needed. Establish the relation and any bounded-use claim separately. Decide what the receiving use may rely on. A shared name establishes none of them.
Minimal vocabulary (this pattern only)
- Legacy label — an earlier expression retained with its exact local meaning and use, or an earlier external label of an identified value, cell, description or row.
- Preferred label — the current label selected for that naming use under F.5 and, when a durable settlement is needed, F.18.
- Alias (local to a naming use) — a read-path from a legacy expression to the preferred one with the same effective scheme, local-sense claim and use, and the same governed value when one is being named. Writing uses the preferred label.
- Continuity relation — one of the descriptive relations below over exact label uses. It records a justified wording or historical-reading decision; it does not establish governed-object, cell, episteme or Bridge identity.
- Epoch note — an informative time marker attached to a legacy label when needed to interpret older text.
Solution — Continuity
Use the least-committing continuity relation that tells the truth after recovering the subject's identity or change under its own rule.
Continuity relations (normative meanings)
-
renames(label_old → label_new)— wording changes while meaning and use remain unchanged. Use when: The effective scheme, local-sense claim and intended use remain the same, and the independently governed value remains the same when one is being named. For an external row or description label, recover its target under F.7, F.4 or F.17 as applicable. Effect: The legacy label becomes a local read-path to the preferred one. IfLocalExpressionchanges, retain both exact F.17 cells; if a NameCard or row ClaimGraph changes, retain the earlier episteme and the corresponding later one. The rename itself proves none of those identities. -
aliases(label_legacy ↔ label_pref)— a legacy synonym is kept for reading. Use when: The two exact label uses have the same scheme, local-sense claim, intended use and, where applicable, governed value. Effect: Keep the needed two-way read-paths; writing useslabel_pref. Preserve historical aliases that readers still need, distinguished from the preferred current designation. When the two labels occupy theLocalExpressionslot, they identify distinct F.17 cells. -
splits(label_old ⇒ {label_A, label_B})— one earlier label covered several senses or rows that are now distinguished. Use when: F.3 recovers the distinct local senses, or F.7 establishes the required row refactor. Effect: Deprecate the old label for new writing and retain a disambiguating read-path. Neither new label is asserted to continue the whole earlier meaning. -
merges({label_A, label_B} ⇒ label_new)— several labels are consolidated for one recovered naming use. Use when: The local label uses have the same scheme, local-sense claim and use, with the same governed value where applicable; or two F.7 rows are duplicate displays of the same named comparison or use, exact entries, independently established relations, losses, basis and receiving-use conclusion. Effect: Keep the old labels' epoch-qualified read-paths needed by historical or current readers. This label or display consolidation does not merge source-local cells, named values or Bridge occurrences. -
retires(label_old)— a name is withdrawn without one successor. Use when: The label misleads and no single successor preserves its earlier use. Effect: Keep a warning for historical reading and point to the relevant meanings or rows. A known single successor for an unchanged meaning belongs to rename, not retirement.
Meaning-change boundary. These five label relations do not exhaust changes to the things being named. For a one-to-one changed criterion, recover the revision or replacement under the subject's rule, preserve the exact earlier claim, and settle the later name under F.18. If the governing continuity decision is unavailable, state that missing decision instead of inventing a split or merge.
A needed relation between different semantic projections is a separate F.9 question.
Invariants (normative)
- Locality of alias.
aliases(-)andrenames(-)preserve the exact naming use's effective scheme, local-sense claim and any governed value. Different values in theLocalExpressionslot identify distinct F.17 cells; an external row or description label keeps its independently recovered target. - Truth over comfort. If meaning changes, recover the corresponding revision, replacement, split, merge or retirement under the subject rule before settling the later name. A rename cannot hide that change.
- Non‑retroactivity. Leave past texts as written and add read‑paths.
- Alias parsimony. For the governed value and naming use, keep one preferred current designation in each register used and the historical aliases needed to recover actual uses. Omit unused synonyms from the main presentation without erasing the naming history.
- Prefer present for writing. In normative writing, use the current preferred label (F.5). Aliases are for reading comprehension.
- Bridge discipline. A different
<ReferenceScheme, LocalSenseClaim>projection is not an alias or pure rename. Use F.9 only for a needed semantic correspondence; recover any obtaining Bridge, bounded-use claim and reliance separately. - Epoch qualification. When declaring continuity, attach a succinct epoch note (“pre‑2023 usage”) if it aids readers.
Self‑checks (mental, not procedural)
- Same-sense test. Do the effective scheme, local-sense claim, intended use and any governed value stay the same? Then a rename or alias may be justified, with distinct cells when expressions differ. If meaning or the target changes, first recover that change under its subject rule.
- Context test. Are the semantic projections different, and is a current correspondence needed? If both hold, open F.9. Different context names alone establish no relation.
- Reader test. Which legacy strings will readers encounter, and under which meanings and editions? Preserve the read-paths they need while keeping the preferred current designation clear.
- History test. Does your “continuity” require editing old claims? If yes, you’re attempting a retroactive rewrite—stop.
- Didactic test. Can the intended reader recover what changed and how to interpret the old label? If not, clarify that account or name the unresolved subject decision; sentence length alone diagnoses neither sameness nor change.
Micro‑examples (illustrative)
Pure rename inside a local service vocabulary
Constructed case: Under the fixed ServiceVocabulary-1 reference scheme, the local-sense claim is “a service indicator threshold together with its assessment window”. The old expression is “SLO”, the proposed expression is “service-level objective”, and both serve the same use of reading a service objective. No change to the objective or assessment rule is proposed.
Relation: renames("SLO" → "service-level objective").
Why. The scheme, sense and use are unchanged. Spelling out the name lets readers see what the abbreviation stands for. A claim about improved recognition by this audience needs evidence of their reading.
Effect: Retain the two exact cells <ServiceVocabulary-1, "SLO", claim> and <ServiceVocabulary-1, "service-level objective", claim> in the naming lineage. Old text retains its earlier expression and meaning; new writing uses the expansion. No Bridge is needed for these identical semantic projections.
Alias for a common legacy synonym (Sys‑CAL)
Illustrative premise: A local state-space control vocabulary has independently established the same scheme, sense and use for “control output” and “actuation”. This premise requires source-local recovery in a real application; the words alone are insufficient.
Relation: aliases("control output" ↔ "actuation").
Effect: Keep the distinct expression-bearing cells and one historical read-path. New text uses the locally selected preferred expression “actuation”.
Split of a muddled local sense (Enactment)
Context: BPMN 2.0.
Legacy label “process” was used to mean both “collaboration” and “executable process” in a team’s prose.
Relation: splits("process" ⇒ {"collaboration","executable‑process"}).
Effect: The single Concept‑Set row becomes two; old label is deprecated with a disambiguation note.
Merge duplicate comparison rows
Suppose the two Concept-Set rows labelled “DBaaS” and “Database-Service” display the same named comparison or use, exact entries, independently established relations, losses, basis and receiving-use conclusion. They are duplicate displays; combining them asserts no new sameness among their entries.
Relation: merges({"DBaaS","Database‑Service"} ⇒ "Database‑Service").
Effect: “DBaaS” becomes a legacy alias with an epoch note.
Not a rename: Cross‑context temptation (forbidden)
Contexts: BPMN (design graph) vs PROV‑O (run activity). Temptation: “Let’s rename process to activity.” Diagnosis: A design graph and a run occurrence have different local meanings. A rename requires the same meaning and use. Action: Keep the labels and their exact meanings. If a semantic correspondence matters, test the relation under F.9. If a proposed use relies on it, assess that use and its current grounds separately. A design-to-run or production relation belongs under the rule for that relation.
Anti‑patterns & remedies
Extended examples
KD‑CAL × Services — metric target labels over time
- Contexts: ITIL 4 (services, design); SOSA/SSN (sensing, run).
- Before: An illustrative local service-objective description used “SLO” (plain “target”), and older texts also used “service target” for the same locally recovered meaning.
- Move:
renames("SLO" → "service‑level objective")(Context: ITIL). Keepaliases("service target" ↔ "service‑level objective"). - Why: The example assumes the same scheme, local-sense claim and use. Retain each expression's exact cell and the later naming settlement; SOSA/SSN labels are unchanged.
- Pay‑off: Runtime Observations (SOSA) are later compared to service‑level objective clauses (ITIL) without Cross‑context aliasing.
Sys‑CAL × LCA‑CAL — separating execution vs actuation
- Contexts: IEC 61131‑3 (run); state‑space control texts (design).
- Temptation: Rename “task execution” to “actuation” “to sound control‑ish”.
- Diagnosis: Different Contexts; different SenseCells (program run vs control output).
- Move: No rename. Keep the labels. Establish any actual Work-to-signal production or output relation under the pattern for that relation, following F.11. Add F.9 only when a separate semantic correspondence between the particular local senses is needed. Then keep the bounded-use claim and reliance separate.
- Pay‑off: Plant narratives stop calling programs “actuators”; runtime vs control semantics stay crisp.
Kind-CAL × method/work stack — false merge avoided
- Contexts: OWL 2 (types, design); SPEM 2.0 (methods, design).
- Issue: A row labeled “Class” tried to absorb “WorkProductKind” by a
renames. - Diagnosis: Not same sense; different calculi (type vs artefact category).
- Move: Reject the proposed rename. If the old row already combined both meanings, apply
splits("class" ⇒ {"type‑class","work‑product‑category"})and preserve the earlier reading path. - Pay-off: Later descriptions can cite the exact local meaning they use. Any SystemRoleKindDescription still describes its independently recovered system-role kind under F.4; a source-cell pointer does not establish that kind.
Enactment × KD‑CAL — replacing a misleading metaphor
- Context: BPMN 2.0 (design).
- Legacy: Team jargon “heartbeat” used for a timer event. Newcomers confuse it with sensor heartbeats (KD‑CAL).
- Move:
renames("heartbeat" → "timer event")for the team’s unchanged BPMN timer-event meaning. Keep a local historical read-path; the sensor-liveness use remains a different local meaning. - Pay‑off: Two different ecosystems stop colliding on the same catchy word.
Concept‑Set row refactor after duplicate comparison is established
- Rows:
{“DBaaS”, “Database‑Service”}representing service notions across several Contexts. - Comparison basis: Both rows display the same named comparison or use, exact source-local entries, independently established relations, losses, basis and receiving-use conclusion. Only the duplicate display is combined; no source meanings or Bridge occurrences are merged.
- Move:
merges({"DBaaS","Database‑Service"} ⇒ "Database‑Service")at row level. Keep “Database‑Service” as the preferred row label and “DBaaS” as its legacy alias with an epoch note. - Pay‑off: One clearer row label; old articles still understandable.
Reasoning primitives (judgement schemas, notation‑free)
Use these schemas to decide which continuity claim the recovered facts support.
Let meaningOf(ℓ) recover the exact <ReferenceScheme, LocalSenseClaim> projection for one local label use; let useOf(ℓ) recover its intended naming use.
Where a value is being named, recover it independently under its subject rule. pref(t) denotes the preferred expression for that exact naming use, not a global preferred word.
Same‑sense & same‑place
Two expressions with the same meaningOf, intended use and independently governed value where applicable may receive a rename. A changed LocalExpression still changes the exact F.17 cell. An external label change preserves a cell only when the target coordinate itself is unchanged.
Local alias
Under the same conditions, a legacy expression may remain a read-path to pref(t). Retain its coordinate and any source basis. Aliasing does not establish cell identity.
Split detection
If one label covered several independently recovered local senses or rows, record the split and prefer the precise later labels. Keep the earlier use resolvable through a disambiguation note.
Merge admission
Consolidate local labels only after recovering their common semantic projection and use, with the same governed value where applicable. Consolidate comparison-row labels only after establishing duplicate F.7 comparison content. Neither move merges distinct cell coordinates or subject entities.
Retirement
Retire a misleading label when no single successor preserves its earlier use. Point historical readers to the relevant meanings or rows.
Cross‑context guard
Different meaningOf projections do not support a pure rename or alias. A needed semantic correspondence goes to F.9; relation obtaining, a bounded-use claim and reliance are separate questions.
Writing discipline
Use the current preferred expression for the exact naming use. Keep aliases for historical reading.
Reading resolution
Resolve a legacy label to its earlier meaning and use first, then to any justified present-name read-path. Include an epoch when it changes interpretation. A successor's changed meaning must not replace the old one retroactively.
Alias budget
Keep the historical aliases needed by the naming use, with their meanings and editions recoverable and the preferred current designation clear.
Row‑level continuity
A Concept-Set display label may change while the exact comparison content and receiving use remain unchanged. Recover changed row content under F.7 first.
When an F.17 UnifiedTermRow's identity-bearing claims change, retain the earlier and corresponding later epistemes under C.2.1/F.17; a stable row designator is not identity proof.
Relations
Builds on: F.1 selects the question-relative sources; F.2 and F.3 recover source-local expressions and meanings. F.5, F.8 and F.18 govern naming choices. F.7 governs comparison displays; F.9 governs exact sense relations and bounded uses. F.17 fixes SchemeSenseCell and UnifiedTermRow identity. F.4 and the direct subject patterns govern any described or named values; F.15 checks the resulting continuity claims.
Constrains:
- F.5 (Naming): when replacing an earlier label, recover its historical continuity before selecting the new preferred label.
- F.7 (Rows): relabel the display when its comparison content is unchanged; recover changed entries, relations or receiving use before deciding revision, split or merge.
- F.9 (Bridges): Cross‑context changes must not be expressed as renames/aliases.
Used by. All Part C patterns when editions shift; all examples and tutorials when teaching with legacy terminology.
Migration notes (conceptual playbook)
- Recover the same meaning and use first. Distinguish the expression from its exact cell and governed value. Use
renamesonly for unchanged meaning and use; recover any subject change before settling the later name. - Keep aliases local. Different semantic projections open F.9 only for a needed correspondence; the difference itself establishes no Bridge.
- Choose for the audience. When proposing a new name under F.5, explain which actual reading difficulty it would resolve. An observed improvement in recognition requires a comparison with the previous form under relevant reading conditions.
- Keep useful read-paths. Preserve the legacy expressions readers need to interpret historical or continuing uses; keep unused synonyms out of the main presentation.
- Keep historical reading clear. For a rename, point from the legacy label to its preferred successor. For retirement without a single successor, point to the relevant meanings or rows.
- Rows before names. If a rename request coincides with a shift in what the row covers, refactor rows (F.7) first, then choose labels.
- Edition changes. Compare the affected definitions. When meaning changes, recover the revision, replacement, split, merge or retirement and preserve the earlier claim. If the source adopted a new designation for the same meaning and use, record that change and retain the historical alias. The new edition establishes its designation even when the audience preferred the old one.
- Teach the delta. In primers, show a mini table with legacy → preferred pairs only where readers will encounter both.
Acceptance tests (SCR/RSCR — concept‑level)
Static conformance (SCR)
- SCR-F13-S01 (local continuity). Every
renames/aliasespreserves the exact semantic projection, intended use and any independently governed value. An external row or description label resolves its target under the applicable rule. - SCR-F13-S02 (Truthfulness). Each rename retains its exact old and new expressions. A changed LocalExpression has a distinct F.17 cell even when meaning is unchanged. Any claim of governed-value or episteme continuity satisfies that subject's rule.
- SCR‑F13‑S03 (Alias purpose). Every retained legacy alias serves an actual reading or retrieval need and resolves to its historical meaning; it is distinguished from the preferred current designation.
- SCR‑F13‑S04 (Non‑retroactivity). No requirement or suggestion to rewrite past texts is present; continuity is expressed as read‑paths.
- SCR-F13-S05 (Row integrity). A display-label rename preserves the exact comparison content and receiving use. Changed row content receives its warranted revision, split or merge under F.7; changed F.17 identity-bearing claims receive the corresponding later row episteme.
- SCR‑F13‑S06 (Bridge discipline). No alias/rename is used to imply Cross‑context sameness; any such relation belongs under F.9.
Regression (RSCR)
- RSCR-F13-E01 (Edition drift audit). Compare affected earlier/later definitions and naming uses. Stable meaning can support a rename with distinct expression-bearing cells; changed meaning requires the subject-governed revision, replacement, split, merge or retirement before fresh naming settlement.
- RSCR‑F13‑E02 (Alias change). After changing aliases, verify that needed historical read-paths remain usable and the preferred current designation remains clear.
- RSCR-F13-E03 (Bridge leak check). A claimed correspondence between different semantic projections is tested under F.9 only when a current use needs it. Retain exact meanings and an unresolved result when obtaining is unestablished; do not create a Bridge by rewriting a note.
- RSCR‑F13‑E04 (Didactic continuity). Sampling of examples shows that readers can resolve legacy labels to current ones without confusion (via the continuity notes).
Didactic distillation (60‑second script)
Preserve what the old words meant. For unchanged meaning and use, record a local rename or alias. A changed expression has a distinct F.17 cell; keep its naming lineage.
If meaning or the named value changes, recover that change under its subject rule before settling the later name. Split or merge only when those relations hold. Keep the historical aliases readers need, preserve the earlier text, and give readers the exact historical meaning and any justified path to the present name.
F.13:End
Anti-Explosion Control for System-Role and Status Name Families
"Name less; recover the governed values first."
Type. Architectural pattern.
Status. Stable.
Normativity. Normative.
Builds on: A.2 for exact context-local system-role kinds; A.2.1 for U.SystemRoleAssignment; A.2.5 for assignment-state predicates and direct state relations; A.2.7 for exact substitution, incompatibility, qualification, and bundle relations among system-role kinds; A.15.1 for performed Work; F.4 for system-role-kind descriptions; F.5 for local naming discipline; F.8 for one mint-or-reuse decision; F.9 for actual relations between exact local senses; F.10 for status families and windows; F.18 for durable naming; and A.6.5 for relation-slot discipline.
Coordinates with: A.2.2 for capability, A.3.1 and A.3.2 for Method and MethodDescription naming, A.10 and B.3 for evidence and assurance use, E.10.D2 for description use, E.24.PUB for publication occurrence, expression form, and carrier, and F.17 only when a public, Core-facing, durable, or cross-local term row is current.
Plain entry cues (informative). Name explosion guard; system-role-name economy; status-name economy; stop before another card or row.
Intent and applicability
Use this when. Proposed names, aliases, cards, local-sense cells, or rows begin to multiply faster than the independently governed distinctions. Before minting a NameCard, SchemeSenseCell, Unified Term Sheet row, or durable name family, ask: does an existing designation, alias, local expression, or direct-pattern name already let the practitioner perform the proposed use?
First useful move. For every candidate expression, name the one independently recovered governed value or relation, its exact kind, its direct pattern, the proposed use, and the effective naming rules (ReferenceScheme).
If no such value or relation is independently recoverable, keep the expression local or keep it with the exact assertion that recovers its subject or value; do not pass a value-less expression to F.8 or manufacture an object so that the name has something to denote.
F.8 receives only an unresolved naming disposition for an already recovered value-or-relation and proposed-use pair, with its exact kind and direct pattern.
Intent. Keep system-role-facing, role-like, and status-like vocabularies small without losing real distinctions. F.14 reviews candidate expressions and name families. It defines no system-role kind, status, assignment, sense, card, row, Bridge, or publication. Use it to decide whether a simpler naming choice suffices.
Primary working object. One candidate family and one proposed use, with its recovered values and direct patterns. A durable control record is optional; no generic context object, selected structure, card, or table row identifies the pass.
Primary working reader. A practitioner deciding whether proposed system-role or status names preserve independently governed distinctions or merely multiply labels.
What goes wrong if missed. System-role-kind labels become capability models, status labels become system-role families, access-control labels become work-facing kinds, and every local wording difference acquires a card, sense cell, row, or identifier. The corpus then contains many near-duplicate naming objects whose apparent precision hides different kinds and uses.
What this buys. A smaller vocabulary with stronger type separation and a way to stop early. Use no durable name, an existing designation, an alias, or a local expression whenever one suffices. Otherwise use the smallest justified durable naming object.
Not this pattern when. Use F.8 to make the final naming disposition for one candidate expression only after its governed value or relation, exact kind, direct pattern, and proposed use have been recovered; F.14 supplies the preceding anti-explosion stop rather than a second decision record.
Assignment claims go to A.2.1. For precise performed Work, A.13 first recovers each exact actual performer and A.15.1 independently admits the dated occurrence; F.6 is added only when the naming case or receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment. Status, evidence, authorization, publication, and other relation claims require exact predicates in their direct patterns.
Add a reader-facing F.17 row only after kind recovery, the F.14 stop, any needed F.8 or F.18 naming decision, and satisfaction of the public-row threshold; treat publication availability as a separate E.24.PUB question.
Recognition versus assurance. Recognition is the visible name-growth pressure plus the first kind-and-use recovery. Assurance is the optional record, invariants, worked countercases, and conformance tests. Neither turns F.14 into naming authority or ontology.
Problem frame
Name explosion usually begins with a helpful shortcut:
- Hybrid-system-role shortcut.
RequestApproverSystemRoleis minted because several local system-role kinds often appear together. - Modifier-as-system-role shortcut.
NightOperatorSystemRoleis minted because a qualifier is visible. - Status-as-type shortcut.
PreValidatedis minted as if time stance or status value established a new kind. - Source-suffix shortcut.
EvidenceRoleis minted because a source tradition uses role-like language. - Prestige shortcut.
SeniorRevieweris minted to bypass a separation, capability, or assurance question. - Locality shortcut. The same spelling under two local-sense bases is treated as one value, or every difference is answered with a Bridge, card, cell, and row before a receiving use exists.
Use F.14 to test whether these shortcuts justify a durable naming object.
Forces
Core idea
Use this sequence before minting a durable name or any supporting naming object:
-
Recover the governed value first. Separate candidate expressions whenever they designate different governed values or relations. Recover each retained value's exact kind and direct pattern; keep an ordinary phrase local when no governed naming case is needed.
-
Name one proposed use and its interpretation basis. State what the reader will do with the expression and the effective naming rules (
ReferenceScheme). An independently selectedBoundedModelUseStructureappears only when that organization changes this naming use; it is never a generic locality field. -
Select the lightest sufficient treatment. Use adequate ordinary wording, an existing designation, an alias, a local expression, a direct-pattern name or a current public row. Reuse an already justified choice; compare alternatives when they could reduce burden without losing a governed distinction.
-
Create only the next object that pays for itself. A local
SchemeSenseCellis useful only when the exact local sense needs a stable address. A NameCard is useful only when the naming settlement itself must endure. Create an F.17 row only for public, Core-facing, durable, or cross-local reuse. Use E.24.PUB only when the selected row edition must actually be made available. None implies the next. -
Use exact relations instead of fused names. Bundles and incompatibilities among system-role kinds remain A.2.7 relations; assignment claims remain A.2.1.
For precise performed Work, A.13 first recovers each exact actual performer and A.15.1 independently admits the dated occurrence; F.6 follows only when the naming case or receiving use expressly consumes precise assignment-bound attribution through the same obtaining A.13 assignment.
Status families and windows remain F.10; qualifiers remain with their direct patterns.
-
Treat cross-local wording as a relation question only when one is current. Resolve the exact local senses first. Same spelling proves nothing; different local-sense projections only open F.9. Cite a Bridge only when its predicate obtains, then state the proposed use and reliance separately. A Bridge does not merge governed values or require a public row.
The result is the smallest naming disposition that preserves the governed value and supports the named use. It does not establish the independently recovered subject or any neighboring relation.
Minimal vocabulary
- Anti-explosion control pass — one bounded review of related candidate expressions before durable naming objects are added.
- Candidate name family — proposed expressions that appear to designate related values or relations.
- Recovered governed value — the exact typed value or relation the expression is trying to designate, under its direct pattern.
- Naming use — what the reader or practitioner proposes to do with the expression.
- Light disposition — no durable name, existing designation, alias, local expression, or existing row reuse.
- System-role-kind relation expression — an expression designating an exact A.2.7 substitution, incompatibility, qualification, or bundle relation rather than another local system-role kind.
- Status-family expression — an expression for a status family, value, window, confidence claim, or status-use relation defined under F.10 or a direct status pattern.
- Blocked minting — the explained decision to keep the candidate as local wording or reuse an existing designation rather than create a new durable name or naming object.
Optional anti-explosion record
Ordinary use needs no record: recover the value, choose the lightest sufficient disposition, and stop. Persist this C.2.1 description episteme only when several related candidates, a contested decision, or later replay makes the family-level reasoning useful.
The record describes the control result. It creates no governed value, naming decision occurrence, designation, local sense, Bridge, row, publication, evidence, system-role kind, status, assignment, or Work.
A field is omitted when its object is not independently current; filling the record is never a completeness goal.
Levers
Recover kind before naming
Reuse before minting
Reuse only when the exact recovered value, kind, direct pattern, proposed use and admitted naming scope match. Before creating a card, cell, row, policy id or new U-kind candidate, establish the receiving need that adequate ordinary wording or an existing designation would not meet. Reuse available grounds for that decision.
Local-sense reuse does not imply sameness with another local sense; row reuse does not widen the row's admitted use.
Use relations among system-role kinds before hybrid kinds
If the use requires assignments to several kinds together, recover the A.2.7 bundle condition and its holder-allocation rule. If one kind requires a separately established qualification, recover that relation. If they must stay apart, recover the exact A.2.7 incompatibility.
When checking actual assignments against one of these rules, identify their occurrences under A.2.1 and any applicable A.2.5 currentness condition. Use F.6 only when the receiving claim separately says that dated Work was performed under one of those assignments. If one kind can satisfy another requirement, recover exact substitution. The relation expression assigns no system and does not become a new kind by name.
Use a status window before multiplying status families
If the proposed name marks evaluation, active use, grace, archival state, confidence, or presentation, keep the status family and use F.10 windows, values, or direct status-use relations. A new status family needs a recovered governed difference, not another adjective.
Keep qualifiers with the claims they qualify
Time, location, object type, seniority, permission, Method, capability, evidence, source, and publication are not system-role-kind or status identity by suffix. Keep a qualifier with the claim it qualifies and use the pattern that defines, constrains, or tests that claim. Retain the qualifier in a durable name only when the already governed value and named use require that designation.
Stop before a naming-object cascade
A candidate can justify one object without justifying all later objects. A durable local expression needs no cell. A stable local sense may need a cell but no NameCard. A durable naming settlement may need a NameCard but no public row. A row may exist without a current publication occurrence. Publication availability creates neither row truth nor governed-value truth. Apply the next gate only when its own use is current.
Invariants
- Governed value first. No durable naming object is added until the exact value or relation, kind, proposed use, and the pattern contribution that defines, constrains, or tests each needed claim are recoverable.
- Lightest sufficient disposition. Prefer the dispositions
no durable name, existing designation, alias, or local expression whenever one supports the use without hiding a distinction. - No status roles. Status, evidence, requirement, source, publication, and access uses do not become system-role kinds by suffix.
- No assignment by name. A designation,
SystemRoleKindDescription, system-role-kind relation expression, card, cell, or row assigns no system and proves no Work. - No hybrid kind by convenience. Exact A.2.7 relations remain relations unless A.2 with C.3 independently admits a different local system-role kind.
- No capability or authority by label. System-role-kind and status names prove no capability, skill, permission, assurance, evidence use, Method validity, or publication authority.
- Local senses do not globalize. Same spelling and different local-sense projections establish neither governed-value identity nor an F.9 Bridge.
- Naming objects remain optional and distinct. Expression, designation, alias, cell, NameCard, row, identifier, publication occurrence, form, and carrier neither imply nor replace one another.
- Selected structure is conditional. A
BoundedModelUseStructureis cited only when its organization changes the naming use under consideration and never becomes a locality slot or naming identity field. - Lineage is not ontology. Historical spelling may be recorded as lineage without carrying its former fused commitments forward.
Reasoning primitives
An obtaining systemRoleKindBundleRelation(K1, K2) does not by itself establish a new K1K2 system-role kind. Any such kind requires independent admission under A.2 with C.3.
The presence of a naming object establishes neither that its governed value exists nor that another naming object is required. Recover the value independently and test the receiving-use condition for the next object.
These are stopping and dispatch rules. They create no values or relation occurrences.
Worked cases
Requester and approver
Candidate family: RequesterSystemRole, ApproverSystemRole, RequestApproverSystemRole, SeniorApprover.
Premise: the local practice has already admitted RequesterSystemRole and ApproverSystemRole as exact local system-role kinds under A.2 with C.3. If that independent basis is absent, the case returns two candidates whose kind status is unresolved; the spellings do not admit them.
Result after that premise:
- Reuse the two admitted kinds and keep separate
SystemRoleKindDescriptionepistemes only when descriptions are needed. RequestApproverSystemRoleis blocked as a fused kind. Use an A.2.7 bundle when the practice requires the two assignments together and specifies how holders may fill them.- If the same holder must not carry both assignments in the same change window, use the A.2.7 incompatibility relation. To check an actual pair against that rule, recover its assignments under A.2.1 and any applicable A.2.5 currentness condition. Use F.6 only if a separate claim says that dated Work was performed under one of them.
SeniorApproveris not proof of independence or assurance. Recover the intended local system-role kind, exact assignment-state predicate or relation, capability, assurance, or policy claim before durable naming.
Operators across shifts
Candidate family: OperatorSystemRole, NightOperatorSystemRole, RemoteOperatorSystemRole, OnCallOperatorSystemRole.
Premise: A.2 with C.3 has independently admitted OperatorSystemRole as an exact local system-role kind. Without that basis, OperatorSystemRole is still a candidate name and the case makes no kind claim.
Result after that premise:
-
Reuse the admitted
OperatorSystemRolekind. -
night,remote, andon-callare qualifiers in the proposed wording. Recover the claim each qualifies—for example, a schedule, location relation,SystemRoleAssignmentStatePredicate, WorkPlan, or policy condition—and use the pattern that defines, constrains, or tests that claim. -
A new system-role kind is blocked unless A.2 with C.3 independently recovers a distinct local kind through its
U.Systemcandidate domain, operative work-facing membership condition, useful member/non-member boundary, and continuity rule.Its criterion may use, for example, a capability, Work, or an assignment established separately, but assignment conditions, a Method, and Work implications are not universal requirements. A practice, source, suffix, or naming ReferenceScheme does not create the kind or its difference.
SLO compliance labels
Candidate family: Compliant, AtRisk, Grace, Breached, Waived.
Result:
- These are not system-role-kind names.
- F.10 recovers status family, status value, status window, confidence, or deontic or policy use.
- Presentation labels may stay local or be named by the direct status pattern. They do not become a system-role kind,
SystemRoleKindDescription, or relation structure among system-role kinds.
Evidence and requirement suffixes
Candidate family: EvidenceRole, RequirementRole, StandardRole, SourceRole.
Result:
- No work-facing system-role kind is recovered from suffix alone.
- Route each recovered claim separately: bounded evidence reliance goes to A.10; an actual named assurance claim goes to B.3; description-use ambiguity goes to E.10.D2 for recovery only; and publication occurrence, form, or carrier goes to E.24.PUB. A requirement, standard, source, access, or policy use goes to the pattern that directly defines, constrains, or tests that exact claim. If none exists, return
missing-governor. - A durable name may be admitted for the recovered relation, but not as a local system-role kind.
Same spelling across two local-sense bases
A plant team uses Operator for one local system-role kind. An access-control team uses Operator for one permission grouping. Recover both independently under their direct patterns; neither spelling nor organizational proximity makes them one value.
For local use, keep the existing expressions and stop. If a cross-local naming use needs a semantic correspondence, resolve the two exact F.17 SchemeSenseCell endpoints and test the proposed F.9 relation. Cite a Bridge only when its predicate obtains, then state the use direction, rule, tolerated loss, polarity, and reliance separately.
A Bridge, NameCard, cell, or row imports no access permission as U.SystemRoleAssignment, capability, authority, or performed Work. Publish an F.17 row only when the public or durable reuse threshold independently holds.
Ordinary composite role-like phrases
A project says: "Vasya is an engineer, he works on musical robots, and he is also a musician who teaches robots to play music."
Result:
-
Ordinary prose may remain
robotics engineer and musicianorengineer-musicianwhen the sentence is clear and no FPF claim relies on either noun as an exact classification. Create no Tech kind merely to explain the phrase, and do not require aSystemRolesuffix in ordinary prose. -
If the sentence supports a load-bearing FPF claim, apply E.10.ROLE and recover only the supported branch: for example, a local system-role kind and classification, an assignment occurrence, a capability, a participation or contribution relation, a Method or Work claim, or a finding that no pattern yet defines the needed claim. Do not infer two kinds from the two nouns.
-
Any claim about an engineering or music-teaching Method, robot-training Work, or performed music Work stays under its direct pattern and remains separate from the ordinary phrase. Such a claim does not by itself justify a system-role-kind name.
-
A durable qualified system-role-kind name becomes a candidate only after A.2 with C.3 independently admits that exact local kind through its candidate domain, operative work-facing membership condition, member/non-member boundary, and continuity rule.
Differences in, for example, assignment conditions, capability expectations, incompatibilities, or Method or Work implications matter only when the
KindSignatureor named use draws on them. A readable suffix does not perform that admission.
Anti-patterns and repairs
Conformance checklist
Regression checks
Reopen only the affected naming use when candidate expressions grow faster than recovered values, when a name starts carrying assignment, capability, method, Work, evidence, status, source, publication, equivalence, or authority, when a row is reused beyond its admitted use, when local wording is silently globalized, or when one naming object begins to imply the next. A changed spelling alone does not require a new governed value or full family replay.
Relations
-
A.2, A.2.1, A.2.5, and A.2.7 define or constrain system-role kinds, assignments, assignment-state predicates and direct state relations, and relations among system-role kinds.
For precise performed Work, A.13 first recovers each exact actual performer and A.15.1 independently admits the dated occurrence; F.6 defines only the later assignment-bound attribution when that relation is expressly consumed. F.14 only blocks names that hide these distinctions.
-
Use F.8 to make one candidate's smallest mint-or-reuse disposition after the F.14 stop test.
-
F.9 defines only an actual relation between exact local senses. Shared spelling and cell presence establish none.
-
F.17 defines the public term-row form and its entry threshold; F.18 defines the durable naming-settlement NameCard form; neither defines the governed value.
-
Use C.2.1 to identify every persisted NameCard, row, or control-record episteme and its
EpistemeEditionRelation; use E.24.PUB to state row publication occurrence, particular expression, any selected reusable form and its conformance, and actual carrier bearing. -
Use F.10, A.10, B.3, E.10.D2, and the direct policy, access, and source patterns for the corresponding status, evidence, assurance, description, policy, access, and source claims that often arrive with role-like suffixes.
SoTA-Echoing
F.14 does not import access-control, terminology, credential, or modeling-language taxonomies as FPF ontology. It uses the sources below only where they change the anti-explosion rule.
Didactic distillation
When names multiply, recover the exact values and the proposed use. Keep adequate ordinary wording or an existing designation; compare other treatments when the choice remains unresolved. Keep relations among system-role kinds, status windows, capability, Method, Work, evidence, source, policy and publication under their direct patterns. Create a cell, NameCard, row, identifier or publication only for the named use that needs it.
F.14:End
Static and Regression Conformance Harness for Naming and Semantic Unification
Type: Pattern Status: Stable
"Prove locality and parsimony first; only then prove composition."
Type: Architectural pattern.
Normativity: Normative.
Builds on: F.17 for exact SchemeSenseCell, local-sense basis, and row epistemes; F.18 for naming-settlement NameCard epistemes and selected designation expressions; F.14 and F.8 for anti-explosion and mint-or-reuse decisions; F.13 for lineage; F.9 for actual cross-local Bridge occurrences and separate bounded-use claims; F.4 for system-role-kind-description epistemes; F.10 or the current pattern that defines the status values and windows; C.2.1 for exact claim and record epistemes; A.2.6 for ClaimScope; A.1.1 and A.22 only when a selected bounded-model-use Structure changes the checked use; and E.24.PUB for publication.
Coordinates with: A.6.1 for exact check-application bindings; A.13 for every precise performer's local core, A.15.1 for independent dated assessment-Work admission, and F.6 only for a current precise assignment-bound attribution; A.10 and B.3 for evidence reliance and assurance; G.11 for currentness; A.2, A.2.1, A.2.5, and A.2.7 for system-role kinds, assignments, assignment-state predicates and direct state relations, and relations among system-role kinds; E.17 and E.10.D2 for view, description, and source-use claims; A.6.5 for relation declaration; and the pattern that defines each non-naming object included in the selected slice.
Plain entry cues (informative). Static or regression check over a finite naming slice; selected-name regression; exact before/after naming continuity check.
Intent and applicability
Intent. Give one compact harness for checking whether a finite naming and unification slice is locally sound now and remains sound across identified changes.
F.15 does not define schemes, local senses, cells, values, relation occurrences, descriptions, rows, system-role kinds or assignments, status families, aliases, names, evidence, or publication. Its application checks exact objects already recovered through their defining or testing rules and records result claims without duplicating F.18 naming settlement.
Applicability. Use F.15 when one receiving use depends on several already recovered items: effective ReferenceSchemes, F.17 SchemeSenseCell values, F.18 NameCards and selected designations, F.17 rows, local system-role kinds or status values, actual F.9 Bridge occurrences, or exact prior and later editions. Include a selected bounded-model-use Structure and its description only when that structure's organization changes this check or receiving use.
Primary EntityOfConcern in plain terms. One exact finite slice version under a declared set of static or regression rules for one named receiving use. The checked scope is not evidence, a work process, result, registry, Bridge, system-role assignment, status value, publication, or universal context.
Admissible move in plain terms. Resolve the finite member refs and exact versions and apply only the triggered rules. Identify an exact check application when the claim needs its identity. Recover dated assessment Work when the claim needs its performance history. Constitute each result claim separately under C.2.1. Cite witnesses and evidence relations separately. Use the defining or testing rule for every failed subject claim, with its PatternID retained only as a locator.
Primary working reader. A practitioner checking whether selected names and semantic-unification results meet the rules for one stated reuse.
Use this when. A slice feels "almost unified", but one or more of these questions remain:
- Does each local expression resolve under its exact effective ReferenceScheme and local-sense claim?
- Does each
SystemRoleKindDescriptionstill describe its exact local system-role kind without becoming the kind, assignment, or NameCard? - Does each F.17 row still pass the entry and result gate for that row, including the valid one-cell case?
- Does every cited F.9 Bridge obtain between exact cells, with its description/Card and bounded-use claim kept separate?
- Do exact earlier and later values, descriptions, rows, names, relations, and status windows support the stated continuity or change claim for this receiving use?
What goes wrong if missed. Shared spelling globalizes local senses; a table row or NameCard looks like value identity; a Bridge description replaces relation truth; record membership becomes evidence; a check record appears to perform work or emit its own result; and an edition label silently proves sameness or difference.
What this buys. A finite, replayable conformance check: selected names remain tied to exact governed values, cross-local use stays relation- and claim-bound, non-naming claims remain governed by their defining or testing rules, and regression closure states which versions, rules, evidence, losses, and receiving use were checked.
Not this pattern when. Use F.18 to settle a name, F.17 to admit a row, F.9 to establish a Bridge, and the direct subject pattern to decide a system-role-kind, assignment, status or evidence claim. Use F.15 when their already-defined outputs must be checked together.
For ordinary checking, apply the F.15 questions directly. A claim about dated assessment Work or an exact operation application additionally uses A.15.1/A.6.1, and publication of a record uses E.24.PUB.
Recognition versus assurance note. Recognition identifies the exact finite scope, versions, triggered rules, and receiving use. Assurance, when needed, concerns reliance on separately constituted result claims through exact A.10 or B.3 paths. Neither a filled record nor scope membership supplies assurance.
Problem frame
Unification work fails when composition is claimed before local meaning, exact object recovery, and continuity are checked:
- Locality leak. Same spelling is treated as one meaning without comparing exact
<ReferenceScheme, LocalSenseClaim>projections. - Row sprawl. F.17 rows or F.18 NameCards multiply although an existing governed value and admitted naming use already suffice.
- System-role or status inflation. Adjectival, temporal, or source-label variants become new system-role kinds or status values without recovery through the pattern that defines them.
- Silent rewrite. An edition or rename changes claim content while a stable id is treated as continuity proof.
- Bridge hardening. A description, Card,
CL, or earlier relation claim is later used as equivalence or use authority without a current obtaining occurrence and separate bounded-use claim. - Check collapse. Scope, rule, application/work, result claim, witness/evidence path, record episteme, publication, and currentness are treated as one object.
- Register split. Tech and Plain designation expressions drift away from the exact current F.18 NameCard, governed value, or local sense.
F.15 catches these failures before the finite slice is used for naming reuse, cross-local comparison, assurance input, or another downstream claim.
Problem
A slice can look stable because labels, cards, rows, descriptions, relation records, aliases, and version ids are arranged in one table. Yet the table establishes none of its listed subject relations, checks, results, evidence uses, continuity claims, or publication occurrences.
F.15 makes the exact static and before-and-after questions inspectable without defining or establishing the neighbouring naming, ontology, checking, evidence-use, or publication claims itself.
Forces
Solution
The harness has two rule families:
- Static Conformance Rules (SCR). Check exact current object and relation refs in one finite slice version. A rule result is a separately constituted claim, not a field value that becomes true because a record is filled.
- Regression and Stability Conformance Rules (RSCR). Compare exact earlier and later refs for the changed member and reconsider dependent claims or uses whose grounds changed. State the governed continuity or change claim, admitted losses, evidence, and receiving use; changed spelling or edition alone proves neither sameness nor difference.
Both families are F.15-local check declarations over already defined objects. A practitioner may apply their questions and obtain a local result without naming the checking activity as Tech U.Work. An exact rule application, when its identity is needed, uses A.6.1.
If a replayable result or example asserts dated assessment U.Work, recover each actual performer's A.13 core and independently admit the Work under A.15.1. Add F.6 afterward only when the result also needs precise assignment-bound attribution.
A short record may omit an assignment identifier unused by its receiving claim only when every relation it consumes remains recoverable. Name the A.6.1 application and bindings when that application is also asserted.
C.2.1 separately constitutes the result claims and optional conformance-record episteme. A.10 and B.3 supply evidence-reliance and assurance rules. E.24.PUB governs publication, and G.11 governs currentness.
Minimal vocabulary
- Finite harness scope - an F.15-local by-value selection of exact current refs, versions, triggered rules, and one receiving use; not a U-kind, relation, evidence set, or selected Structure by default.
- Static Conformance Rule (SCR) - an F.15-local declared predicate over exact current inputs.
- Regression and Stability Conformance Rule (RSCR) - an F.15-local declared predicate over exact earlier/later inputs plus the continuity or change claim and receiving use.
- Check application - an actual A.6.1 operation application with exact rule and object bindings, when current.
- Dated assessment Work - a specific
U.Workoccurrence used only for a replayable performance claim. Each performer must already have the A.13 core and the Work must already be independently admitted under A.15.1. F.6 is additionally required only when the receiving claim needs precise assignment-bound attribution. - Result claim - one C.2.1 episteme asserting
pass,fail, orundeterminedfor one exact rule application, scope version, and use; not a general status value. - Witness - an exact example, counterexample, invariant, trace, or edition note cited by the result claim; its presence is not the result or an evidence-use relation.
- Conformance record - an optional C.2.1 episteme that packages refs to the scope, applications/work, result claims, witnesses/evidence paths, non-admitted uses, and reopen conditions; it performs no check.
- Changed member - one exact prior/later pair whose governed identity, relation truth, description, designation, status use, or publication availability may affect the receiving use.
Objects under check
F.15 lets a practitioner check the following objects together. Each keeps its own definition:
- effective reference scheme values and exact prior/later editions;
- independently governed local-sense claims and F.17
SchemeSenseCellcoordinates; - exact governed values and relation occurrences together with the rules that identify each value or say when each relation obtains, and the PatternIDs that locate those rules;
- F.4
SystemRoleKindDescriptionepistemes and their exact local system-role kinds; - F.18 NameCard epistemes, selected Tech/Plain designations, aliases, and lineage;
- F.17 UnifiedTermRow epistemes and exact row editions, including admissible one-cell rows;
- actual F.9 Bridge occurrences, with Bridge descriptions or Cards referenced separately when current;
- status families, values, targets, scopes, windows, source conditions, and uses recovered through F.10 or another applicable status rule;
- selected bounded-model-use Structures and their separate descriptions only when structural organization changes the checked use;
- exact source, evidence, currentness, and publication relation occurrences needed by the result's receiving use.
A description, Card, row, label, shared table, stable id, selected scope, or earlier pass makes none of these subject relations obtain and grants no continuity, equivalence, conformance, authority, system-role kind or assignment, status, or evidence use.
Finite scope and conformance record
Declare the finite scope before applying a rule:
SelectedStructureRefs is empty unless an independently selected A.1.1/A.22 structure changes interpretation for the receiving use. A Structure description never replaces the Structure, its obtaining membership relations, or another scope member.
Use an optional record only to package already identified neighbors:
The checked scope, rule declaration, ordinary checking action or admitted dated assessment Work, exact application, result claim, witness, A.10 evidence-provenance path, conformance-record episteme, E.24.PUB occurrence, publication form, carrier, and G.11 currentness relation remain distinct.
A result ref is included only after its C.2.1 claim exists.
The optional record may cite an already admitted Work ref; it does not restate the Work's performer, Method, assignment, time, or containing System. Publication and currentness refs are neighbouring claims, not record identity shortcuts.
Static conformance rules for local material
SCR-F15-S1 (Finite exact scope). Every selected member resolves to one exact governed value, occurrence, episteme, or by-value scheme at one exact version; the receiving use and triggered rule refs are explicit. Scope membership is selection, not evidence or conformance.
SCR-F15-S2 (Local-sense basis currentness).
Each relied-on local-sense claim names its effective ReferenceScheme and exact expression. If a LocalSenseBasisRelation is cited, resolve its exact occurrence under F.17 and any description as a separate episteme. A source title, carrier, NameCard, or row does not replace the relation.
SCR-F15-S3 (SchemeSenseCell identity).
Each cell is the exact F.17 value <ReferenceScheme by value, LocalExpression, LocalSenseClaim>. No cross-local items, description fields, source labels, or selected Structures are merged into one cell.
SCR-F15-S4 (Two selected registers). When Tech and Plain designations are current, both are the exact expressions selected by the same current F.18 NameCard for the same governed value and admitted use. Register difference does not create another value or sense.
SCR-F15-S5 (Minimal gloss). A local gloss states the needed sense and includes a blocked use only when F.19's grounded-contribution test requires it. It does not smuggle behavior, permission, evidence, source authority, publication status, global sameness, or a check result.
SCR-F15-S6 (Local reuse before Bridge).
Another expression under the same <ReferenceScheme, LocalSenseClaim> projection is a designation or alias question. Different projections open the F.9 question only when a named semantic-correspondence use is current; scheme difference alone proves no Bridge.
Static conformance rules for composed material
SCR-F15-S7 (SystemRoleKindDescription boundary).
An F.4 SystemRoleKindDescription is one C.2.1 episteme about one exact local system-role kind under one effective ReferenceScheme. It makes the C.3 candidate domain, operative membership condition, intended member/non-member boundary, continuity rule, and current KindSignature recoverable.
Practice or source provenance may locate the definition but does not identify the kind. The description is not the kind, NameCard, SchemeSenseCell, assignment, status, evidence template, method, or work; a cell is cited only when the naming use needs one.
SCR-F15-S8 (Name discipline without F.18 duplication). Every candidate or selected name cites the recovered governed value and the pattern containing its defining or constraining rule. Apply the F.14 and F.8 criteria to decide whether naming work continues. If the naming use requires a durable settlement, check the NameCard and designations established under F.18. If a term row is claimed, check its admission under F.17. Name selection remains with the naming decision.
SCR-F15-S9 (F.17 row truth).
Each cited row is one exact F.17 UnifiedTermRow episteme that records one value, its direct kind, the locator where that kind or value is defined, its NameCard, selected designations, effective scheme, one or more exact SchemeSenseCell refs, admitted uses, any grounded blocked uses required by F.19, and reopen condition. One cell is valid when the row use is not cross-local; a row-shaped local note or table position is not a row episteme.
SCR-F15-S10 (Cell and neighbor purity). Each row cell remains an exact SchemeSenseCell. NameCard, local-sense basis relation, Bridge, Bridge description/Card, selected Structure, source publication, row id, and carrier remain separate refs and substitute for no cell component.
SCR-F15-S11 (Reuse before minting). When an existing NameCard or row supports the same governed value and admitted use, reuse it or record the exact F.8 decision that justifies another naming settlement. A new label, table, project, or edition is not a visible value difference.
SCR-F15-S12 (Actual Bridge before Bridge use). A cited F.9 Bridge has two exact endpoint cells, one exact relation-semantic profile, a currently true kind-defined predicate, and all required dependencies. Its assertion/description episteme and optional Card remain separate.
A separate C.2.1 claim affirms or rejects the proposed use of that Bridge in the stated direction, under the stated correspondence rule and loss tolerance. Use A.10 for ordinary reliance on the claim. Use B.3 when the receiving use needs a named assurance claim.
For a use through several Bridges, follow F.9:14.7. State the direction of use at each link, the rule for using them together and the cumulative effect on the receiver’s decision. Two signed shifts of +0.4 satisfy a local tolerance of 0.5 each and exceed a final tolerance of 0.5 together. Established cancellation may meet the final tolerance. The chain can thus support a particular use. A claim that a direct A-to-C Bridge obtains has its own predicate test.
SCR-F15-S13 (Cross-local locality).
Use F.9 only for different <ReferenceScheme, LocalSenseClaim> projections and one named current correspondence use. Same-projection expression reuse stays with designation; different projections do not themselves establish a relation; when no current correspondence use exists, add no Bridge or bounded-use claim.
SCR-F15-S14 (Status basis). A status-shaped item resolves through F.10 or another applicable status rule to the exact family and value definitions, target, scope, window, source condition, and intended use. Adjective, time, scale, phase, confidence, row presence, or display label creates no status family, value, assurance, gate decision, or evidence use.
SCR-F15-S15 (System-role-kind relation preservation).
Every exact incompatibility, monotonic kind order, residual qualification, bundle, requirement, or selected SystemRoleKindRelationStructure remains an independently identified relation occurrence or selected structure. A description or convenient fused name creates neither another system-role kind nor an assignment or performed Work.
SCR-F15-S16 (Rule and locator boundary for non-naming claims). Assignment, work, result, evidence, source, publication, currentness, assurance, gate, decision, method, capability, policy, structure, and subject-relation claims cite the rule that defines or tests each exact claim and the PatternID that locates it. When a rule fails, re-evaluate that subject claim under the rule; an F.15 result neither decides nor absorbs the claim.
SCR-F15-S17 (Public naming and publication separation). Public or Core-facing naming cites an exact F.17 row only after its current gate passed. Row currentness is not availability: E.24.PUB separately governs any publication occurrence, form, carrier, audience, and bounded use, and rendering/upload work remains separate.
Twin-register checks
Use these checks when the F.18 naming result records both a Tech and a Plain designation.
SCR-F15-T1 (Same exact settlement). Both expressions resolve through the same current NameCard to the same governed value, effective scheme, local-sense claim, and admitted naming use. The NameCard, expressions, value, and any F.17 cell remain distinct.
SCR-F15-T2 (Same governed kind). The Plain expression does not suggest a different kind, relation truth, system-role kind or assignment, status, work, evidence, or permission from the Tech expression's exact governed object.
SCR-F15-T3 (Ambiguous head guarded). A high-risk Plain head receives a kind head or short recognition gloss at first use without turning the gloss into a second selected designation.
SCR-F15-T4 (No normative displacement). Reader-facing Plain wording does not silently replace the selected Tech designation in normative Core claims; both remain expressions, not the governed value.
SCR-F15-T5 (Projection-aware reuse).
Same-projection reuse is a designation/alias question. A named reuse between different <ReferenceScheme, LocalSenseClaim> projections cites an obtaining F.9 Bridge, a separate affirmative bounded-use claim, and current A.10 or B.3 reliance. A public row, copied label, Card, or earlier pass supplies none of those premises.
Regression and stability rules
The RSCR family compares exact earlier and later refs for each changed member and rechecks dependent results whose grounds changed. Every result names the continuity or change proposition, admitted losses, receiving use, and evidence path. It does not infer identity or difference from spelling, path, stable id, table position, timestamp, or edition label.
Schemes, versions, and known confusions
RSCR-F15-E1 (Exact before/after and no silent replacement).
For each changed member, resolve exact @t0 and @t1 refs and versions. A changed effective ReferenceScheme changes interpretation-bearing content; an unchanged label or shared designator does not prove continuity. State the exact identity, continuity, split, retirement, or replacement claim and cite the rule that defines or tests it.
RSCR-F15-E2 (Known confusion check). For each prior confusion, blocked use, and nearest counterexample affected by the change, recheck it or explicitly retire it. A new edition does not erase an old trap.
Local senses and SchemeSenseCells
RSCR-F15-E3 (Reconstructible local sense).
When the basis episteme, source unit, or attestation changes, the @t1 local-sense claim remains recoverable from exact current basis relations and descriptions. Changed witnesses or source publication do not silently rewrite the sense claim.
RSCR-F15-E4 (SchemeSenseCell value identity).
The exact F.17 cell value is <ReferenceScheme by value, LocalExpression, LocalSenseClaim>. Changing any component yields another coordinate value; keeping a label or id does not preserve it. Same sense under a renamed expression is handled through designation/lineage; it does not preserve the cell.
UnifiedTermRows
RSCR-F15-E5 (Row episteme identity and edition). Compare the exact C.2.1 row epistemes and their ClaimGraphs, EntityOfConcern values, and effective schemes. Changed governed value, NameCard, selected designation, cell, Bridge ref, admitted use, or rationale creates the corresponding later row claim content; an edition id cannot hide it.
RSCR-F15-E6 (State the row change). When a changed value, sense, or use alters row support, preserve the exact earlier row and state the warranted later disposition, admitted losses, and receiving use under F.13/F.17. A one-to-one change uses its subject-governed revision or replacement; split or merge only when those relations hold. Do not mutate a shared table cell as continuity proof.
SystemRoleKindDescriptions and names
RSCR-F15-E7 (SystemRoleKindDescription continuity).
Compare exact F.4 description epistemes and the described kinds' candidate domains, operative membership conditions, intended member/non-member boundaries, continuity rules, current KindSignature editions, effective schemes, and claim content. Source or practice provenance is a cue to compare those definitions, not an identity key. A label-only change cannot prove that the described kind or description episteme stayed the same.
RSCR-F15-E8 (Alias for expression change; direct recovery for meaning change). If only a selected expression changes while the exact value, scheme, sense and use are preserved, F.13 and F.18 may record an alias or rename.
When a described kind, candidate domain, membership distinction, boundary, continuity rule, scheme, local sense or description claim changes, recover the affected object or episteme under its subject rule. Reopen the naming settlement when that change affects the named value, interpretation or receiving use; reuse an unaffected settlement. A practice or source change by itself triggers comparison and proves neither continuity nor a split.
Bridges and bounded uses
RSCR-F15-E9 (Exact Bridge change). Compare exact prior/later endpoint cells and relation-semantic profiles. A changed endpoint or profile concerns another Bridge candidate and obtaining test; reconsider every bounded-use claim that depends on that Bridge. Changed assertion, description, Card, evidence, reliance, or bounded-use claim does not by itself reidentify or negate a fixed obtaining occurrence.
RSCR-F15-E10 (No drift to equivalence or use authority).
A later equivalence claim requires an exact Equivalence profile, true predicate, required dependencies, and a separately identified obtaining occurrence. A new witness set, high CL, Card, or earlier partial relation is insufficient. Any later substitution still needs its own bounded-use claim and reliance.
Status and system-role-kind relation structure
RSCR-F15-E11 (Status-window and status-use stability).
Compare the exact status family and value definitions, target, scope, window, source condition, and intended use at @t0 and @t1. Changed time, scale, confidence, or edition does not create a new family or preserve an old result automatically.
RSCR-F15-E12 (System-role-kind relation stability).
Preserve, retire, or restate each exact incompatibility, monotonic kind order, residual qualification, bundle, requirement, or selected SystemRoleKindRelationStructure before using it in a naming, assignment, or Work claim. No later description or fused label substitutes for the relation occurrence.
Public naming, publication, and currentness
RSCR-F15-E13 (Public name continuity). F.13/F.18 record the exact selected-expression lineage and NameCard change; F.17 separately records the later row episteme and admitted use. E.24.PUB publication occurrence/form/carrier and G.11 currentness are rechecked only when their exact refs or receiving use changed. A local rename, row edition, or upload does not prove public-name continuity or publication.
Reasoning primitives
staticSliceOK(...) may be asserted only as a C.2.1 summary claim over those exact positive results. Scope membership, a filled record, or an absent failure row does not establish it.
changedSliceOK(...) may summarize only exact current changed-member results that pass for the named receiving use; fail or undetermined leaves the affected use unadmitted.
Reuse prior results for unchanged members after a direct contradiction check. One changed member does not trigger a full-slice rerun unless its dependencies invalidate the other results.
An F.15 result may report the failed check. Writing another record field neither repairs nor decides the subject claim.
The Bridge, use claim, evidence/reliance, authorization, and any receiving occurrence remain separate. CL, a Card, or record membership is not a use result.
Archetypal Grounding - worked cases
Activity and task under two run schemes
The slice resolves activity under PROVORunScheme-2026 and task under IEC61131RunScheme-2026 as two exact F.17 SchemeSenseCells. A named comparison use is current.
F.15 result:
- SCR-F15-S3 checks each exact triple; shared run-language does not merge them.
- SCR-F15-S12 requires an obtaining F.9 occurrence before the comparison uses a semantic relation. Its Card is optional and its bounded-use claim is separate. When the claim concerns use of the whole chain, check the combined rule and tolerance. A commission confined to the local checks can finish there.
- Any F.17 row must pass its own gate. It may contain the exact cells needed by the row use; table shape does not create the row.
- An
ExecutionSystemRoleKindDescriptionremains an F.4 episteme about one exact localExecutionSystemRoleunder one scheme; it does not describe both cells, assign a system, or prove work. - If a later
tasksense becomes cyclic while theactivitysense remains non-periodic, RSCR-F15-E4 and E9 compare exact later cells and Bridge candidates; evidence may change the use claim or reliance without silently rewriting the prior Bridge.
Suppose CheckRun-17 is dated assessment U.Work, CheckMethod-17 is its semantic U.Method, CheckInterval-17 is the Work interval, and HarnessSystem-17 is the containing System. Evaluator-17 is the admitted U.System that performs the Work using that Method during CheckInterval-17.
First recover Evaluator-17's A.13 core for this action, including declared assignment species EvaluatorAssignmentSpecies-17 and one obtaining occurrence EvaluatorAssignment-17 with every required participant value, Evaluator-17 as holder, and interval coverage.
A.15.1 then independently admits CheckRun-17 from its performance history, enacted Method, interval, and containing-System relation. Because this example also claims performance under EvaluatorAssignment-17, F.6 afterward links the already admitted Work to that same assignment.
ApplySCR-S12-17 is the exact A.6.1 rule application and bindings. BridgeRuleResult-17 is a separate C.2.1 result claim; WitnessTrace-17 and its A.10 path are separate again. The record UnificationConformanceRecord-17 cites these independently admitted items. Publishing the record requires an E.24.PUB publication occurrence with a particular expression borne by a carrier. If a reusable publication form is selected, check that expression’s conformance to it.
Service availability across service and observation schemes
The slice contains one service-management status value/use and one uptime-observation claim under different effective schemes, plus exact cells only for the naming use that addresses them.
F.15 result:
- SCR-F15-S14 requires F.10 for the status family/value, target, scope, window, source condition, and intended use, or the exact defining or testing rule for the current status claim.
- If the comparison uses a semantic correspondence, check it under SCR-F15-S12 and S13. To use the uptime measurement to judge service fulfilment, check the applicable measurement and acceptance rules. A shared label or row establishes neither claim.
- Observation evidence and A.10 reliance are not the status value, comparison result, assurance claim, or F.15 result.
- Use B.3 only when an actual named assurance claim is current. If a direct rule requires that claim for this use, recover it first; materiality or slice membership alone establishes no assurance claim.
Rename a SystemRoleKindDescription without changing the described kind
IncidentReviewerSystemRoleKindDescription@t0 and ServiceIncidentReviewerSystemRoleKindDescription@t1 describe the same exact IncidentReviewerSystemRole only if F.4's candidate domain, operative membership condition, intended member/non-member boundary, continuity rule, current KindSignature, effective scheme, and description claims support that continuity. A changed source, practice, or name alone decides neither sameness nor difference.
F.15 result:
- RSCR-F15-E7 compares the two exact description epistemes and the described local system-role kind.
- RSCR-F15-E8 permits F.13/F.18 alias or rename treatment only for expression change with value, scheme, sense, and use preserved.
- F.18 updates the NameCard; F.17 updates a public row only if that row use is current and its gate passes.
- If the described system-role kind or description claim changed, F.4 and C.2.1 identify the affected value or episteme; F.18 rechecks naming when the change affects its value, interpretation or use. F.15 does not infer continuity from the label.
Partial Bridge later claimed as equivalence
An exact Partial-overlap Bridge once obtained between an OWL subclass sense and an FCA order-edge sense. A later formal result claims equivalence inside one constrained fragment.
F.15 result:
- RSCR-F15-E9 keeps the prior occurrence fixed and identifies the exact later endpoint/profile candidate.
- RSCR-F15-E10 requires the Equivalence predicate and dependencies to be true for a separately identified occurrence; new witnesses or
CLdo not suffice. - The constrained-fragment substitution is a separate bounded-use claim with its own rule, tolerance, polarity, and reliance.
- C.29 governs the mathematical-lens claim; F.15 checks that no description, Card, or result label silently strengthens the relation.
Peak-hours status proposal
A team proposes PeakHoursAvailabilityStatus as a new family because one existing status is used in another time window.
F.15 result:
- SCR-F15-S14 fails if F.10 or the applicable status rule shows only a changed window or use.
- RSCR-F15-E11 compares the exact family/value, target, scope, window, source condition, and use rather than the suffix.
- Use F.10 or the applicable status pattern for the status claim; F.14/F.8/F.18 block a new durable name until a distinct governed value is independently recovered.
Bias-Annotation
F.15 counters unification bias by comparing definitions, local senses and continuity grounds. Shared spelling, table membership, a stable id, an earlier pass, a Bridge description or a NameCard does not establish common meaning or continuity.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Closure conditions
A finite slice is locally admissible for its named receiving use only when:
- every scope member and exact version resolves under its identity rule and PatternID locator;
- every triggered static rule has an exact current C.2.1
passresult for the named receiving use; afailorundeterminedresult leaves the affected use unadmitted; - every changed member has an exact prior/later pair and a current passing RSCR result supporting the stated continuity/change, admitted losses and receiving use; dependent results whose grounds changed have also been rechecked;
- every failed subject claim needed by the use is repaired and re-evaluated under its defining or testing rule, and the affected F.15 rule then passes before reuse;
- witness refs and any relied-on A.10/B.3 path are current for the exact result and use, without becoming the result;
- the optional record cites, but does not replace, applications/work, result claims, evidence, Bridge occurrences, descriptions, publication, or currentness;
- a non-admitted use is explicit when its rejected reading has an independent local ground and the boundary changes the intended reader's interpretation or action under F.19; and
- the closure statement names the exact slice versions, rule set, currentness basis, and receiving use.
Closure is local. A later change reopens only the affected rule results and their dependents after contradiction checks. It does not authorize a full rerun by habit or a global claim that all names, rows, relations, evidence, and publications conform.
Consequences
Benefits. F.15 makes interpretation locality, exact naming settlement, Bridge truth, check execution, result identity, and edition continuity visible before reuse. The applicable patterns retain their definitions while the finite slice gains one replayable check surface.
Costs. A slice that looks unified by spelling or table shape may remain open until exact object refs, rule applications, result claims, evidence paths, and prior/later continuity claims are recoverable. The harness limits this cost by triggering only relevant rules and reusing unaffected results after contradiction checks.
Failure avoided. F.15 prevents row-, card-, or record-shaped notes, alias-only rewrites, Bridge optimism, system-role and status inflation, evidence collapse, and publication or currentness labels from becoming hidden global meanings or conformance authority.
Rationale
Cross-local reuse depends on the exact local meanings and supported relation. Regression compares the relevant earlier and later values for a named use. F.15 brings those checks together in a finite slice and returns unresolved subject questions to the patterns that can settle them.
SoTA-Echoing
Currentness rule: when the defining or testing rule for a value—or F.17/F.18, F.9, C.2.1, A.10/B.3, A.15.1/A.6.1, G.11, or E.24.PUB—changes an exact input, relation, result, evidence, or receiving-use boundary, reopen only affected SCR/RSCR results and their dependents. A label, carrier, record layout, or unrelated edition change does not reopen the whole slice.
Relations
-
F.17 and F.18. Supply exact scheme-based cells, basis relations/descriptions, NameCards, selected designations, rows, and editions. In an F.15 check, verify those values without selecting or publishing a name.
-
F.14, F.8, and F.13. Govern anti-explosion, mint-or-reuse decisions, and lineage before F.15 checks the resulting exact refs.
-
F.4 and exact system-role patterns. Define system-role-kind-description epistemes, local system-role kinds, relations among them, assignments, and work claims that the harness cannot absorb.
-
F.9, C.2.1, A.10, and B.3. Govern actual Bridge occurrences, separate bounded-use claims, evidence reliance, and assurance. Descriptions, Cards,
CL, and witnesses are not relation truth or use authority. -
F.10 or the applicable status pattern. Use it for status family, value, target, scope, window, source, and use claims.
-
A.1.1 and A.22. Supply an optional independently selected bounded-model-use Structure only when its organization changes the checked use; description and membership remain separate.
-
A.13, A.15.1, F.6, and A.6.1. A.13 recovers each exact actual performer and A.15.1 independently admits dated Work; F.6 adds only an expressly consumed precise assignment-bound attribution through the same obtaining A.13 assignment, and A.6.1 governs exact rule application.
Ordinary checking need not be admitted as
U.Work, and missing or failed F.6 alone leaves independently admitted Work intact. Recheck any admission premise invalidated by the same finding, with its dependent conclusions, under F.6:4.2. -
E.24.PUB and G.11. Govern publication occurrence/form/carrier and currentness separately from the checked record.
-
C.34. Supplies architecture-specific preservation or equivalence adequacy when exact selected architecture structures and losses are the live subject; F.15 carries only the finite regression check and result refs.
Didactic distillation
Use F.15 as a small check over exact already-governed objects. First pin the finite scope, versions, rules, and receiving use. Then check locality and naming: schemes and cells are exact, the F.18 result records the selected names, an F.17 result records any admitted row, and actual Bridges remain separate from Cards and use claims. Next check execution and result: an application or dated Work is not its C.2.1 result, witnesses are not evidence use, and a record does not perform or publish anything.
For change, compare exact prior/later refs and state continuity, loss, and use. When a rule fails, re-evaluate that subject claim under its defining or testing rule; do not patch the label or record field.
F.15:End
Worked-Example Template (Cross-Domain)
“Show one claim, the actual values and relations that make it true or false, and enough evidence for a reader to replay the example.” Type: Architectural pattern.
Status: Stable
Builds on: E.10.D1 Recovering What “Context” Means in Use; F.1 for the source cut; F.0.1, F.2, F.3, and F.17 for exact source-local meaning when needed; F.7 for an optional comparison surface; F.9 only for actual semantic relations and bounded-use claims; F.10 for windows; F.15 for checks. Coordinates with. F.12 when a case evaluates promise acceptance; A.10 for evidence use; B.3 only for an actual named assurance claim; C.16.P and A.6.RCD when indicator or proxy wording hides a missing relation; E.13 only for optimized or decision-driving proxies; the direct Part C pattern for every illustrated subject; and the A.3, A.15, A.6.1, and B.1.5 method-and-work stack.
Intent & applicability
Intent. Provide a one-page worked-example shape that starts with a recognizable question and practical gain, names the actual values and obtaining relations, exposes exact sources and evidence, and leaves a clear boundary.
Optional lexical cells and comparison tables help navigation only when they reduce reader effort.
Use this when. A pattern needs a compact example that crosses disciplines, source schemes, design-time and run-time boundaries, or several relation families and prose alone would hide the hand-offs.
Do not use this when. A shorter direct example already shows the claim. F.16 does not require a cross-source relation, table row, lexical cell, system-role claim, window, or SoD unless the case contains one.
Scope. The template organizes the explanation presented to the reader.
Problem frame
Cross-domain examples fail when:
- The claim is missing. Sources and facts are listed, but the reader cannot tell what is being demonstrated.
- Words replace subjects. Process, role, service, or execution is used without identifying which value or relation the word denotes.
- Relations hide in prose. “Basically the same”, “implements”, “uses”, or “governs” conceals several different relation families.
- A table is asked to prove. Co-placement or row membership becomes sameness, permission, or evidence.
- Evidence and limits disappear. The page reaches a result or status without showing the evaluation, observation, source, loss, or boundary that supports it.
Core idea (didactic)
A useful worked example is a compact case with nine visible parts:
- Question and gain — the recognizable practical problem and what changes when it is solved.
- Claim — one sentence that can be accepted, rejected, or qualified.
- Actual subjects — the values and entities to which the worked claim applies, with their relevant kinds and relation positions.
- Direct patterns — the patterns that define, constrain, or test each substantive value or relation.
- Source basis — exact sources and editions and the passages or interpretation schemes that matter.
- Obtaining relations — each named directly, with participants, direction, and limits.
- Evidence, result, and reliance — what supports the claim, what the evaluation returned, if one occurred, and what remains uncertain.
- Optional aids — F.17 cells when a local meaning needs a stable address and an F.7 table when comparison needs a display; their presence establishes no fact.
- Boundary and checks — where the example stops and the few checks a reader can replay.
Minimal vocabulary
- Worked claim — the exact proposition demonstrated by the page.
- Actual subject — the value or entity to which a substantive claim applies.
- Obtaining relation — a relation supported for these exact participants; its appearance on the page does not create it.
- Source basis — exact source and edition, passage, and effective reference scheme needed by the claim.
- Evidence basis — the observations, source claims and independently established direct relations supporting the conclusion, represented in A.10's descriptive evidence-provenance account with its bounded-reliance disposition. A separately current B.3 assurance claim retains its own result and limits.
- Evaluation result — when the case evaluates a promise or criterion, the result of applying its rule to the selected inputs, on the result scale that rule declares. Recover an exact A.6.1 application or dated evaluation Work when the case claims it or the governing evaluation rule requires it.
- Receiving use — the concrete decision, explanation, comparison, or action for which the result is used.
- SchemeSenseCell — the exact F.17 coordinate
<ReferenceScheme by value, LocalExpression, LocalSenseClaim>. A separate address resolves that coordinate when the receiving use needs it. - Comparison table — an optional F.7 display of exact entries and already obtaining relations.
- Window and separation of duties — F.10 or F.14 constraints included only when time, population, phase, or separation of duties changes the case.
The one-page Worked-Example Canvas
Each item is a thought the reader needs, not a mandatory form field.
-
Title, working situation, and gain. Name the situation in ordinary language and say what practical error or decision the example resolves.
-
Worked claim. State the decision the reader can make in this case. If the page teaches a method for other cases, give the operation and the grounds for its stated range. You can do this within the story. When you have only a particular solution, use C.39.RO to distinguish the case values from the conditions the solution depends on, construct the operation for other admissible inputs, and establish where it works.
-
Actual subjects. Name the values and entities the claim concerns, distinguishing its subject, relation participants and result.
-
Direct pattern route. Beside each substantive claim, cite the pattern that defines, constrains, or tests the value or relation used by the claim. This prevents one vague word or generic Bridge from taking over several relation families.
-
Exact source basis. Cite the source and edition and, where local wording matters, the expression, claim, and effective scheme. Add an F.17 cell only if the exact local meaning needs a stable address for this use.
-
Obtaining relations. Name the participants and state how they are related, including the orientation of a directed relation. Identify the inference the reader can draw in the intended direction of use. Show what supports that inference and which information, if lost in a transfer, could change the receiving action. Use F.9 when the example tests a proposed relation between local meanings with different interpretation bases. Assert a Bridge only if its predicate holds.
Use the defining or testing pattern for MethodDescription membership, enactment, performed Work, measurement, evidence, fulfilment, kind, assignment, publication, transformation, and any load-bearing indicator relation. Use C.16.P to recover the proposed indicator relation. If no rule defines or tests it, return A.6.RCD
missing-governor. When a rule exists, state what the case evidence supports under it. -
Evidence, result, and limits. Show the independently established direct relations by which observations or sources support the claim. A.10 describes that path and classifies reliance on it for the bounded use; the cited basis must meet the evidence requirement of the claim and its direct rule.
When evaluation is part of the case, show the selected inputs, evaluation rule, result and any declared result scale. Recover the evaluation Work, enacted Method and exact application and result binding when the case claims them or its governing rule requires them, including the performance-history branch of A.2.3/F.12 promise evaluation.
Add F.10 status and a durable C.2.1 verdict assertion only for their selected uses. Use B.3 only for an actual named assurance claim. State the consequential limits of the conclusion.
-
Optional comparison surface. If two or more local meanings or source claims are hard to compare in prose, add one compact F.7 table. State the receiving-use conclusion separately. Omit the table when it adds ceremony but no clarity.
-
Micro-narrative and checks. Walk from the working situation through the relations to the result. For a transferable method, show which of its decisions the example illustrates. If independent choice is being assessed, establish what the reader chose and what the task, a diagram or a helper already supplied (E.8:4.2.1.3). A prior independent performance may suffice. End with the subject checks that let the reader replay the claim. Add a non-use boundary when F.19 justifies it, and F.15 when a finite naming slice needs its checks.
When a case needs several pages. Keep the decision the reader came to make. A request about one case can finish with its supported answer. If one method is meant to work across a range, identify the conditions and the operation that remain valid as case values change. A derivation for arbitrary admissible inputs may fit on one page. If several pages establish results about different parts, kinds of holons or model abstraction levels, show where those results enter the same decision and why they hold together. Check shared inputs, handovers or other dependencies that can change it. A claim about each member, a claim about a relation between members and a claim about the whole can require different grounds (C.39.QR and C.37.MM). If the working example has displaced the argument for the range, use C.39.GA to recover and repair that argument.
Invariants
- Claim first. The reader can identify the worked claim and its practical use. For a method to be reused, the reader can recover what to do when case values change, which premises must still hold and why they suffice.
- Actual subjects. World, claim, evidence, status, and use assertions attach to actual values, not lexical cells or rows.
- Direct relations. Every substantive relation cites the pattern that defines or tests it; F.9 is conditional on a real local-meaning relation, and the word proxy never supplies an indicator relation.
- Optional aids. F.17 cells and F.7 tables appear only when they reduce reader effort; their presence or layout does not supply evidence.
- Source precision. Source and edition and the effective scheme are explicit where wording or interpretation changes the claim.
- Time and object distinctions. MethodDescription, Work, observation, and output remain distinct; windows are stated when they change the result.
- Agency precision. Every asserted dated Work uses the performer's recovered A.13 core and independent A.15.1 admission. The core includes its obtaining assignment; add F.6 only when the worked claim needs precise assignment-bound attribution. Recover a separately needed local-kind or description claim under its own rule.
- Evidence, evaluation, and boundary. The page shows why the claim is supported. An evaluation retains its inputs, rule, result and applicable scale; any claimed or required evaluation Work, exact application, status and durable verdict assertion remain distinct. State the consequential limits of the result.
- Didactic parsimony. Keep what the reader needs for the worked answer or the promised method. A condition may leave the answer in this case unchanged and determine whether the method works elsewhere.
Optional comparison panel
When a table helps, use a small F.7 panel:
The panel may also be a two-source contrast with no relation asserted. A source name becomes a column only when the comparison benefits from it. A row does not make entries the same, and the page does not compute a row-wide permission or assurance score.
Worked micro-example
A complete constructed availability case
Question and gain. Does the supplied June observation support the service promise's 99.9% availability threshold? The example separates the measurement and acceptance result from a runbook's approval.
Inputs and direct rules. This is a constructed case, not a report of observed service performance. Take as independently admitted inputs the service-delivery Work Delivery-June and evaluation Work Evaluation-July-1, with their performers' A.13 cores and A.15.1 admissions already established.
The exact A.2.3 promise clause Availability-June concerns the delivery Work's outcome. Its acceptance specification uses all 43,200 minutes in the 30-day June window, no exclusions, and the rule availability = availableMinutes / 43200; Met means at least 0.999 and NotMet means less. These two values are the declared result scale.
Observation and basis. The supplied C.16 observation Availability-Trace-June directly concerns the promised available/unavailable characteristic of Delivery-June. Under its stipulated complete-coverage measurement basis, exactly the first 40 minutes are unavailable and the remaining 43,160 minutes are available.
The observation's subject, characteristic, window, unit and complete coverage are premises of this case; an alarm log without that measurement basis could not replace it. No distinct indicator relation or F.9 correspondence is needed.
Application and result. During Evaluation-July-1, the evaluator enacts the declared evaluation Method through the A.6.1 application Availability-Application-June with availableMinutes=43160, totalMinutes=43200 and threshold=0.999. The application calculates 43160/43200 = 0.999074074..., compares it with 0.999, and returns Met on the declared acceptance scale.
If a separately recovered F.10 rule maps Met to RequirementStatus=Satisfied for this clause and window, that status can be reported; the application result alone is already the worked conclusion.
Evidence and bounded use. The descriptive A.10 account cites the exact observation-subject/measurement basis, promise-outcome fit, evaluation application and result-binding facts. Those independently established premises supply the evidence demanded by this acceptance rule, so pass is limited to explaining this June result. There is no separately named assurance claim, future-availability conclusion or permission claim.
Replay and boundary. The available and unavailable minutes sum to 43,200. The 40 unavailable minutes are below the permitted 43.2, and the computed ratio exceeds 0.999. With 50 unavailable minutes the same rule would return NotMet.
With missing observations, bound the unavailable time over the full window. Forty confirmed unavailable minutes and two unclassified minutes, with every other minute confirmed available, give at most 42 unavailable minutes. The threshold is met. Forty-four confirmed unavailable minutes already establish NotMet. If a qualifying and a failing history both fit the evidence, leave this threshold decision open. Apply any separately stipulated complete-trace or missing-data rule to the obligation it governs.
Applying the canvas to a multi-source availability case
The following is an application sketch. Supply the named source passages, exact relations, observations and result before claiming it is a completed worked example.
Title and situation. An alarm log does not by itself prove monthly uptime. Operations has an approved runbook and a month of IEC task and alarm logs; a service report must judge the exact ITIL promise-content claim.
Worked claim. June uptime is judged from admissible observations of the promised service outcome over the stated population and window. Alarm and command records may contribute evidence only through explicit relations and coverage limits.
Actual subjects and routes. The ITIL promise content and its promise-use, delivery, and fulfilment relations use A.2.3; the service-delivery Work and separate evaluation Work use A.15.1; the exact observations, availability characteristic, scale, and values use C.16; A.6.1 identifies the evaluation application and result binding; F.12 supplies the evaluation shape; A.10 describes the independently established evidence relations and qualifies the bounded reliance; and B.3 applies only when an actual named assurance claim is current.
To recognize the runbook as a MethodDescription under A.3.2, identify its one admitted Method as its exact subject and a substantive claim about that Method as a way of doing. Name the edition here only when it changes the evaluation result or replay.
Source basis. Cite the ITIL edition and promise passage, IEC edition and task and alarm passages, observation source and procedure, and any source-local meaning needed to interpret availability or alarm.
Relations and limits. State which observations concern which Work. First ask whether the observation and its measurement model directly concern the promised availability characteristic. If they do, use C.16 and A.10 and add no proxy.
If alarm-state intervals instead indicate a distinct unavailable-service characteristic, name both participants and the pattern that defines or tests that relation, with covered modes and blind spots. Use C.16.P to recover the relation and stop at A.6.RCD missing-governor when no such rule exists.
If the case uses an indicator, apply E.13 only when that indicator is optimized or drives a target, incentive, gate, release argument, reputation signal, repair, or decision. F.9 is needed only if the exact local meanings of alarm state and unavailable service are themselves related.
Result. A System performs evaluation Work, enacts the evaluation Method, and applies the declared availability rule to June's in-scope observations. The A.6.1 application binds those inputs and returns a result on the declared acceptance scale.
Map the result to RequirementStatus=Satisfied or RequirementStatus=Violated only through the exact F.10 rule. When the available evidence permits both threshold outcomes, state what remains open. Report EvidenceStatus=Inconclusive or RequirementStatus=Pending only when the corresponding F.10 rule supports that status. Preserve any result the local acceptance scale prescribes.
Create a verdict episteme only if another use needs it. State whether the promise was met, was not met, or cannot yet be judged from the evidence. The approved runbook establishes none of these results or statuses.
Checks. Actual subjects; a defining or testing pattern for each relation; direct-measurement-before-proxy; evaluation Work, application and result binding; declared result scale; separate EvidenceStatus and RequirementStatus; matching window and population; the indicator’s limit when one is used; and no row-created fact.
Relations
Builds on:
- Use F.1 for the exact source cut and F.0.1 and F.17 for each local claim or durable address needed by the case.
- Use F.7 only for a readable comparison surface and F.9 only for relations that obtain between exact local meanings, together with the separate bounded-use claim.
- Use E.10.D1 to replace vague context wording with the source, scheme, scope, model use, working situation, comparison basis, or other value that changes the case.
- Use F.10 for status claims and their windows. Use each direct subject rule for the worked checks; add F.15 only for the example's finite naming slice.
Coordinates with: A.2 with C.3 for a local system-role kind, A.2.1 for an assignment, F.4 only for a separately needed kind description, and F.6 only for precise performed-Work attribution; A.10 for evidence use; B.3 only when an actual named assurance claim is current; and the direct Part C pattern for every illustrated subject and relation.
Constrains: A cross-domain example may use this canvas or a faithful reduction. It names claim and gain first, actual values and relations next, and optional aids last. It never treats a lexical cell, table row, page layout, or generic Bridge as proof.
Didactic distillation
“Start with one practical question and one claim. Name the actual values involved and the pattern that defines each relation. Cite the exact source and evidence. Use a local-meaning cell only when you need a stable address; use a comparison table only when it makes the argument easier to read; use F.9 only when a real relation between local meanings is part of the case. State the result, its limits, and a few checks. The page displays the reasoning—it does not create it.”
Anti-patterns & remedies
Cross-domain application sketches
These sketches select claims and routes for a later worked case. They become replayable examples only after their exact local meanings, subject facts, source passages, observations and results are supplied.
OWL class and FCA formal concept
Situation and claim. A product catalogue uses an OWL class named Pump and an FCA formal concept with a similar label. The example asks whether their covered products may be compared for one catalogue query.
Recover both exact local claims. State the actual relation, if any, between their extensions for this data set and the difference between OWL subclass semantics and FCA lattice order. A small comparison table may display the evidence and limits. Similar labels and table co-placement do not establish identity or class inclusion.
System role and RBAC role
Situation and claim. The same System performs Work under a local operator system-role assignment and also has permissions grouped by an RBAC role. The claim is that these are different subjects even when both use the word role.
Recover the local system-role kind under A.2 with C.3 and the exact assignment under A.2.1. For claimed dated Work, recover the performer's A.13 core and independently admit the Work under A.15.1; add F.6 only for precise attribution through that same obtaining assignment. F.4 supplies a kind description only when one is needed.
Use the access-control pattern for the permission grouping, E.10.ROLE and F.0.1 for the trigger word, and the direct rule for any separation-of-duties constraint. A table row or shared spelling establishes neither sameness nor disjointness.
Method description, Work, and service promise
Situation and claim. A build MethodDescription contains a target duration; dated build Work occurs; observations report actual duration; a service promise is evaluated for a calendar week.
Use A.3.2 for MethodDescription membership, A.15 and B.1.5 for delivery Work and any claimed evaluation Work or Method enactment, C.16 for observations and measured values, and A.2.3 with F.12 for promise evaluation. Use A.6.1 when the exact evaluation application and result binding are current; use A.10 for a descriptive account of independently established evidence relations and bounded reliance, and B.3 only for an actual named assurance claim.
Cite the MethodDescription edition only if it changes the result or replay, and state separately whether the Work used it. A comparison table may show promised and observed values, but it establishes neither Work, result, status, nor a verdict episteme.
Safe reasoning moves
-
Recognize. State the working situation, failure, and practical gain.
-
Bound. Write one claim and one receiving use.
-
Name actual subjects. Replace vague words with the Systems, epistemes, Methods, Work, claims, observations, values, kinds, or assignments involved in the claim.
-
Route relations. Name the defining or testing pattern for each relation. For an alleged proxy, first test direct measurement; if a distinct indicator relation is needed and no rule supplies it, return A.6.RCD
missing-governor. -
Recover local wording. Cite the source, edition, and scheme; add an F.17 cell only when the receiving use needs a stable local-meaning address.
-
Show evidence and evaluation. Recover the direct evidence-use relations, their descriptive A.10 account and its bounded-reliance disposition.
When evaluation occurs, show its inputs, rule and supported result with any declared scale. Recover Work, Method and exact application/result bindings when claimed or required by the governing rule; add F.10 status or a durable verdict episteme only for their selected uses. Use B.3 only for an actual named assurance claim.
-
Use aids sparingly. Add an F.7 table only when it lowers reading cost; never infer from layout.
-
State limits. Include direction, loss, window, population, uncertainty, or non-use boundary that changes the conclusion.
-
Replay. Run focused checks for the case's failure risks, including result and status separation when evaluation is present.
Acceptance tests
Static conformance
- SCR-F16-S01 (recognition). Situation, failure, gain, and worked claim are clear to a cold reader.
- SCR-F16-S02 (actual subjects). No cell, row, label, or record substitutes for an actual value.
- SCR-F16-S03 (direct relations). Every substantive relation names its exact participants and cites the pattern that defines or tests it; use A.6.RCD
missing-governorfor a proposed indicator when no rule defines or tests its relation. - SCR-F16-S04 (conditional F.9). F.9 appears when the worked claim needs a test of semantic correspondence between local meanings with different interpretation bases. Any asserted Bridge must meet its predicate.
- SCR-F16-S05 (source and evidence). Exact sources and editions and the A.10 evidence-use basis are visible where they affect the claim; B.3 appears only for an actual named assurance claim.
- SCR-F16-S06 (optional aids). Cells and tables are omitted when they do not reduce reader effort; their presence or layout is not treated as evidence.
- SCR-F16-S07 (time and agency). Windows, Work, System, assignment, and attribution are explicit when material.
- SCR-F16-S08 (one-page parsimony). Each item helps the reader establish the worked claim or use the promised operation. A split case preserves access to its joint premises and conclusion.
- SCR-F16-S09 (evaluation architecture). When evaluation is present, its inputs, rule, supported result and applicable result scale are recoverable. Any claimed or rule-required performing System, evaluation Work, enacted Method and exact application/result binding satisfy their direct patterns; any selected EvidenceStatus, RequirementStatus or durable verdict assertion remains separate. For a claim about a class or a whole, the reader can recover the grounds for each aspect claimed. When the claim concerns interactions across a range, explain why the connection holds across that range. For an individual case, the evidence may settle that case directly.
Regression
- RSCR-F16-E01 (edition drift). Changed editions reopen only affected local claims and relations; no silent rewrite.
- RSCR-F16-E02 (relation change). A changed relation updates dependent conclusions and limits, not unrelated parts of the page.
- RSCR-F16-E03 (sense split). A changed local claim leaves no ambiguous F.17 address or table entry.
- RSCR-F16-E04 (window change). Changed cadence or population creates an explicit new evaluation boundary.
- RSCR-F16-E05 (plain-language guard). Repairing ontology does not replace readable sentences with avoidable symbolic or bureaucratic language.
Migration notes
- Refactor source tours. Recover the receiving use and its worked claim. Keep the sources needed for that answer or the method the reader is meant to reuse.
- Replace mandatory rows. Retain a table only if it improves comparison; move substantive relations and conclusions back to their direct statements.
- Replace cell subjects. Name actual kinds, Systems, Work, claims, observations, values, and evidence; leave cells as optional wording addresses.
- Split generic Bridges. Restore MethodDescription membership, description use, enactment, performed-Work, measurement, evidence, reliance, fulfilment, kind, assignment, publication, or transformation claims as applicable. Add an indicator relation only when a defining or testing pattern supplies it; otherwise retain the blocker.
- Repair trigger words. Recover what role, function, process, service, or context denotes in each sentence before rewriting it.
- Keep the gain. After every ontological repair, reread the page as a cold practitioner and simplify any wording that became harder without gaining precision.
Teaching variants
- Single-source direct case. One source, one actual value, one obtaining relation, one boundary; no table or F.9 relation.
- Two-source comparison. Two exact local claims, an optional F.7 row, and one separately supported conclusion.
- Triangulated evidence case. Promise, Work, and observation sources joined by their direct fulfilment, measurement, and evidence relations.
- Difference lesson. Two trigger-word uses kept distinct, with no relation asserted unless one obtains.
- Window primer. The same promise and Work evaluated under different explicit windows to show why result, status-use, and verdict-episteme identity must be checked separately.
Didactic checklist
- Recognizable situation, failure, and gain?
- One bounded worked claim and receiving use?
- If the method is meant to transfer, which inputs and choices can change, and why does the operation still work? If several parts or pages supply the answer, what holds their results together across the stated range?
- Actual values named?
- Defining or testing pattern named for every substantive relation, with direct measurement checked before any proxy?
- Exact sources and editions and evidence where material?
- F.17 cell or F.7 table only if it lowers reading cost?
- If the example tests a relation between local meanings with different interpretation bases, does it distinguish the F.9 test result from any asserted Bridge?
- Evaluation Work, application result, declared scale, status family, optional verdict episteme, window, population, agency, loss, and uncertainty visible where they change the answer?
- Plain language retained after ontological repair?
Closing distillation
A useful worked example is one replayable argument: situation and gain → claim → actual subjects → direct relations → exact sources and evidence → result and limits → a few focused checks. Cells and tables are optional reading aids. They never replace the values, relations, evidence, or judgement that make the example true.
F.16:End
Make a Settled Naming Decision Recoverable for Reuse (Unified Term Sheet)
Type: Lexical row pattern (F) Status: Stable
Use this when. An independently governed value has a settled durable name and public, Core-facing, durable, or cross-local reuse needs a reader-facing term row.
First useful move. Point to the exact governed value, its kind, the pattern that defines or constrains it, one proposed row use, the selected Tech designation and any Plain designation needed for that use. Then apply F.14 at the row gate. If no durable row is needed, reuse the designation, alias, local expression or name already supplied for that value and stop.
Primary working object. One C.2.1 UnifiedTermRow episteme whose exact EntityOfConcern is the independently governed value. Its claim graph cites the separate F.18 naming-settlement episteme and selected designation expressions. Section 5.1 explains the identities and relations needed to recover this decision and distinguish the row from its publication.
What goes wrong if missed. A table entry becomes an ontology claim; a stable identifier looks like identity evidence; one source title or file stands in for a local sense; a NameCard automatically creates a cell and row; or a row is mistaken for the publication occurrence that makes it available.
What this buys. A compact, durable navigation row through which readers can recover the naming decision and the rules that define or constrain the governed value without letting the row create, merge, prove, or publish that value.
Not this pattern when. Keep private wording, local synonyms or aliases, and names already supplied where the value is defined or constrained in their local use. Use F.14 before every naming object, F.8 for one unresolved mint-or-reuse choice, F.18 for the durable naming settlement, F.9 only for an actual relation between exact cells, and E.24.PUB only when a selected row edition must be made available.
For any stronger ontology, obtaining, equivalence, authority, system-role-kind, assignment, relation-position, status, evidence, Work, or subject-use claim, use the pattern that defines or constrains it.
Intent and applicability
UnifiedTermSheet is a reader-facing collection of independently identified term-row epistemes for one useful naming thread. Each row makes one selected naming decision easy to find: it names the governed value, its kind, where that value is defined or constrained, selected designations, exact local senses, any actual Bridge needed by the declared use, admitted citation use and any grounded blocked-use claim, and reopen condition.
Use it especially for:
- public system-role-kind and status names whose underlying values are already governed;
- durable relation, slot, interface, signature, or FPF kind names;
- Core-facing names used by examples, training material, project standards, dashboards, checks, tool interfaces, Part G search packs, architecture, transformation, or evaluation work;
- one exact naming use between independently recovered local senses;
- row identifiers that must remain usable across row-episteme editions.
F.17 constitutes the row episteme. Claims about the named value and its use remain governed by their defining or constraining patterns. A table can display the row, but its position, filled cells, suffix, source prestige, or row count does not establish the value's identity or permitted use.
Problem frame
Naming work often succeeds locally and then fails in reuse. A term looks stable, but the receiving reader cannot recover which value was named, which pattern defines or constrains it, which naming decision selected the expressions, which effective scheme and local-sense claim are current, or whether a cited Bridge obtains.
Five shortcuts follow:
- shared spelling is treated as shared value;
- a row combines unlike system-role-kind, assignment, status, relation, Work, evidence, or publication concerns;
- a card, cell, row, id, and publication are minted as one automatic chain;
- a source title, document, or table layout substitutes for the exact sense and basis relation;
- the row itself is said to make the term public, current, authoritative, or obtaining.
F.17 repairs those shortcuts by making every row a separately identified claim-bearing episteme whose references lead back to the exact naming settlement and governed value.
Problem
The practical problem is to make one durable naming decision recoverable without turning its row, representation, or availability into the named object. If a row is needed, it carries one decision. Split a row that combines decisions. For any stronger claim, use the rule that defines or constrains it.
Forces
Solution
Recover the naming objects needed for the declared reuse, then constitute a row only when its separate gate is met:
- Recover the value. Identify one exact already-governed value or relation, its kind, the pattern that defines or constrains it, its identity or obtaining semantics, and one proposed use. Split a mixed candidate before naming.
- Apply the anti-explosion gate. Use F.14 to establish what the receiving use needs before minting a card, cell, row or family. Reuse an adequate local expression, existing designation, alias or admitted row directly. A stronger object needs a contribution that those adequate treatments do not supply.
- Settle only the durable name that is needed. If one expression remains unresolved, use F.8. If a durable naming settlement is justified, F.18 constitutes one C.2.1 NameCard and selects the Tech designation and any Plain designation needed for a stated didactic use. The card creates neither the value nor its kind and does not require a cell or row.
- Address a local sense only when useful. Create one
SchemeSenseCellonly when the exact local expression and sense claim need a stable address under an effective by-valueReferenceScheme. Cite a selected bounded-model-use Structure only when its organization changes this naming use. The cell does not require a NameCard or row. - Open the public-row gate independently. Apply F.14 again when public, Core-facing, durable, or cross-local reuse needs a row. The current F.18 public-row interface supplies the exact NameCard, selected designations, governed value and kind, the locator for its defining or constraining pattern, effective scheme, and exact cell. None of those inputs alone requires the row.
- Add a Bridge only for an actual cross-local relation. Compare the exact
<ReferenceScheme, LocalSenseClaim>projections. When the proposed row use relates different projections, cite an obtaining F.9 Bridge between the exact cells and separately cite the affirmative C.2.1 use claim plus its current A.10 or B.3 reliance. Same spelling, scheme difference, or cell presence proves no Bridge. - Constitute one row episteme. Its C.2.1 EntityOfConcern is the exact independently governed value; its claim graph cites the separate naming-settlement episteme, selected designations, admitted citation use, any grounded blocked-use claim, rationale, and reopen condition. Split unlike governed values or independently different uses into separate rows.
- Keep succession and availability downstream. Use
EpistemeEditionRelationonly when a later row episteme historically continues an earlier one under C.2.1. When availability is current, use the exact E.24.PUB expression, bearing, and publication relations. A row, row id, form, carrier, upload, or rendering establishes neither succession nor publication by itself.
Apply the static and regression checks to the affected row, then stop. The result makes the naming decision recoverable for its stated citation use. Establish claims about the named value, its use or publication through their governing patterns.
Minimal vocabulary
Scheme-based local-sense coordinate, basis relation, and row episteme
A selected expression, an exact local sense, the episteme supporting that sense, the naming decision, and the reader-facing row answer different questions. Keep them independently recoverable.
SenseCellAddressRef designates one SchemeSenseCell; it does not create that cell or a universal context object. A legacy address is usable only through an explicit lossless adapter to the exact effective scheme, expression, and local-sense claim. Otherwise stop the row.
Resolve each reference in the source edition selected for that use. The enclosing publication can supply this selection; retain it when transferring the reference. A pattern locator such as E.17.0 supplies an exact basis only when the selected edition and cited claims recover one episteme. If they do not, leave the reference unresolved instead of substituting the currently available edition.
The basis relation has exactly two participants. basisEpistemeRef resolves the exact selected basis-episteme edition; its exact kind is derived from that referent and is not copied as another participant.
A relation reference resolves the exact LocalSenseBasisRelation occurrence rather than its description or designator.
basisPublicationUnitRef, when present, is a provenance qualifier that narrows the supporting episteme; it neither participates in nor identifies the relation. A source publication occurrence, its form, and its carrier remain separate E.24.PUB objects.
The relation says only that this basis episteme supports this cell's exact sense claim for the admitted use. Its description states the supported uses, any grounded blocked-use claim, and any exact source-unit qualifier.
A changed NameCard reopens the selected expression. A changed scheme, expression, sense claim, or basis-episteme edition identifies another cell or basis-relation participant pair. A changed source-unit or supported-use claim creates another relation-description episteme without silently changing the basis relation.
Any description of a SchemeSenseCell is a separate C.2.1 episteme whose EntityOfConcern is that exact cell. The cell's identifier, description, source publication, NameCard, and basis relation neither replace nor identify the cell.
UnifiedTermRow is another C.2.1 episteme, not a root U-kind, value container, or publication occurrence. Its EntityOfConcern is the exact governed value. Its displayed identity-bearing row claims jointly constitute the complete ClaimContent; a scalar graph-ref line need not be repeated in the readable fixture when that graph is recoverable from them.
The claim graph cites the separate NameCard and the governed value's kind, locates the rules that define or constrain that value, and projects the selected designation expressions.
The row, card, designations, governed value, external row reference, and UTSRowId designator remain distinct; UnificationThreadId, Block, and RowEditionId are navigation or edition designators rather than additional identity discriminators.
To establish historical continuation between two row epistemes, test C.2.1's EpistemeEditionRelation(earlierRowEpisteme, laterRowEpisteme). An independently established refinement or supersession claim does not by itself supply that continuity. Stable row spelling, id, table position, shared carrier, or later publication establishes no such relation. A CurrentnessCondition is row claim content; it is not the edition relation and does not make itself true.
When a selected row edition must be made available, E.24.PUB supplies PublicationExpressionRelation(selectedRowEdition, expression, boundedUseDeclaration), PublicationExpressionBearingRelation(carrier, expression) and EpistemePublicationRelation(selectedRowEdition, audienceDeclaration, boundedUseDeclaration, expression, carrier).
A selected reusable form constrains that expression through form conformance. Form-mediated availability uses the same expression as witness for expression, bearing and conformance; broader availability retains its permitted alternatives.
The row is the claim-bearing episteme, X is its particular expression and C is independently identified under its carrier-domain rule. Any asserted rendering or upload Work has its separately established performer and occurrence.
GovernedValueRef and GovernedValueKindRef are separate. A kind token has kind U.Kind. An exact local system-role kind, obtaining system-role-assignment or other relation occurrence, status value, slot kind, representation position, or local concept retains its own kind; the row points to the pattern that defines or constrains that value. A row or card cannot admit a U-kind or make a direct relation obtain.
NameCardRef resolves the F.18 C.2.1 naming-decision episteme consumed by the current public-row gate. UnifiedTechName and any UnifiedPlainName are designation expressions selected by that decision, not values or references. Aliases and rejected candidates stay in the NameCard or local lexicon rather than becoming rival selected names in the row.
BridgeRefs cites only actual F.9 occurrences between exact cells. Direction, use-specific rule, loss tolerance, polarity, evidence, reliance, permission, and receiving action remain in their own claims and relations. Local senses do not globalize; same spelling or a different scheme provides neither governed-value identity nor Bridge obtaining.
A.22.CGUS:4.4 permits a separately constituted demonstrative-slice episteme after CGUS qualification. The token DemonstrativeUnfoldingSlice@Context is neither a U.Kind nor an exact slice by itself. F.17 records a row only after one exact C.2.1 slice episteme and its current F.18 naming settlement are recoverable; a local phrase or seminar expression alone creates neither.
UnifiedTermSheet is the reader-facing collection or layout through which rows are found. A selected table layout, optional block plan, or carrier is not the row episteme and does not prove that every needed decision is present.
When to create or update a UTS row
Create or revise one row only when all entry objects are exact and at least one receiving need is current:
- public or Core-facing citation of the selected naming decision;
- durable reuse outside the immediate local repair;
- cross-local reuse whose exact cells, any actual Bridge, separate use claim, and reliance are recoverable;
- stable citation from examples, checks, dashboards, training material, a project standard, or a tool interface;
- a change to the rules that define or constrain this value, or an F.18 change, that alters this row's value, name, sense, admitted use, or blocked use.
Before the row, apply F.14 again. A noticed word, accepted designation, stable local sense, NameCard, Bridge description, source publication, or desire for a tidy table does not by itself meet the gate. A durable local NameCard can remain local; a cell can remain a cell; an existing row can be reused only within its admitted use.
Row schema
Use these positions when they are current. Presence means that the exact referenced object or claim is independently recoverable; it is not a form-completion target.
For SenseCellRefs, recover the exact by-value scheme, expression, and local-sense claim. Cite LocalSenseBasisRelation only when the basis relation obtains. A NameCard selects designations; it does not fill the cell or basis positions. A source title, file, carrier, locality label, selected structure, row id, or description substitutes for none of them.
Publication availability is not a row column. When current, maintain the exact E.24.PUB relation occurrences, particular expression, carrier, audience and bounded use beside the selected row edition.
When a reusable publication form is selected, identify it and establish the expression's conformance. Form-mediated availability uses that same expression as witness. Publication change does not silently change the row episteme or its C.2.1 edition relation.
Optional block plan
A block plan is an optional navigation aid for a sheet with enough rows that grouping helps a reader. Use few memorable blocks and omit the plan when direct row search is clearer. Neither a declared plan, the number of blocks, nor filled row count proves coverage, completeness, usefulness, or semantic adequacy.
Example navigation plan for a system-role, Method, Work, and status thread:
- governed values and naming decisions;
- system-role kinds and their descriptions;
- system-role assignments and performed Work;
- methods, method descriptions, and work plans;
- status families and status windows;
- relation, slot, interface, and Bridge terms;
- evidence, assurance, source, and publication terms when those are the governed values.
This list defines no ontology. A sheet may use another small navigation plan for architecture, transformation flows, evaluation characteristics, Part G search packs, or another receiving use.
Layouts
F.17 admits two common layouts.
Layout A, scheme-first: keep the left rail fixed and add one exact reference-scheme column per selected interpretation basis. Use this when the reader's comparison concerns local senses under named schemes.
Layout B, comparison-column: keep the scheme, local expression, and sense claim inside SenseCellRefs and use presentation columns such as tradition, discipline, language, publication family, or project family. For each compared entry, the reader must still be able to resolve its exact by-value scheme and local-sense claim. The column label helps organize the comparison; that recovered scheme and claim determine its interpretation.
Never mix a scheme column and a discipline or project-family column as if they had the same kind. A ReferenceScheme is an interpretation basis carried by value; a comparison column is a didactic view.
Static conformance rules for a UTS
Use these checks before citing a row outside its immediate sheet.
A row succeeds when intended readers can recover the correct naming decision, governed value, and applicable defining or constraining rule for the declared use, and distinguish any grounded excluded use. Schema conformance and maintenance aids such as row counts, label uniformity and stable identifiers do not establish that recovery.
Regression and stability rules
Recheck only the rows affected by the changed object, name, scheme, sense, Bridge, basis, or source.
Archetypal Grounding - worked cases
System-role-kind name becomes public across two project contexts
One project has an exact local DesignReviewerSystemRole kind and another has an independently governed ExternalAuditReviewerSystemRole kind. Both local expressions say reviewer, but one classifies an admitted System that may perform design-review Work and the other classifies an admitted assurance System that may produce an audit report. Any actual assignment and Work are separately identified.
Create two rows, one for each independently governed local reviewer kind. If a named use needs correspondence between their two exact sense cells, cite an obtaining F.9 Bridge and an affirmative C.2.1 claim that names direction, label rule and tolerated loss. Each row cites the pattern that defines or constrains its local system-role kind, its SystemRoleKindDescription when current, and the F.18 NameCardRef. Use A.10 for ordinary reliance on the use claim and B.3 when a named assurance claim is current; the row and card establish neither assignment nor review Work.
Status label looks like a system-role-kind name
A team proposes BlockedReviewer as a public label. F.17 does not accept it as a row until the two governed values are separated. ReviewerSystemRole is a local system-role kind; blocked is a status-family or status-window value. The sheet may record one system-role-kind row and one status row, with a note that a local UI may render their labels together. The table creates neither a BlockedReviewerSystemRole kind nor an assignment. If either exact row edition must later be made available, use a separate E.24.PUB publication package.
Learning is an anti-row and split prompt
A DPF or dashboard proposes one public Learning row, perhaps with LearningProgress as its value. Apply E.10.LRN first. Teaching or practice Work, a holder's capability, a fitted model edition, an inference result, acquired information, a representation relation, cultural change, and a course or other product are independently governed values and claims. Performance, capability evidence, information gain, model fit, prediction error, compression, and representation change are likewise different progress coordinates; shared spelling does not make them one governed value, one SchemeSenseCell, or an F.9 Bridge.
F.17 therefore creates no umbrella Learning or LearningProgress row. If one recovered value later needs a durable public designation, apply F.14 and F.18 to that exact value and create at most one row for its admitted use; another recovered value receives another row only under its own gate. When the direct claim is already readable or durable reuse is absent, stop with no UTS row.
Relation and slot names become reusable
An architecture pattern needs names for interfaceSlot, providedPort and requiredPort. Recover each governed meaning: the slot in the relation signature and what makes a port provided or required in this interface. Use an existing designation when it serves the receiving work. A name used once can stay local. When F.14 establishes a separate need for repeat public use, give the governed value its own row with the F.18 technical designation and a plain designation when useful. Keep other aliases with their source in the NameCard or lexicon. Cite A.6.5 for slot discipline, A.6.RSIR for the relation-signature-interface boundary and F.18:5.6 for the naming decision. A project may use port differently. Test any proposed cross-source Bridge under F.9 and justify its intended use separately.
Misleading evidence-role row
A sheet has a row labelled Evidence role. Recover the governed object before choosing a name. For an episteme's evidence use, establish the direct relation under A.2.4 or its other subject rule; A.10 describes the evidence-provenance path and classifies the bounded reliance, and B.3 enters only for an actual named assurance claim.
For dated evidence-producing Work, recover each precise performer's A.13 core, including the same obtaining assignment, and independently admit the Work under A.15.1. Add F.6 only when the row or receiving use needs precise assignment-bound attribution. A separately needed local-kind claim uses A.2 with C.3.
The UTS may expose selected names for these distinct values; it does not admit a generic evidence-role row that fuses them.
Manufacturing batch across material and planning contexts
A furnace team uses batch for one physically handled set of shafts that shares a heat-treatment run and traceability basis. A planning dashboard uses batch for a grouping of intended PlanItems. Spelling does not make these one governed value. Recover the physical batch under the material or production DPF pattern that supplies its identity and part-whole rules when the proposed comparison relies on either; recover the planning grouping and its relation to intended PlanItems under A.15.2.
When the row gate is met, record separate rows for the distinct governed values. A needed semantic comparison may cite an obtaining F.9 Bridge and a separate affirmative C.2.1 claim stating direction, correspondence rule and tolerated loss with current reliance; those facts do not combine the values into one row.
If either selected row edition must be made available, apply E.24.PUB separately. A batch row cannot turn a PlanItem grouping into a physical holon or make the physical batch a WorkPlan.
Clinical discharge wording
A clinical publication proposes one row for discharge and discharge-ready. First separate the governed values. A patient-state classification uses A.19.SPR plus the clinical DPF pattern for its bearer, state frame, evidence, qualification window, and use. An accountable discharge decision remains a decision relation under the pattern that defines or tests that decision. A completed discharge is dated Work under A.15.1.
If the naming use needs a row for a patient state, discharge decision or completed discharge, record that value separately. Connect rows only through relations established for the clinical use. A later publication operation uses E.24.PUB for each row edition that must be made available. One familiar label does not make state, decision, and Work interchangeable.
Demonstrative walkthrough and bounded mantra
A.22.CGUS:4.4 permits a separate C.2.1 episteme that shows one traversal through an already qualified CGUS for a declared teaching or comparison use. It does not define a demonstrative-slice U.Kind, and the token DemonstrativeUnfoldingSlice@Context does not identify one exact slice. A proposed seminar reference-scheme name must also resolve to an effective by-value scheme. These labels alone therefore supply no basis for a demonstrative-slice row, seminar SenseCell, Bridge or bounded-use claim.
Use demonstrative walkthrough as ordinary readable wording for a shown traversal when the sentence makes the exact slice clear. Keep mantra as bounded seminar or pattern-local recall wording when repetition and attention are the point. Neither expression creates a kind, episteme, scheme, cell, Bridge, row, or publication occurrence. mantra move remains E.10.MOVE Plain wording for an E.11.PUA practice-continuation description when such a description is shown; ordinary long and local mantras receive no F.17 row.
If a later use needs a durable public name, first recover one exact slice episteme under C.2.1 from its claim content, the qualified CGUS it concerns, and its effective scheme. F.18 may then record one naming settlement and F.17 may record one row after the ordinary gate. A second cell and F.9 Bridge are justified only when another exact scheme-and-sense projection and a named current correspondence use both exist. Availability of any selected row edition still requires a separate exact E.24.PUB publication package.
Bounded model-use structure public row
This row records the selected A.1.1/F.18 names for the dependent U.Structure specialization. A.1.1 separately governs the identity and membership of structure individuals and the obtaining of their relations; E.24.PUB governs availability of the row edition.
This row records BoundedModelUseStructure for public reuse; availability of its selected edition requires the separate E.24.PUB publication relations. A.22 retains CrossContextRelationStructure as a provisional local designator and governs when a positive structure member may be claimed. Any later naming settlement or public row must be established for its exact governed value under F.18 and the row gate.
Three bounded-model-use direct relation-kind rows
These rows record the three A.1.1 relation-kind names used by E.24.UK. Each row makes one naming decision recoverable. A.1.1 separately defines the obtaining and reidentification tests for relation occurrences, and E.24.PUB governs availability of a selected row edition.
Do not create a public F.17 row for ModelExpressionCoherencePredicate: that label remains local to A.1.1 and names the five-part criterion ValueKind rather than any of the three relation kinds.
Viewpoint, view, and conformance-relation public rows
Use three separate rows for the different receiver needs below. E.24.UK admits U.Viewpoint and U.View as same-individual dependent kinds under U.Episteme; E.17.0 defines their positive membership predicates and EpistemeViewpointConformanceRelation.
Stable Core citation and typed-reference use justify the rows. Retain the existing Tech designations under F.14 rather than creating a synonym family; establish membership, relation obtaining and publication through their separate governing rules.
The two existing dependent-kind designations use the following F.18 naming-settlement epistemes. They remain distinct from the E.24.UK admission results, the governed kinds, their members, every reference or designator, and the F.17 rows that cite them.
U.Viewpoint
U.View
EpistemeViewpointConformanceRelation
Each row retains its C.2.1 episteme identity and refers separately to the governed value, naming decision and local sense. The displayed fields preserve the distinct designators, references and relation occurrences. When readers need access to a selected row edition, recover the E.24.PUB expression, bearing and publication occurrences outside the row; displaying this block alone does not establish them.
Make a settled row available only through a separate publication operation
Do not perform an E.24.PUB publication operation on a placeholder. First require the governed value, its lexical classification, the reference scheme's selected name and permitted scope, and the intended reader use to pass F.18 and the ordinary F.17 row gate. If any input is unresolved, keep it as naming work rather than representing it as a current row.
When availability is current, state how the selected row edition is available to its audience for the bounded use through an expression and carrier. E.24.PUB governs that claim at its required granularity; name reusable-form conformance only when it matters.
A NameCard, scheme-sense cell, basis relation, row reference, or E.24.PUB occurrence substitutes for none of those decisions. Keep predicate definition, actual use, basis analysis, naming settlement, row admission, and downstream availability separate.
Role-Precision Core Rows
These eight rows expose the F.18 designation pairs for Core citation. Each names one value defined or constrained by its subject pattern and uses one FPFCoreReferenceScheme sense cell. Its currentness depends on that value and subject-pattern rule, the E.10 token classification and allowed-use rule it consumes, its exact NameCard, sense cell and cited use.
The rows use these exact scheme-based sense cells; the cells name no additional value and require no Bridge merely because both Tech and Plain designations exist.
Architectural Reference term row
This row makes the F.18:4.2e settlement reusable for the architectural-reference form already governed by E.4, E.11.PFP and E.11.DSG. The governed value is the functional publication-unit form, not a newly minted root kind.
A single row is useful because the same function is cited across these rules and applied at framework, Suite and Library levels. The row identifies that function without identifying all its instances as one publication.
For example, FPF Library Architectural Reference and Engineering Suite Architectural Reference identify different explained wholes. An embedded Method Engineering Principle Framework Architectural Reference performs the same function for that framework. Whether the account is independently managed depends on its use and change boundary. Its title alone does not decide this. The row makes the designation recoverable, while E.24.PUB separately governs any publication of this row or a particular architectural reference.
Model-comparison direction and level rows
Use these three rows to cite model refinement, its converse and the levels established by comparing models. B.5.MR gives the tests for the two model relations. The level row gives both directional readings of one order used in project documents and teaching.
In the shared ReferenceScheme a model's completions satisfy its assertions. The inquiry and its interpretation determine the answer read from each completion, including probability claims when relevant. C refines A when at least one completion of C exists and every completion of C satisfies A under that interpretation. A is then an abstraction of C. Strictness requires an answer that A admits and C excludes. Within a chain established by this test, models with the same answers share a level for that inquiry. Models on different branches can give the same answer while imposing incompatible constraints.
For each row, UnificationThreadId is ModelComparisonDirections, GovernedValueKindRef is U.Kind and ReferenceScheme has the value given above. The stable row designators are retained. RowEditionId for this edition is 2026-11-18. Each sense cell gives its expression and local sense under the shared scheme.
For a source that says “level”, “scale” or “layer”, use these model rows when it compares models for one inquiry. Library.Reference:6.2.2 gives the companion F.17 rows for the composition of each kind of holon, functional layers, measurement scales and the scope kinds of USM. A Method stack can be read for both its functional organization and its constituent Methods where both relations are established. To recover the range of a speaker's “all” before choosing a row, use C.39.QR:4.1.1.
The rows cite no F.9 Bridge: they make no correspondence claim between different local meanings. A concrete comparison still needs its models, inquiry, interpretation and B.5.MR test; citing a row supplies the selected name and its meaning.
Bias-Annotation
F.17 blocks table-bias: a row does not make the named object real, global, reusable, equivalent, or authoritative. It also blocks label-bias: the public name is a designation for a governed value, relation, slot, or local concept, not a substitute for the rules that define or constrain it, the scheme-based local-sense coordinate, Bridge, admissible-use statement, or currentness condition.
Conformance Checklist
Common Anti-Patterns and How to Avoid Them
Closure conditions
One row is ready for its declared citation use only when:
- the governed value, exact kind, the pattern that defines or constrains it, and the proposed use are explicit;
- each current card, cell and row supplies a receiving need established under F.14;
- the exact F.18 NameCard, selected Tech designation and any Plain designation are current for this public-row gate;
- every
SchemeSenseCellresolves to one by-value ReferenceScheme, local expression, and local-sense claim; - every relied-on local-sense basis is an actual two-participant
LocalSenseBasisRelationand every description/source/publication object remains separate; - any row use that relies on semantic correspondence between different local meanings cites an obtaining F.9 Bridge for the exact endpoints, a separate affirmative use claim and current A.10 reliance for that use or, when a named assurance claim is needed, the corresponding B.3 assurance result;
- the row has one decision, an admitted citation use, any grounded blocked-use claim, an edition designator, and a reopen condition;
- any historical continuation is an exact C.2.1
EpistemeEditionRelationrather than shared id or title; - any claimed availability has E.24.PUB grounds at the granularity asserted; an ordinary availability statement is sufficient when no finer identity or change question is current; and
- every ontology, obtaining, equivalence, authority, system-role-kind, assignment, relation-position, status, evidence, Work, and other subject-use claim uses the pattern that defines, constrains, or tests it.
No other row needs to be filled before this one can close. A sheet's row count or optional block plan says nothing about whether another naming decision is substantively needed.
Consequences
Benefits. Readers gain a stable route from the selected designation or designations to the naming decision, governed value, defining or constraining rules, local sense, and admitted use without treating the table as ontology or publication proof.
Costs. A tempting row waits until the independently governed value, F.14 disposition, naming settlement, exact cell, and any actual Bridge are current. Publication availability adds its own E.24.PUB objects only when needed.
Failure avoided. F.17 prevents global glossary drift, card-cell-row cascades, label-based sameness, row-shaped ontology, optional-layout completeness claims, and authority or publication smuggled through a table.
Rationale
Terms travel farther than the reasoning that produced them. F.17 carries only the reopening hooks needed for that travel. The pattern that defines or constrains the governed value, together with F.18, F.9, C.2.1, A.10/B.3, and E.24.PUB, supplies the separate rules to which those hooks lead.
SoTA Decision for One Reader-Facing Term Row
Question and selected answer. At the effort of settling one reusable term, what must a reader recover beyond a familiar label, and what apparatus can be left out? Under E.8:11, use one readable row that returns separately to the named value, naming decision, exact local sense, permitted citation use and any grounded blocked-use claim, and the condition that reopens the row. Create it only when that reader-facing route is needed.
The lightweight serious rival. The OntoLex Community Report, Overview, Purpose, Core, and Semantics permits the core module alone. A fair one-term comparator is a lexical entry with a form, a sense and a reference to the already defined ontology entity, plus only the usage notes this application needs. It does not require a full lexicon, morphology or syntax graph; it is not intended to define the ontology itself.
Adapt its expression/sense/reference separation in 5.1 and 7. Its extensibility also permits notes or a view to expose the same decision content as an F.17 row.
Compare the row with the lightweight lexical entry. Hold the term, target, use conditions, and maintenance obligation fixed. A bare label or bare form/sense/reference chain lets the reader find the expression and target, but leaves the naming choice and reason to revisit it unstated, along with any blocked use that the reader needs to distinguish.
Adding those statements to the lightweight rival closes that gap; a generated readable view of those same statements is an acceptable way to express the F.17 row. A different vocabulary or file format is not, by itself, an improvement or a second naming result.
For an ordinary reader inspecting one term, select the explicit row because the decision and return are visible together. The trade-off is maintaining a truthful reader-facing projection instead of asking that reader to reconstruct it from linked lexical statements. When lexical tooling already produces that projection, keep it and avoid a second editable source.
When morphology, translation, or machine lexicon exchange is the actual use, the core model and only its needed extensions may be the better carrier. No measured speed advantage, blanket cost superiority, or replacement of OntoLex is claimed.
Exact effect and countercases. Steps 1–5 of section 4, together with sections 6 and 14, stop at ordinary wording, a local card, or a cell when no row is needed; steps 7–8 of section 4, together with sections 5.1 and 7, make the row's returns explicit and keep later publication separate.
The countercase in 12.2 prevents a state label from becoming a system-role kind, 12.4c leaves ordinary mantra wording without rows, 12.4h distinguishes a reusable structure kind from one selected instance, and 12.4i returns the Reference name to its product-form rule without creating a product or lookup Work.
These cases supply the FPF-local reason for the added decision content; the OntoLex report supplies the rival's lexical modeling capability, not validation of FPF ontology. Reject mandatory full-lexicon modeling, raw label familiarity as sufficient reader recovery, and the inference that a row makes its referent or publication real.
Reopen. Recompare when an equally small lexical entry or other presentation exposes the same decision, relevant use limits, and exact returns with less reader and maintenance effort; when a generated view drifts from its source; or when the current use needs linguistic distinctions or machine exchange omitted by the row. If the added reader projection no longer changes use, retain the simpler expression of the same content.
Currentness rule: when F.2, F.3, F.5, F.7, F.8, F.9, F.10, F.14, F.15, F.18, C.2.1, E.17.0, E.24.UK, E.24.PUB, A.1.1, A.2, A.2.1, A.2.7, A.6.5, A.10, B.3, E.10.D2, or the pattern that defines or constrains the governed value changes the value, kind, membership or obtaining rule, designation, scheme, cell, basis relation, Bridge, bounded-use claim, reliance, status and system-role boundary, edition relation, reference typing, or publication boundary, recheck only the affected rows and worked examples.
Relations
Builds on: F.2 and F.3 for local-sense discovery probes; C.2.1 for row and NameCard epistemes plus exact EpistemeEditionRelation; F.14 for the anti-explosion gate; F.8 and F.18 for naming disposition and settlement; F.9 for actual cell-to-cell Bridges; F.5 for designation form; and F.7/F.15 for neighboring unification and conformance decisions.
Coordinates with: E.10.LRN for the learning anti-row and split boundary; A.2, A.2.1, A.2.7, A.6.5, A.6.P, A.10, A.15.1, A.19.SPR, B.3, C.2.P, E.10, E.10.D2, E.17.0, E.24.UK, E.24.PUB, F.4, F.6, and F.10, plus every row's SubjectPatternLocator.
Row-local review after a changed value, membership or obtaining rule, designation, cell, Bridge, reference typing, edition, or availability rechecks the exact defining predicate and any neighboring subject assertion. Use G.11 only when an actual refresh plan, edition orchestration, telemetry, freshness, or decay claim is current. F.17 does not inherit a generic context-holon identity reading from earlier terminology practice.
Constrains: every public, Core-facing, durable, or cross-local term row that cites FPF values, local senses, relation names, slot names, system-role names, status names, or Bridge occurrences.
Didactic distillation
Recover the governed value and settle only the name needed for the use. If readers need a durable route, make a row that leads back to the naming decision, exact local sense and defining rules. Revisit the row and its dependent uses when those grounds change; establish publication availability separately.
F.17:End
Local-First Unification Naming Protocol
Status: Stable For practitioners who need a durable reusable name for an already governed value.
Use This When
Use F.18 when a name must become stable, public, Core-facing, reusable under more than one named source, practice, or reference scheme, or durable enough that later work can cite it without guessing. Typical cases:
- a local expression becomes a durable name for an already governed value;
- two teams use different words for the same candidate sense and need one reusable term plus preserved local wording;
- one tempting head word is useful under one recovered local meaning but misleading under another;
- a proposed name implies a different kind from the value it is meant to designate.
First useful move.
- Recover the exact value and the pattern containing its defining or testing rule.
- Decide whether ordinary local wording is enough or later use needs a durable name.
- For a durable name, reuse a current naming settlement that already covers this value and use. Otherwise compare the plausible names and record the Tech label, any Plain designation needed for a stated didactic use, the selection reason and the condition for reconsidering the choice in a
NameCard.
If bare claim-bearing role still hides the object, use E.10.ROLE; if relation, slot, interface, port, or signature wording hides it, use section 5.6. Open section 4.4 only for a public, Core-facing, durable-across-context, or cross-context use. A public row is a later result, never part of the first naming move.
For one-off wording repair, use F.19. Return to E.10, E.10.ARCH, A.6.P, A.6.RSIR, C.2.P, or the subject pattern only if the object or claim remains unresolved. Use F.18 when a reusable name is needed.
For example, say in ordinary words whether one exact Bridge is suitable for one named use; do not create a NameCard, public claim kind, or durable CamelCase head merely to abbreviate that C.2.1 claim. Reopen F.18 for that claim only when an independent later use needs a reusable name beyond the local statement.
Context
A name lets people refer to an independently governed value. The naming decision selects designations for that value; its subject pattern supplies the rules for identifying it and testing claims about it.
FPFCoreReferenceScheme names the EpistemeInterpretationScheme value used to interpret current FPF Core Tech labels and relation names; a name under another scheme carries that scheme by value. C.2.1:4.1.1–4.1.2 identifies the actual rules, domains and semantic dependencies and supplies the range. Recover those rules for the naming claims being made; the scheme's label or a document about it does not supply the value.
Most naming work stays within one <ReferenceScheme, LocalSenseClaim> projection and needs no Bridge. If one named use must relate different projections, follow the later cross-projection branch in section 4.4.1. Shared spelling, another scheme, or a selected model-use structure alone creates no Bridge, use claim, reliance, governed-value identity, or U.BoundedContext.
F.18 supplies the naming discipline for Part F and for any FPF pattern that needs a durable public term. It coordinates with:
F.5for type-name and system-role-kind-description label form;F.8for the prior decision that an expression should become a durable name rather than remain local, reused, or aliased;F.9for an actual sense Bridge between different<ReferenceScheme, LocalSenseClaim>projections;F.13for renames, aliases, splits, and merges;F.14for anti-explosion control;F.17only as a later public-row consumer whose current entry and result must accept the exact F.18 objects named below;E.10.ROLEwhen bare claim-bearing role hides its object;A.6.5andA.6.RSIRwhen relation, signature, interface, or slot wording hides the governed object;A.6.P.WMRwhen Work and Method boundary wording still hides the exact relation; andA.15.1when a candidate performed-work name still lacks occurrence grounding.
One F.18 naming decision concerns one independently governed value. F.18 supplies the candidate comparison, selected Tech designation, any Plain designation needed for a stated didactic use, declared naming use and reopen conditions. The value's direct pattern retains its kind, identity, obtaining and other subject semantics.
The NameCard's complete claim graph records the selected designation expressions, exact local sense, covered and rejected alternatives, rationale, lineage, and reopen condition.
Problem
FPF texts fail when names are treated as if they carried ontology by themselves.
- A short label appears in another context and gets treated as the same value although no obtaining Bridge establishes the exact sense relation, no separate claim says that Bridge suits this reuse, and no current reliance supports that claim.
- A role-looking name bundles a system-role kind, assignment occurrence, capability, method fit, work evidence, or authorization.
- A status-like or evidence-like phrase becomes a fake role or fake type because the row says "evidence role", "status role", or similar wording.
- A relation, declaration-local slot, interface, port, or signature name hides the exact governed object, relation-participant meaning, or rules that define or constrain the claim.
- A term chosen for convenience becomes a permanent Core-facing name without candidate comparison, rejected alternatives, or lineage.
- Local names proliferate until the corpus has several almost-synonyms and no recoverable reason for choosing one.
Recover the governed value, then record a naming settlement whose kind, effective reference scheme, exact local sense, intended use, and selected designations remain visible. Publication is a separate later relation.
Forces
Solution
Use a local-first naming protocol:
-
Recover the governed value, its kind, and its subject pattern.
-
Decide whether the expression should remain local or the current use needs a durable reusable name; apply
F.14before adding a card, cell, or row. -
For a durable name, constitute one
NameCardepisteme underC.2.1; keep the value, its kind, the card, selected designations, exact local sense, and any basis or Bridge relation distinct. -
Choose the Tech label and any Plain designation needed for a stated didactic use from the smallest candidate set that covers the live head-term families and plausible neighbouring objects.
-
Record the covered alternatives, rejected candidates, selection reason, lineage, and the condition that reopens the settlement.
-
Only for public, Core-facing, durable-across-context, or cross-context reuse, test the then-current
F.17entry. It must accept the exact governed value and kind, NameCard episteme, by-value scheme, local sense, and any actual Bridge. Public or durable reuse alone creates no Bridge.When the named use relates different
<ReferenceScheme, LocalSenseClaim>projections, F.17 must also accept the separate affirmative C.2.1 claim and current A.10 or B.3 reliance through the row rationale or notes rather than treating either as NameCard content.The F.17 result must supply the required public row. If any required input or result is absent, retain the durable name and NameCard locally, mark the public row pending, and stop.
-
Keep each cited subject, relation and later operation under its own rules. In particular, distinguish the Bridge, the separate claim about its named use, A.10 or B.3 reliance, and authorization. F.18 governs the naming settlement.
Naming Invariants
Every durable name must satisfy these invariants.
NameCard Fields
A NameCard is complete when its full interpreted naming-settlement content (ClaimGraph) is recoverable under C.2.1. That value satisfies NonemptyClaimContent; membership alone does not supply the narrower naming-settlement claims required here.
A compact card states the governed value and subject pattern, effective scheme and local-sense claim, the selected Tech designation and any Plain designation needed for a stated didactic use, candidate set, rejections, rationale, lineage, and reopen condition. The subject pattern can make the governed kind unambiguous. These claims together constitute the complete semantic content; an omitted expanded field contributes no hidden claim.
The compact cards NC-U-RELATION, NC-PROBLEM-CRITERION-APPLICABILITY-RELATION, and NC-PROBLEMATIC-FOR-RELATION use this form. Section 4.2a carries four expanded bounded-model-use cards.
Use the expanded form only when the current naming use needs the additional position:
Field discipline:
-
The card is a
C.2.1episteme.GovernedValueRefresolves to its exactEntityOfConcern;ClaimContentis the complete interpreted naming-settlement content inNonemptyClaimContent; andReferenceSchemeis the effective by-valueEpistemeInterpretationSchemeunder which those claims are interpreted. Changing any of those three values identifies another card episteme. Changing only a content or card designator, carrier, field order, or layout does not. -
The expanded form's
ClaimContentnames the complete semantic value expressed by its readable sibling fields. The examples'.ClaimGraphlabels are local content designators; a label or graph encoding is not the value.When a representation instead supplies a stored constituent designation, declare a C.2.1:4.2.6
ClaimContentDesignationwhose exact source edition and selection resolve to that complete naming-settlement content. The source may supply its whole content or an intrinsic semantic selection; resolve it before applying the content-range and naming-completeness tests.Changing a selected Tech/Plain name asserted by the card, declared use, local-sense claim, coverage, rejection, rationale, lineage, or reopen claim changes the content and therefore the card episteme even if the displayed content or card designator stays the same.
-
NameCardIddesignates the card episteme. It is not another identity discriminator and does not create a card kind. -
GovernedValueRefresolves to the exact already-governed object or value being named.GovernedValueKindRefis added when the kind is not already unambiguous from that value and its subject pattern, or when a receiving use needs the exact kind reference. For relation-facing wording the value reference resolves to exactly one of the objects distinguished in section 5.6; a field label, card, table row, or local phrase is not a proxy for that object. -
subjectPatternLocatornames the pattern description containing the exact ClaimGraph that defines or constrains the value.F.18defines only the naming-settlement predicate recorded in the card; a pattern that merely presents or teaches the name defines neither the value nor this settlement. -
LocalSenseRefin a progressive-minimum card states the exact local-sense claim directly under the card's by-value scheme.LocalSenseCellRefin an expanded card resolves to the current F.17 coordinate<ReferenceScheme by value, LocalExpression, LocalSenseClaim>and does not require a context holon.LocalSenseBasisRelationRefis present only when a separately governed relation to a basis episteme is current; a source title, card field, or publication is not that relation. -
CandidateSetrecords the plausible labels considered by head-term family. When family coverage or an exception is not already recoverable from the set, rejections, and rationale, addCandidateCoverageto state which live families and neighbouring-object readings were tested and whether any plausible alternative remains open. -
RejectedCandidatesrecords why tempting names were not selected. A usable alias is recorded in lineage as an alias, not left as a second selected Plain label. -
BridgeRefscontains only actual F.9 Bridge occurrences whose relation-semantic profiles obtain for the exact endpoint senses. It carries no naming-use direction, use-specific rule, tolerated loss, polarity, reliance, or permission. When naming across different semantic-context projections relies on a Bridge, recover the separate C.2.1 claim and its current A.10 or B.3 reliance outside the NameCard; omitBridgeRefswhen the settlement makes no Bridge claim. -
PublicRowStatusis exactly one oflocalOnly,pending, orcurrentwhen public-row use is current.UnifiedTermRowRefseparately resolves to the exact row and is present only when status iscurrentafter the section 4.4F.17entry/result gate passes. Omission in a compact card claims no row. A pending public use does not imply that a row already exists. -
RefreshConditionnames a change to the value, kind, scheme, local sense, Bridge, subject pattern or receiving use, or a repeated reader error, that requires reconsidering this settlement.
An eligible partial scheme can leave a naming claim or reference uninterpreted; resolve that gap before the exact settlement or public-row use. Known typography-only rules fail the scheme criterion. Content and scheme membership establish neither a supported local sense, an obtaining basis relation or Bridge, suitability for a named cross-local use, nor actual use.
Names such as "foundational principle pattern set", "FPF Core", "domain principle framework", and "local practice framework" require ordinary NameCard work before public stabilization under an effective reference scheme. Source aliases such as ZPF, SPF, TPF, or broad xPF labels remain intake aliases until F.18 has settled the governed value and kind, by-value reference scheme, exact local sense, rejected candidates, and admissible short form.
Current Bounded-Model-Use NameCards
The four expanded cards below record the FPFCoreReferenceScheme naming settlements used in F.17:12.4d-12.4e. Each resolves to one scheme-based F.17 cell and its separately governed local-sense basis relation. The cards state selected designations for already governed values. Establish the subject relations under A.1.1 and use E.24.PUB when a row edition must be made available.
These four naming settlements require no correspondence between different <ReferenceScheme, LocalSenseClaim> projections, so they add no Bridge or use claim.
If a named current use relates different <ReferenceScheme, LocalSenseClaim> projections, apply the F.9 predicate to the possible Bridge, identify the affirmative bounded-use claim separately under C.2.1, and apply A.10 or B.3 to the relied-on evidence or assurance claim; without that use, add no Bridge or use claim.
For ModelExpressionCoherenceRelation, an A.1.1 predicate may require an obtaining Bridge in its bridged interpretation branch; a receiving assertion or structure selection that relies on that occurrence still needs its own bounded-use claim and reliance path. The NameCard records the naming settlement; a row may cite the separate Bridge, bounded-use claim and reliance when its use needs them.
Current Role-Precision NameCards
The eight cards below record Core-facing designations for values defined or constrained by their subject patterns. They share FPFCoreReferenceScheme and need no Bridge for these naming uses. Each card cites the E.10 token-class, allowed-use and collision rules on which its designations depend.
Each card has one exact governed value and one selected Tech/Plain pair. No card is created for the SystemRole morphology, J_kindUse, a declaration-local slot, or a context field.
Demonstrative wording without a fabricated value or scheme
A.22.CGUS:4.4 permits one exact C.2.1 episteme to show a traversal through an already qualified CGUS. It does not define a demonstrative-slice U.Kind, and DemonstrativeUnfoldingSlice@Context does not identify an exact slice by itself. Before using a label such as FPFSeminarTeachingReferenceScheme-2026-07-11 for another scheme, recover that scheme by value and identify any interpretation that differs from FPFCoreReferenceScheme.
Keep demonstrative walkthrough as ordinary readable wording when a sentence already makes the exact shown slice clear. Keep mantra as bounded seminar or pattern-local recall wording when repetition and attention are the point. These local recall uses need no durable naming settlement or public row.
If a later use needs stable citation of one exact slice, first recover that C.2.1 episteme from its claim content, the qualified CGUS it concerns, and its effective scheme. Then make one NameCard only if durable naming is useful. When a second exact scheme-and-sense projection changes the interpretation, reuse its naming settlement or make a card if it needs a durable name. If a named use depends on correspondence between the projections, test that relation under F.9 and follow §4.4.1.
Availability remains a separate E.24.PUB operation. mantra move stays E.10.MOVE Plain wording for a shown E.11.PUA continuation description; it is not a durable value or a second scheme.
Rule-content names with unresolved lexical prerequisites
Consider the following R7 naming candidates under the by-value FPFCoreReferenceScheme. Keep each governed U.NameToken separate from the predicate or designation value it names.
Suppose the E.10 TokenClass, reserved-name and allowed-scope prerequisites have not yet been established for these candidates. They remain naming candidates. There is no current NameCard or public row to cite until the lexical prerequisites and naming decisions for that row are settled. These naming uses require no semantic correspondence and make no Bridge claim.
The examples show how to read each candidate expression under its stated rule. To justify an expression for a wider vocabulary, establish which defining relation its uses share and where that reading fails. For a proposed universal Kernel kind, use the independent projections required by A.8. For the name of an already qualified predicate or designation, justify that lexical choice for its intended uses. A current NameCard or public row also requires the lexical prerequisites and naming settlement.
Architectural Reference NameCard
This card names the architectural-reference form governed by E.4:4.1, E.11.PFP:4.7 and E.11.DSG: a public account of a named method language's architecture, reasons and shared explanation. The governed value is this functional publication-unit form, not one framework, Suite, Library, product series, edition, answer, lookup activity or publication occurrence. The form can be embedded or separately carried.
This naming use requires no correspondence between different local senses, so the settlement adds no F.9 Bridge or correspondence-use claim. FPF Library Architectural Reference, Engineering Suite Architectural Reference and Method Engineering Principle Framework Architectural Reference qualify the same functional designation by the explained whole. They do not name new kinds or independently maintained series.
The full title, filename, declared publication-unit key and stable SectionRef are separate; changing a title does not by itself preserve an old Markdown fragment.
Candidate Selection
Build the smallest candidate set that covers the live head-term families and plausible neighbouring-object readings that could change the decision. A cited external standard may fix the label under §4.1. One supported candidate is enough when examination leaves no live alternative that could change the choice. Stop when the relevant comparison is sufficient. If a deadline forces closure while a plausible alternative remains untested, record that exception in CandidateCoverage and make it part of RefreshCondition.
Judge candidates on:
- semantic fidelity: does the label preserve the governed value without adding or losing required conditions?
- reader ergonomics: can the intended reader recognize, say, and remember it in the current situation?
- morphology fit: does the word shape fit the kind being named?
- alias risk: will a careful reader import a wrong sense from nearby FPF patterns or external practice?
Use these as ordinal comparisons. Do not average them into one score. If a Pareto-front or quality-diversity method is used, the dimensions and dominance rule must be visible on the card.
A candidate can be selected despite a remaining weakness. The SelectionRationale must say what it buys, what risk remains, and why the covered set is sufficient for this use.
Public Term Rows
A durable local name needs no row. When public, Core-facing, durable-across-context, or cross-context reuse is current, test the then-current F.17 entry with the exact objects already recovered here. Public or durable reuse alone creates no Bridge.
The F.17 entry must be able to recover:
- the governed value and its kind;
- the locator for the pattern containing its defining or testing rule;
- the NameCard episteme, selected Tech designation and any Plain designation needed for the naming use;
- the effective by-value reference scheme, exact F.17 scheme-based SenseCell, and any separate local-sense basis relation;
- any F.9 Bridge that obtains.
If the row use relates different <ReferenceScheme, LocalSenseClaim> projections, its rationale or notes must cite the separate affirmative C.2.1 claim for the exact action, direction, rule, and tolerance, plus that claim's current A.10 or B.3 reliance.
The F.17 result must contain one row for one naming decision and state its supported citation use. Add a blocked-use claim only when an independently grounded overreading changes understanding or action, as F.19 explains.
If the entry cannot do this, keep the durable name and NameCard local and mark the public row pending. Do not repair or emulate the missing row inside F.18.
Cross-Projection Use and Reliance
Open this branch only when one named reuse must relate different <ReferenceScheme, LocalSenseClaim> projections. Compare the exact F.17 cells. Another expression under the same projection is a designation question and gets no Bridge. Different projections open the F.9 question; a different scheme is only one way projections can differ and proves no relation.
Test the F.9 predicate and cite a Bridge only when it obtains. With no current correspondence use, create no Bridge or use claim regardless of scheme count.
State the proposed naming use in a separate current C.2.1 claim whose EntityOfConcern is that Bridge. Record the action, direction, correspondence rule, tolerated loss, and polarity.
For an affirmative named-use claim, recover the independently established direct evidence relations and their descriptive A.10 evidence-provenance account. RelianceDisposition=pass supports only this bounded use and requires the evidence demanded by the claim and its direct rule.
Use B.3 only when an actual named assurance claim is current. If a direct rule requires such a claim and it is missing, return assurance-needed and recover that claim before opening B.3; materiality alone creates neither a claim nor a positive result.
A failed or non-passing claim or reliance stops or narrows the use. Neither the A.10 nor the B.3 route authorizes the use or proves that it occurred.
If the reuse did occur, recover the claimed result under its own rule: actual Work through section 5.3, an assertion episteme under C.2.1, publication under E.24.PUB, a direct relation under its predicate, or an operation application under A.6.1. Name a BoundedModelUseStructure only when that selected structure changes the sense or naming use.
Until the Bridge, separate claim, and required reliance are current, keep the names local or record the unresolved alignment. A reference-scheme or model-use-structure difference alone supplies neither a premise nor governed-value identity.
Model refinement, abstraction and relative levels
Use these NameCards when project or teaching texts need a shared name for a model comparison. B.5.MR defines two relation kinds and one kind of inquiry-relative level, each a local U.Kind. A pair of models can stand in one of the relations. Constructing or selecting a model is Work.
For the shared ReferenceScheme fix an inquiry and an interpretation. Each admissible completion satisfies its model's assertions and yields an answer to that inquiry. Include the probability claims when they matter. Refinement requires at least one completion of C, with every completion of C satisfying A under the interpretation. A is then an abstraction of C. For strictness A must allow an answer C excludes. Within a chain established this way, models admitting the same answers share a level. Equal answers across branches can coexist with incompatible constraints.
In ordinary speech a team can say “the local model is more concrete about where the check may take place” and name the choice it has fixed. Refining a model and abstracting a model name the work of constructing or selecting one. The comparative adjectives describe the models once a strict difference has been established.
F.17:12.6 gives the three citable rows and the exact sense cells for this comparison. Source-specific uses of concretization, idealization and lifting retain their own definitions.
System-Role-Kind, Assignment, Slot, and Status Naming Settlement
This settlement keeps naming aligned with the object already recovered. Bare role is a trigger handled by E.10.ROLE, not a reusable kind head.
System-Role-Kind Names
A durable system-role-kind name designates one exact local kind admitted through C.3 and A.2. Recover that kind through its candidate domain, operative membership condition, intended member/non-member boundary, and continuity rule. A practice or source reference can locate the definition or signal that two definitions should be compared; it does not identify the kind. Candidates are entities already admitted under A.1 as U.System.
The Tech designation normally ends in ...SystemRole, for example ReviewerSystemRole. SystemRole is compound morphology, not a universal governed value. The name creates no system admission, kind membership, assignment, agency, capability, or Work.
A system-role-kind name preserves the distinctions in its operative membership condition. Keep a current holder, particular assignment or Work occurrence, temporary status, and current evidence or assurance claim separate from the kind name. A skill, Method-family or other qualifier belongs in the label when the admitted kind's definition makes that distinction necessary, under §6.
If a phrase such as SeniorReviewer, NightOperator, or source wording such as evidence role appears, recover the current claim first. The result may be an exact local system-role kind, one direct assignment occurrence, a status assertion, an evidence-use relation, a Work admission condition, another governed value, or a local source phrase. Do not force all of them into one system-role-kind name.
System-Role-Assignment Names
A system-role-assignment name designates one already recoverable obtaining occurrence of an exact direct species under U.SystemRoleAssignment and A.2.1; the system-role-kind name does not identify that occurrence. Recover the admitted holder system, the exact assigned local system-role kind, and only additional participants needed to distinguish that direct species. A taxonomy, reference scheme, description, display, or generic context episteme is not a mandatory assignment participant.
Assignment extent follows uninterrupted predicate truth; an assertion or occurrence-description episteme may state a known interval separately.
A durable assignment name uses a NameCard whose GovernedValueRef resolves to that occurrence. If public or cross-context reuse is needed, apply section 4.4; until it passes, retain the card locally and mark the row pending. Neither a name, card, row, nor publication occurrence makes the assignment obtain.
Holder#Role:Context@Window is source notation only. Recover the holder System, local system-role kind, assignment occurrence and its declared species when one exists, and any separately applicable context, schedule, interpretation, or Work relation. The source token is neither a Tech name nor proof of assignment, capability, or performed Work.
Capability, Method, and Work Names
Keep these separate:
ShipbuilderSystemRolenames one exact local system-role kind;ShipbuildingCapabilitynames a capability of an admittedU.System, including an acting holon admitted as a system for that capability claim;ShipbuildingMethodnames a method or method family;HullAssemblyWorknames a work family or planning-level work label until an exact performed occurrence is current.
A role-derived or role-method-coupled expression is only a naming cue. First recover the exact value it refers to. If that value is an exact Method or Method family under A.3.1, choose a Method name. If it is an exact U.MethodDescription, U.WorkPlan, or dated U.Work occurrence, name that description episteme, plan episteme, or occurrence separately under A.3.2, A.15.2, or A.15.1; those names are not Method names. If the expression refers to another value, use the rule that defines or tests that value.
After identifying the value, use F.18 if a durable name is needed. An exact relation involving a system-role kind or assignment may constrain who may use a Method or perform Work; it neither creates nor names the Method, description, plan, or Work occurrence.
Treat an action nominal such as testing as a morphology cue, not a governed kind. Placement in function- or flow-structure prose identifies no U.Function. If the function-like use remains claim-bearing while its exact object or relation is hidden, apply A.6.F; if it is already recoverable, name that value under the rule that defines or tests it. Only then use F.18 to choose a durable name.
In a plan or assignment, a WBS element, activity or Work Package can identify intended work. To identify a performed occurrence, recover its A.15.1 basis.
A durable name for performed Work points to one dated occurrence already grounded under A.15.1. An action word, plan row, local work-family label, or U.WorkPlan does not create that occurrence or an assignment.
Every System claimed as an actual performer must already have its A.13 core, including the obtaining assignment for the exact action and any scope, situation and window required by the receiving use. A.15.1 independently admits the dated Work from its Method, temporal extent, containing System and other required direct facts.
Add F.6 only when the naming account or receiving use needs precise attribution through that same assignment; the assignment must cover the Work interval, have the recovered performer as holder and retain every participant required by its declared species. Missing or failed F.6 attribution alone leaves independently admitted Work and its durable name intact; if the same finding invalidates a consumed A.13 or A.15.1 premise, recheck that premise and its dependent conclusions.
A compact naming account cites only the identities its receiving use needs. Add a continuity policy only when interruption, retry, a changed Method or binding, or competing designators make occurrence identity material.
Name any other result separately under its subject pattern, such as A.15.PROD for a production claim or C.11 for a decision.
When the underlying boundary wording still hides the relation, apply A.6.P.WMR. Use F.18 only after an exact governed value and its use are recovered through a direct subject relation, an exact A.6.1 application binding, or an exact local A.15.PROD/A.6.RCD claim.
An exact non-assertability result retains its reason under A.6.P.WMR:4.3; it does not authorize durable naming of the attempted relation. A missing subject rule may identify a future pattern or declaration need; unavailable constructor semantics retain their separate derivation question. This section selects and tests a name. It does not define a second work-occurrence or work-result recovery algorithm.
For a method-relation or method-composition name, first decide whether the phrase names a Method, relations among Methods, a selected structure, or an episteme describing one of those objects.
- If admitted submethods make one composite way of doing, name the composite
U.Method. Use A.3.1 for that Method and B.1.5 or another direct composition pattern for its exact composition relation. - If the phrase names relations among methods without making one whole Method, name the relations first. When a named use depends on how several such relations are organized, select that
U.Structureunder A.22 and designate itMethodRelationStructure; state the question and the limits that bound the selection. Identify and constrain each included method-side relation through its A.3.1, G.5, or other defining rule. A claim-bearing episteme that describes such a relation remains separate under C.2.1, and any Work-use relation remains under A.15. - If the current object is a separately identified episteme that describes one exact admitted Method,
A.3.2may classify it asU.MethodDescription; F.18 names that episteme separately from the Method. - If an episteme instead describes the selected relation structure,
C.2.1keeps that structure as its exactEntityOfConcern; the episteme is not thereby aU.MethodDescription.
F.18 settles a durable name only after one of those exact objects has been recovered. Name a mathematical lens or representation only when it is the governed value.
System-Role-Kind Relations, Method Relations, System-Role–Method Relations, and Lens Names
System-role-kind-relation expressions remain ordinary expressions or direct relations unless their exact local kinds and relation predicates are already admitted. A mathematical description can represent a selected SystemRoleKindRelationStructure; distinguish that description from the object it represents.
First recover what the name is for:
Ordinary speech may say “reviewer” when the local sentence makes the intended system-role kind or assignment obvious. Use the concrete ...SystemRole designation when stable technical reference to the kind is needed. Do not infer kind identity, assignment, relation, Method, capability, or Work from the ordinary word alone. Add a qualifier only when it distinguishes a live neighboring reading.
Status, Evidence, Source, and Publication Names
Status-like and evidence-like wording must go to direct patterns:
- status value or status assertion:
F.10orA.19.SPR; - evidence-use relation:
A.2.4or its other direct subject pattern;A.10for the descriptive evidence-provenance account and bounded reliance; - assurance use:
B.3; - source use:
E.10.D2or source-use patterns; - description-episteme identity:
C.2.1; - multi-view publication face or form:
E.17; - availability of one selected edition through a particular expression and carrier, with reusable form conformance and form-mediated claims when current:
E.24.PUB; - gate or admission result: the relevant gate, decision, or assurance pattern.
Do not name these as system-role kinds or assignments unless that separate work-facing classification or direct assignment occurrence is current. “This standard plays the role of evidence” is repaired to the appropriate evidence-use, source-use, or status-use relation; it is not an assignment of the standard.
Relation, Slot, Interface, Port, and Signature Names
Use the direct subject pattern when the named relation, slot, interface, port, boundary, protocol, API or signature is already clear. Open A.6.RSIR when that wording still hides which object or relation is being named.
- Use A.6.5 for relation slot discipline and
SlotSpecdeclarations. - Use A.6.0 for signatures and law-defined declarations.
A.6.Mand architecture patterns define or constrain module interfaces and architecture interfaces.A.6.F, transformation, and architecture patterns define or constrain functional ports and functional structures.A.6.C, protocol, service-access, and commitment patterns define or constrain API, protocol, and service-access cases.- Use C.2.1 for the identity and content of a claim-bearing interface-description episteme.
- Use E.17 for a multi-view publication face or form.
- Use E.24.PUB for availability of the selected edition, exact expression and carrier-bearing relations, and reusable-form conformance when needed. A selected form-mediated availability claim retains one shared expression witness.
Before naming a relation-facing object, keep these settlements distinct:
One token may be reused only where the reference scheme and local sense preserve these distinctions; it cannot collapse definition, kind, occurrence, representation, and designator into one object.
F.18 can settle a durable name for the recovered value. It does not decide which value the interface word names, create a public row, or make that row available.
Words such as member, membership, belongs to, and in do not by themselves identify one reusable relation. First use E.10 to recover whether the sentence concerns mathematical inclusion, kind classification, relation participation, collection belonging, or constructive parthood.
For a collection, an ordinary sentence such as “this edition belongs to this product series” is enough unless another use needs a reusable relation name. Name a reusable predicate only under the pattern that states who or what may belong, what makes belonging begin and end, and how recurrence and past belonging are handled. Do not create a NameCard or public name for generic MemberOf merely to abbreviate the ordinary sentence.
What Belongs In The Label
Choose a head that helps the reader recognize the governed value and only the qualifiers needed to distinguish its local meaning. Use SystemRole morphology for an exact local system-role kind, and relation, slot, Method, Work or characteristic morphology only for the corresponding recovered value.
For a kind or value-family name, do not pack a current holder, temporary admission state, particular Work occurrence, evidence claim or gate result into the label as though it defined that family. A stable number, threshold or other qualifier belongs only when the subject's definition makes it part of the distinction the name must preserve.
When naming an exact assignment or performed Work occurrence under §§5.2–5.3, an occurrence designator may include an identifier, holder cue or date when needed to distinguish that occurrence. For example, the worked-case designator HullAssemblyWork-42@2026-07-15T09:10–11:35 still resolves through the independent Work-admission and identity basis. The token does not establish that basis.
Quick check: identify the value being named, remove each optional word, and retain it only when the reader would otherwise lose a needed distinction. Put current evidence, permission, status and other neighboring claims under their own rules.
Worked Cases
System-Role Kind, Assignment, Capability, Method, and Work
A shipyard team wants one reusable name for the local system-role kind used in shipbuilding Work. It first separates the values that the source word shipbuilder could hide.
Recovered values:
ShipbuilderSystemRole, one local C.3 kind whose admitted-system candidates count when they satisfy the current shipbuilding condition; the member/non-member probes and continuity rule expose the boundary, while the ShipyardProduction source only locates the definition;- one direct assignment occurrence under A.2.1 whose admitted holder system and assigned
ShipbuilderSystemRolekind are explicit, while any work area, schedule, interpretation, or reference scheme remains separate unless that direct species needs it for occurrence identity; ShipbuildingCapabilitywith envelope and measures under the capability pattern;ShipbuildingMethodor a method family under A.3.1; if a separately identifiedShipbuildingMethodDescription : U.MethodDescriptionepisteme is current, name it separately under A.3.2 only when its exactEntityOfConcernis that Method;HullAssemblyWorkunder the Work patterns.
Here HullAssemblyWork is a work-family label or a label in a plan or assignment episteme. A designator such as HullAssemblyWork-42@2026-07-15T09:10–11:35 names performed Work only when each exact actual performer has its A.13 core and A.15.1 independently admits the occurrence from the Method actually used, temporal extent, containing System, affected hull referent, material bindings, resource-use facts, and any current continuity policy.
If the naming record also expressly represents which assignment covered that Work, it cites that same obtaining A.13 assignment and adds the F.6 attribution relation; missing or failed attribution alone leaves independently admitted Work and its name intact, subject to the premise recheck in section 5.3.
A changed hull state, measurement result, evaluation verdict, delivery occurrence, or acceptance verdict remains a separately defined and separately named value.
The local card is:
The candidate comparison meets the section 4.3 stopping condition: each live head family is represented, and the recovered Method and Work objects are not synonyms for the local kind. If public or cross-context reuse becomes current, apply section 4.4; until it passes, keep this card local.
Reviewer in a Journal Context
ReviewerSystemRole designates the local kind whose admitted-system candidates count when they supply a substantive review judgment that meets the current JournalReview acceptance conditions. The candidate range, operative condition, member/non-member probes, and continuity rule recover the kind; JournalReview-2026 provenance only locates the definition. A review assignment, responsibility, authority, capability, permission, and performed review Work remain separate claims.
No F.17 row is created without a named public or cross-local reader use.
Engineer-Roboticist and Musician
A lab says: “Vasya is an engineer, does robot engineering, is therefore an engineer-roboticist. These are musical robots, and Vasya is also a musician, performs music, and teaches robots music.”
Recovered values:
- Vasya as an admitted system;
MusicalRobotLab_2026is the lab and Work locus in its direct relations, not a generic assignment participant; RoboticsEngineerSystemRole, one local system-role kind whose admitted-system candidates count when they satisfy the current robotics-engineering condition, boundary probes, and continuity rule; MusicalRobotLab provenance locates the definition but does not identify the kind;- robotics as the qualification that distinguishes this local engineering kind, with any non-monotonic restriction retained as a separate A.2.7 relation;
MusicianSystemRoleas another exact local kind when its own music-performance condition and boundary matter separately;- any current engineering or musician assignments as occurrences of their declared A.2.1 species;
- robot-engineering Method or Work, music-performance Work, and robot-music-teaching Method or Work under their direct patterns;
- an optional representation only if the project uses it to describe the selected system-role-kind relation structure.
If the exact robotics-qualified local kind has been admitted, its local naming settlement is:
If no durable qualified kind is admitted, keep engineer-roboticist as local ordinary wording rather than filling the card. Ordinary project communication may say “Vasya is our engineer-roboticist and musician” when the separate claims about his engineering and musicianship remain recoverable; any assignment is another claim.
Name a current Method, MethodDescription, or performed Work through A.3.1, A.3.2, or A.15.1. If public reuse becomes current, apply section 4.4; do not infer an F.17 row from this local card.
Method Relation Structure and Method Algebra Name
A lab says: "Use the robot-engineering method algebra: choose scouting, then calibration, then training; fall back to teleoperation if training fails."
Recovered values:
- one or more robot-engineering methods or method families under
A.3.1; - a method-family registry or selector outcome under
G.5when the family registry or selector result is current; MethodRelationStructurefor the namedMusicalRobotLab_2026use when the current claim concerns serial composition, guarded fallback, or family selection among exact methods;- a C.2.1 description episteme whose EntityOfConcern is that selected MethodRelationStructure; it is not thereby a U.MethodDescription;
- a
C.29mathematical-lens use when "algebra" is the selected representation for checking composition, fallback, or preserved/lost structure; - work plan or dated work only when a concrete plan or occurrence is current.
F.18 settlement: RobotEngineeringMethod names a Method or method family only when that is the governed value. RobotEngineeringMethodRelationStructure may name the selected method relation structure when durable naming is needed.
RobotEngineeringMethodAlgebra names the lens only when the algebraic representation itself is the governed value. Do not use a system-role-kind label such as RoboticsEngineerSystemRole to name the method relation structure, and do not use method algebra to hide a WorkPlan or performed Work.
Evidence-Like Source Phrase
A review table contains the phrase "model card evidence role".
Recovered values:
- a model-card episteme;
- an evidence-use relation to a target claim;
- possible source-currentness and assurance-use relations;
- no system-role kind, assignment, or acting system merely because the episteme is used as evidence.
F.18 settlement: mint no system-role-kind or assignment name. Recover the exact evidence-use relation under A.2.4 or another direct subject rule, including its participants and obtaining condition. A.10 supplies only its descriptive evidence-provenance account and bounded-reliance disposition.
If durable naming is needed, compare names for that independently recovered relation; ModelCardEvidenceUse can be a candidate, not a relation created by the phrase or card. Then apply §4.4 only when a public row is needed; otherwise retain the local settlement.
Interface-Like Source Phrase
A software team says "the payment interface owns customer identity".
Recovered candidates:
- module interface under
A.6.M; - API description or protocol under
A.6.C; - signature or SlotSpecs under
A.6.0andA.6.5; - claim-bearing interface description under
C.2.1; - multi-view publication face or form under
E.17; - publication availability, exact expression, reusable form conformance, form-mediated expression, or carrier bearing under
E.24.PUB; - a system-role assignment under A.2.1 only when an occurrence belongs to a declared species, has an admitted System as holder, and has the local kind as assigned-kind value; any responsibility or authority relation remains separate.
Do not mint PaymentInterfaceRole from the ambiguous phrase. If the speakers mean that the API specifies a customer identifier, name that field through its declaration. If they mean responsibility for the identity data, identify who holds that responsibility and under which rule. Ask which claim they mean if the phrase leaves it open. Apply F.18 when the recovered value needs a durable name.
Cross-Context Name
Two teams use component, module, and unit for nearby meanings.
Recovered values:
- structural component under architecture and part-whole patterns;
- deployable module under module-interface patterns;
- management unit under organizational patterns.
F.18 settlement: first keep the three recovered values and their local labels separate. If only local speech is needed, stop there; do not name a claim merely because one team wants to explain the difference.
If a public term use is proposed between different <ReferenceScheme, LocalSenseClaim> projections, identify the exact source and receiving F.17 cells and test the F.9 Bridge predicate between them. The same scheme with different LocalSenseClaim values qualifies; a different scheme only opens the question and never establishes the relation.
When the Bridge obtains, state in ordinary C.2.1 wording whether it is suitable for this naming use, naming the direction, label-correspondence rule, tolerated loss, and polarity, and establish the current A.10 or B.3 reliance required by section 4.4.1. The Bridge does not choose the Tech label, the claim does not identify the governed value, and neither authorizes or performs publication.
For this cross-projection use, apply F.17 as specified in section 4.4 only after the Bridge, the use claim and the required reliance are current. If that public-row use lacks a required input, keep the name and card local and mark the row pending. A team that wants to publish its own term without claiming correspondence can use §4.4 directly. If only local speech is needed, keep the local wording.
Anti-Patterns And Repairs
Conformance Checks
Use these checks before a durable name is reused in a pattern. If an F.17 row is current, run its own row checks after the section 4.4 gate; these F.18 checks neither create that row nor establish a publication occurrence for it.
Regression checks:
- When either the effective reference-scheme edition or the
LocalSenseClaimchanges, compare the resulting semantic-context projections. Re-check any obtaining Bridge, the separate claim about the named use between different projections, and that claim's current reliance; same-projection expression changes stay with designation, and no current correspondence use creates no Bridge or use claim. - When a system-role-kind description changes in a way that may alter the C.3 candidate domain, membership distinction, member/non-member boundary, continuity, or the naming settlement's reader meaning, re-check the local kind name and any assignment name that depends on it. A provenance-only edit does not split the kind.
- When a subject pattern changes the named value or the interpretation of borrowed morphology, re-check the affected name.
- When repeated reader errors occur, reopen candidate comparison instead of adding aliases indefinitely.
SoTA-Echoing
Question and selected answer. How should one already identified value receive a durable name without turning a convenient word into a different object or making every local phrase into a maintained record? Under E.8:11, the best-known answer for this bounded use is a local-first settlement. Stop at sufficient ordinary wording. Otherwise compare plausible head-term families against the same value and reader situation, keep the Tech designation and any Plain designation needed for a stated didactic use, and record why it was chosen and what would reopen it.
Serious alternative. A low-cost option is a compact preferred-label entry with alternatives and a scope note. SKOS Reference, lexical labels, documentation, and mapping properties supplies that comparator and the useful separation of labels, concepts, notes, and mappings. It can carry a careful explanation and does not claim that a shared label proves identity. F.18 does not reject it for lacking FPF field names or require a different storage format.
The remaining choice is about the naming decision. A preferred label and scope note can state what a term means while leaving unclear why a neighbouring head was rejected, whether ordinary wording would suffice, and which later use needs a durable settlement. F.18 makes those distinctions explicit in the naming decision.
The additional comparison and note are needed only for a reusable name. When a terminology entry already carries the same value, candidate comparison, use boundary, and reopen reason, reuse that content; neither a second naming decision nor corpus-wide normalization is warranted.
Adapt and reject by value. Section 0 and steps 1–3 of section 4 keep the value and ordinary-wording exit before the card. Sections 4.1–4.3 require semantic fidelity before reader familiarity and make candidate coverage and the remaining risk inspectable.
Cases 7.1 and 7.2 expose the concrete cost of a short head that confuses a system-role kind with an assignment, capability, Method, or Work; 4.2e compares the architectural-reference function with introductory placement, instructional, decision-history and finding-aid readings for one publication-unit form. The cases show which object distinctions the candidate comparison preserves.
Section 7.5 and the public-use branch in 4.4/4.4.1 preserve local wording and test a needed correspondence separately; a label or generic mapping is not authority for that use. Reject choosing by familiarity alone, adding a card for every phrase, or treating a source's preferred label as the identity or admission rule for the named thing.
Judge reader ergonomics in 4.3 against the actual candidate and intended readers. Prefer a shorter label when it still fits the governing naming rule and helps the intended readers recognize the same value for the same admitted use. C.18 supplies comparison discipline only if Pareto or quality-diversity methods are used; it is not independent evidence that this name wins.
Reopen. Compare again if a lighter naming procedure preserves the same object distinctions and later reuse with less effort; if readers still infer the wrong object from the chosen head; or if the actual use needs linguistic or multilingual modeling that the simple settlement does not support. Use evidence about the naming procedure and its receiving use to make that comparison.
Currentness rule: when the pattern containing a value's direct rule, C.2.1, F.9, A.10, B.3, or E.24.PUB changes the value, card, sense, Bridge, bounded-use claim, reliance, or publication boundary, reopen only the affected invariant, field, case, or check. A future F.17 edition is consumed only through section 4.4; its change does not reopen local NameCards unless their supported public citation use or object references change.
Relations
Builds on F.0.1, F.1, F.2, F.3, F.5, F.8, F.9, F.13, F.14, F.15, C.2.1, and E.24.PUB.
Coordinates with:
A.2,A.2.1,A.2.5,A.2.7,A.15,A.15.1, andF.6for system-role kinds, system-role assignments, assignment-state predicates and direct state relations, relations among system-role kinds and selectedSystemRoleKindRelationStructure, system-role–Method–Work alignment, performed-Work occurrence grounding, and the separate Work-to-assignment attribution;A.3.1for method and method-family names;A.3.2for a separately identifiedU.MethodDescriptionepisteme whose exactEntityOfConcernis that Method, and for the description episteme's separate name;A.6.P,A.6.P.WMR,A.6.RCD,A.6.REL,A.6.5,A.6.RSIR,A.6.0,A.6.M,A.6.F, andA.6.Cfor relation-claim settlement, work/method-boundary relation recovery, relation-kind and occurrence boundaries, slot, signature, interface, port, and protocol names;A.2.4or another direct subject pattern for evidence-use relations;A.10for their descriptive provenance and bounded reliance;B.3for an actual named assurance claim;F.10,E.10.D2, andC.2.1for status-use, source-use, and description-episteme names;E.17for multi-view publication-face and publication-form use;F.17only after its current entry accepts the exact F.18 value, kind, card, and sense result and, for reuse between different semantic-context projections, the separate obtaining Bridge, affirmative C.2.1 use claim, and current A.10 or B.3 reliance; otherwise the local NameCard remains sufficient and the public row stays pending;E.24.PUBfor the separate occurrence, form, carrier, audience, bounded-use, and currentness objects needed when an exact row-episteme edition is made available;C.16,C.18, and Part G search patterns when candidate comparison uses Pareto or quality-diversity vocabulary.
Constrained non-use:
F.18names independently governed values. ANameCardis a separately constitutedU.Epistemeunder C.2.1; its constitution does not establish the value it names.- Do not use
F.18to decide whether two locally interpreted values are identical. A Bridge between exact F.17 cells can obtain only under the F.9 predicate; a separate C.2.1 claim states one proposed naming use between those cells, A.10 or B.3 supplies its reliance rule, and any governed-value identity claim must independently satisfy the direct value rules. F.18does not turn a publication row, card, table, or glossary entry into the thing being named.
F.18:End
Ontology-First Plain Technical Rewriting
Type: Plain-technical precision-restoration pattern Status: Stable Normativity: Normative for FPF-governed technical prose unless marked informative; informative for external source prose until it is rewritten for FPF use
Plain-name. Ontology-first plain rewriting.
Intent. Repair technical prose that is grammatically plausible or locally true yet makes the reader supply an unsupported relation, participant, alternative, list meaning, or rhetorical branch. First recover the governing object, claim, action, required operands, referents, and kinds; then remove apparatus and other structure that contributes nothing to the intended use.
The normal result is repaired text, not an audit form. Preserve every technical distinction and operational detail that changes truth or action, and route unresolved FPF wording to E.10, E.10.ARCH, E.10.ROLE, A.6.F, F.18, or the subject pattern that defines it.
Builds on. E.8, E.10, E.10.ARCH, F.18, A.6.P, A.7, E.18, E.21, and the subject patterns governing the repaired span's claims.
Coordinates with. E.19, E.22, E.23, A.19.SPR, C.2.P, C.16.P, C.30.P, and E.11.
Use this when
Use F.19 when a bounded piece of technical prose is harder to understand or use than its intended claim requires.
The sentence may be grammatical and every isolated statement may be true, yet the reader still has to invent a missing operand, accept an implausible relation, guess what a pronoun or relational noun refers to, interpret a list with no stated purpose, or wait through caveats and ornament before reaching the governing message.
Common signs are:
- a verb or relational noun whose needed participant is not cheaply recoverable;
- a grammatical subject that cannot bear the asserted predicate, even when that predicate appears inside a denial;
- a contrast, warning, or guard against a reading that no plausible intended reader has reason to make;
- one head or predicate imposed on unlike members;
- examples presented as a classification, or a catalogue presented instead of a proposition; and
- coordination repeated inside phrases and across clauses, or stacked modifiers, when one governing statement would do.
Item count is only a cue. Two coordinated members can already be needless, while a long inventory can be exact and useful when its kind, membership rule, and closure matter.
Matching kinds and individually relevant members do not by themselves justify a series: the reader must need to distinguish or retain the members together for the intended use.
Apply the same method to FPF pattern prose and to other technical prose whose accepted domain terms, relations, claim boundaries, or use conditions must survive simplification.
What goes wrong if missed. The prose looks careful while introducing relations, alternatives, or branches that the work does not need. A later author or generator may then copy that shape as an acceptable technical style.
What this buys. The passage puts the supported object, claim, and action where the reader can find them without first working through unnecessary caveats. The rewrite preserves required technical distinctions and removes invented foils, false agency, reference puzzles, and catalogue rhetoric.
First useful move. State in one plain sentence what the intended reader must recognize, understand, decide, or do. Then read the whole natural span against that sentence before changing individual words.
Not this pattern when.
- If only one already-visible FPF word or head has an unresolved technical use, follow the
E.10entry for that use. - If the question is a durable reusable name, use
F.18. - If clear advice still demands work whose contribution, feasibility or requirement merits are unresolved, use
C.11.DUAto repair that advice. Return here for any wording repair it needs. - If source prose is only being observed and not admitted into governed technical prose, keep the observation source-side.
- If evocation, rhythm, ambiguity, or parallelism is the declared work of a poem, quotation, ceremonial passage, or other expressive genre, do not flatten it into technical instruction. Apply
F.19only to the technical claim or action that must remain recoverable. - If a language-specific grammar or idiom remains after the common semantic repair, use the applicable language profile.
Primary EntityOfConcern in plain terms. One sentence, row, paragraph, list, or small coherent section being repaired into precise plain technical prose.
Problem frame
Local truth is necessary but not sufficient for useful technical prose. “A mouse is not the Eiffel Tower” is true, but it introduces an Eiffel-Tower reading that the reader had no reason to construct. “The evidence does not notice the error” is also locally true, yet the denial makes evidence the subject of an impossible noticing relation. “The result is not a final scheme of the world” invents a grand alternative before the sentence reaches its result.
The same failure appears without negation. “Then pour” can omit the thing or destination that determines the operation. “Bearer” can leave the reader asking bearer of what. A grammatical series can give examples, methods, activities, and outcomes one false common head. Several individually valid pairs can create catalogue rhythm while never stating the proposition they are meant to support. Scenic or defensive detail can delay an urgent event or requested action.
One connected repair therefore answers two questions:
- Semantic completeness: can the reader recover the predicate, its required participants or operands, local referents, member kinds, and the relation actually asserted?
- Pragmatic contribution: does each explicit alternative, modifier, guard, list member, and extra proposition change what the plausible intended reader can recognize, understand, decide, or do?
A word class or syntactic form alone does not establish a defect. Negative polarity can be the claim. A documented anti-pattern can quote the real error. A visible diagram feature can make one overreading plausible. Ordinary metonymy and ellipsis can be clearer than formal expansion. Judge the supported relation and the receiving use, not the presence of not, a comma, or a particular verb.
Problem
How can a practitioner repair technically plausible prose that asserts unsupported relations or makes the reader invent missing structure, while preserving the kinds, claim boundaries, operational detail, and established terms that the intended use needs? How can the practitioner do this without building a controlled language, a universal ontology of speech, a prohibited-word list, or a form for every correction?
Forces
Solution
Use OntologyFirstPlainRewrite as one connected reading and repair over a natural sentence, row, paragraph, list, or small coherent section. Take the intended reader and use from the surrounding work; do not invent an adversarial reader or a persona form.
One connected reading and repair
-
Establish the reader's use and state the governing message. From the passage, the request it answers and the material normally available to the reader, establish what this reader needs to recognize, understand, decide or do. A stated question or task, heading or surrounding explanation can supply the grounds. Keep a passage's local function distinct from the larger explanation the recipient requested. If different supported uses need different content, retain that uncertainty and obtain the missing purpose before judging sufficiency; continue semantic repairs that do not depend on it. A clear claim can have an unclear use. If the decision depends on composition scales or levels of model abstraction, apply F.19:4.3.1.1 during this reading. Then say what object, claim, action, event or distinction the span needs to convey. Mark process traces, status language, reference boilerplate, quality proof, defensive caveats, ornamental detail and other apparatus that may be displacing it. Truth or polish alone does not make the apparatus useful.
-
Recover the predicate and its participants. Identify the operation and every participant that changes it. This may be an actor, object, source, target, result, or another required operand. Apply the same question to verbal and relational nouns: recover of what, for what, between what, or another required participant. Leave an argument implicit only when one intended value is cheaply and uniquely recoverable from the local span.
-
Check predicate compatibility. Recover what the complete claim asserts under its negation, modality, or conditions, and test the subject and participants under the intended literal or metonymic reading. A positive assertion must assign its predicate to a compatible kind. A denial may correct an evidenced type mistake under the plausible-reader test below; without that ground, “evidence does not notice the error” introduces an idle alternative.
Keep ordinary metonymy when the relation is established and the capable participant or work remains recoverable: a diagram may show, a framework may help, a reminder may cue, and a constraint may limit.
-
Resolve referents and kinds. Pronouns, demonstratives, omitted heads, and repeated labels must select one locally appropriate referent. Preserve the object kind, claim or relation kind, slot or relation position, use and publication boundary, and flow distinction whenever one changes the claim. A shared grammatical position does not make different FPF kinds interchangeable.
-
Test contribution. Read the complete span with and without each optional contrast, guard, modifier, example, coordinated member, or extra proposition. Remove it when the plausible intended reader can recognize, understand, decide, and act in the same way. Local truth and grammatical fit do not earn a phrase a place by themselves.
-
Resolve coordination and lists on two axes. Recover the proposition or action served by the series, its intended scope, and the relation among its members before choosing the form. Use one kind or predicate only when it fits every member. Distinguish a closed set, illustrative examples, alternatives, a sequence, several direct relations, and a failed ontology. A closed set needs its kind, membership rule, and closure. Connect illustrative examples to the proposition or kind they illustrate, with a non-exhaustive cue when a plausible reader could mistake them for a classification or a boundary of applicability.
Usually state the proposition first. When an explanation starts from cases, make the transition to its general claim explicit and supply grounds for that claim's scope.
Then decide whether the reader needs to distinguish or retain the members together. If so, preserve the series and the differences that matter to understanding or action. Otherwise state the governing claim or relation directly, retaining its scope and needed grounds. A question about one case may be answered by that case; a general explanation retains its general proposition and reasoning when illustrations are shortened.
Test discourse load: keep each optional member for its distinct contribution; reduce repeated coordination and modifier chains that make the reader retain needless branches or postpone the governing message. Length is a reason to inspect the series. Decide from what the series contributes. If a series hides an FPF kind, relation, or structure that ordinary reading cannot recover, use the pattern that defines or tests it, or return the unresolved meaning as a blocker.
-
Foreground, rewrite, and compare. Put the governing event, claim, requested action, or decision before optional atmosphere, examples, caveats, and catalogues. A prerequisite may come first when it is needed for safe interpretation or action. Write the shortest ordinary technical sentence that preserves every live predicate and participant, established term, polarity, and action-changing detail. Such detail can include quantity or threshold, sequence or timing, criterion or tolerance, exception, and applicability.
Compare before and after: any unsupported change of kind, relation, scope, use, currentness, or operational effect is a loss and blocks the rewrite unless another accepted decision authorizes it.
Keep ontology visible only where it carries the sentence. A term-source or type annotation is needed only when it changes how the reader identifies the object, kind, relation, slot, use, publication boundary, admissible use, or applicable rule. A record, card, table, schema, data structure, dashboard, or named form remains apparatus unless it carries one of those values.
If ordinary domain wording already preserves them, keep the ordinary sentence. "The aircraft flies" is better than a typed expansion unless the flight function, system kind, or slot relation is under repair.
Keep what a passage claims and teaches through a change of wording
Read from the opening question to the decision, instruction or conclusion. What does the reader have to identify? Is the passage about one item, a kind and its criterion, a model or specification, a lot, or a family with a stated grouping rule? Is it reporting a dated action, explaining a repeatable way to act, or assigning the next job? The nouns can all be correctly typed while the answer slips from a rule for further cases into the story of one successful job.
If a second reading is plausible, try both on the work this reader has to do. Could a technician repair the wrong unit, a designer revise the product drawing when one unit needs repair, or a reviewer accept one successful check as grounds for every unit? Establish what the source actually supplies. State the missing identifier, criterion, condition or relation at the point where the readings diverge. If the evidence cannot decide between them, keep both readings open and obtain the particular fact the decision needs. Where the document already identifies the right unit or operation, use its ordinary name.
Follow the method as well as the object. If the text promises to teach a way to check an item, recover the operation, the conditions under which it applies and what would count as success. When repeatability is in question, try the wording on another item to which the rule should apply. The very same case may admit two ways to do the work. If the passage says which method was chosen or prescribed, identify the separate grounds for that choice. An account of what the technician did on Monday supplies one performance. Recover what permits repeating the check and which choices must be made again. Conversely, a report that one item passed calls for the record of its test. General instructions alone cannot establish that occurrence. Keep the field's own rule in view when moving from a transdisciplinary distinction to a discipline, or from that rule to a local procedure or job.
Write for the decision in front of the recipient. A furniture designer can write “Use drawing T, revision 4, for the seat joint.” The workshop order can read “Repair the loose joint on stool 17.” For an order to inspect the May lot, give the lot reference. If inspecting any available stool is sufficient, say so. One agent may need any of these forms. In “giraffes have long necks” the plural can express a characteristic of the kind. In “three giraffes came to the waterhole” it reports observed individuals. Singular wording can also speak of a kind or of one individual. Quantity, criterion and referent come from the statement in use. An explicit “all”, a particular set of items, or an exception belongs in the text when it changes what the reader may conclude or must do.
Reread the whole passage after the edit. Which object would two competent users pick, what claim could each support, and what operation would each perform? Can they still say why moving from the case to the wider rule, or applying that rule here, is warranted? If both readings already lead to the intended action for the same reason, the familiar wording is enough. Use C.3 to test a candidate against the kind criterion. Use A.2.6 to check whether a claim covers the conditions of use, and A.3.1 to identify a repeatable method. When a change of subject affects the conclusion or action, use C.39.WA:4.3 to find the relation and the grounds the subject practice supplies. When rewriting a model or specification, compare what the earlier and revised words allow under the same conditions. An example allowed by both can conceal a lost alternative or a newly permitted one. Use B.5.MR if the difference changes a decision.
First establish whether the passage lets the reader recover what each model asserts. Giving an ambiguous term its supported meaning may reveal the same assertions; it does not by itself make the model more concrete. Use B.5.MR when a claim about refinement or abstraction needs to compare the answers admitted by models whose assertions and interpretation can be recovered. The delivered wording is enough to begin this recovery; its production history need not be known.
If a needed assertion or relation still cannot be recovered from the passage and its available definitions, identify that gap and the inference or action it prevents. Compare each operation promised to the reader with what the passage enables. Use C.39.GA when the argument covers only a narrower class or when the promised way of explaining, choosing or constructing is missing even though every stated conclusion is sound. Keep a sufficient method at the reach it actually covers. For a missing subject premise, derivation or operation, obtain what the recipient needs from the corresponding practice. For the whole explanatory attempt, EXD.Reference:3 follows the conclusion to its premises and operations and then forward to the answer. Bring back the supported assertion or operation, its grounds and any unresolved condition; express it in the passage and repeat local recovery. A clearer sentence alone does not establish the missing premise.
Check which composition scales and model abstraction levels a claim covers
Read the instruction as its recipient will receive it. Could someone follow it exactly and still do different work from the work promised? When wording such as “at all levels” affects the answer, recover what is ordered, which members the speaker included and what is asked about each or their connections. C.39.QR:4.1.1 gives the echo question and shows how to handle an emphatic use, a literal universal and unresolved readings. “What can vary?” lets a reader substitute another case. For a conclusion about composition, recover the relevant kind of holon, its whole and constituents, and the range of scales for each kind. Which ranges does the commission give, and which must the reader choose? When that selection is open, use C.39.SA. Look for alternatives the model merges although the recipient needs to distinguish them, and for choices the passage fixes although the recipient needs to keep them open. If the conclusion connects different kinds, recover the relation it needs, such as which dated work enacted the Method. If the decision uses several levels of model abstraction, ask what each model leaves open, which conclusion carries between them and in which direction. A more detailed account may leave the same alternatives open. Several models may supply different premises for one conclusion. Distinguish a rule that holds at each composition scale, a dependence across scales and a result of parts working together. A short question intended for separate use must carry the distinctions the reader needs there. A familiar term with an established meaning in the accompanying text can suffice.
When the question changes, return to the original question (C.39.QR:4.1). Read a restatement, delegated instruction or returned result as its recipient will encounter it. What could the recipient actually do or conclude by following it, and what work did the current commission call for? A passage can change the subject, who chooses, the population, a composition scale or what the model distinguishes while using the same terms throughout. Make the reason for the change clear to whoever will use the result for the original question. Establish what the new result warrants and whether it settles that question, supplies a premise or leaves it unresolved. Restore a displaced assignment or conclusion. A deliberately bounded inquiry can finish with its own answer. Give its performer the task and information appropriate to their part, preserving any choices or discoveries entrusted to them. The person or system drawing the wider conclusion is responsible for any grounds still needed. Use the revised commission if an authorized decision has changed it.
If the intended operation or its reach is unclear, restore it from the commission. If the meaning is clear but the argument offered for the whole range only works for a case, subclass or one permitted arrangement, use C.39.GA, Restore a Missing General Argument. Bring the repaired inference back to this passage and check the conclusions that depend on it. A general argument can already be complete inside one example. A report of progress can give a supported partial answer and explicitly leave the broader question open.
If the disputed step concerns a different relation, check that relation. For “layer”, C.30.STRAT:4.2 distinguishes a functional grouping from a scale of composition. To establish that a Method is part of another, use the construction of the encompassing Method under B.1.5.CM. For a claimed abstraction or refinement of models, compare what each admits for the same inquiry under B.5.MR. A transfer of one conclusion may need only a ground for that transfer. C.37.MM:4.1.1 helps when several kinds, scales or models must support one decision together. A claim that a permitted Method was actually performed needs evidence about the work under A.15.1. State what the available grounds establish when the stronger conclusion remains open.
Restore a technical head when the meaning is still ambiguous
Precision before a coarsened rendering. When head kind, qualifier claim, or comparison basis remains unresolved in FPF-governed prose, use this working order:
Restore the head kind first. A narrowing qualifier such as comparative, safe, interactive, or reliable does not by itself restore that kind. Unpack the qualifier claim and then check whether the comparison or escalation basis is homogeneous. Only after that may a later Plain, didactic, or coarsened rendering relax the sentence, while keeping the more precise upstream interpretation recoverable.
Judge homogeneity against the comparison or condition rule actually used; recover distinct relations separately when that rule combines them. The basis may be a homogeneous claim-kind criterion, threshold, or named defining, constraining, or source-relation condition.
Treat exact, direct, current, governed, subject, owner, defining, and similar qualifiers as content only when they distinguish live alternatives. Remove them when no such contrast changes the truth, action, stop, or reliance.
A PatternID may remain an ordinary citation; expand it into a claim-bearing episteme, ClaimGraph, U.MethodDescription, U.Method, actor, assignment, U.Work, or another formal identity only when the current claim or a named later use depends on that distinction.
Keep ordinary practitioner action and instrumental pattern-use wording ordinary when it does not assert a particular dated Work occurrence. “Use E.9 to record the decision” and “the framework maintainer compares the editions” need no invented Method, MethodDescription, performer, assignment, or Work identity.
Open the identity-bearing branch only when the sentence deliberately asserts a particular dated U.Work occurrence. Then point to its basis: A.13 first, independent A.15.1 Work admission second, and F.6 afterward only for precise assignment-bound attribution. Add a local system-role kind or a separate System-classification judgment only when that neighboring claim matters.
Treat a pattern episteme as a U.MethodDescription only after A.3.2 establishes that it has an already admitted Method as its exact EntityOfConcern and makes at least one substantive claim about that Method as a way of doing. Otherwise cite the applicable pattern content as guidance and use A.3.1 for the Method itself.
Plausible-reader guards and cold-reader recovery
Use two reader tests for different decisions.
- The plausible intended reader has the knowledge and task presupposed by the text. Use this reader to decide whether a foil, guard, warning, or contrast deserves mention. Do not substitute an adversarial reader who can imagine any false inference, or the author who already knows the answer.
- The cold intended reader lacks the author's private context and unpublished notes. After the rewrite, ask whether this reader can recover the object, predicate, participants, relevant kind or ordinary status, relation, action-changing detail, and next useful action.
Recover the instruction from the publication, its stated prerequisites and the knowledge presupposed for its intended reader. An application-test scenario may supply task-specific inputs, such as workload or available resources. Distinguish those inputs from explanations of the instruction supplied only for the test.
If only the test scenario explains the action denoted by B2, successful application demonstrates use of that supplied explanation; the publication's wording question remains unanswered. Name or reference the action in the publication, then check recovery without the test-only explanation.
Retain a negative alternative, denied consequence, warning, or non-use statement only when three conditions hold. The particular reading being rejected has an independent local ground. A plausible intended reader could adopt that reading here, including through an evidenced type mistake. The distinction changes truth, understanding, selection, safety, stop, reliance, or action.
An earlier or source claim, an observed recurring mistake, a serious competing position, a visible representation feature, or an applicable safety risk can supply the ground. The guard itself cannot.
Use the smallest clear correction for a grounded guard. When actor allocation is the useful content, state it positively: “On receiving new evidence, the reader decides whether to reopen checking or revision.” When currentness is the useful content, state the direct use: “This guide conveys the seminar of 1 February 2026; check current rules against the current FPF edition.” Keep material negation, documented anti-patterns, fair disputes, and safety stops when their polarity or boundary is itself the claim.
Basis and coverage of a recovery judgement
State whether the judgement rests on an expert walkthrough, an actual reading, or a formal-model estimate, using the distinctions in C.2.8:4.5.
An expert can judge recoverability from the public text for the intended reader. An actual response establishes what that reader recovered under those reading conditions. A formal-model estimate retains its stated observer, selected structure and model conditions under C.2.8:4.6.
Use expert reading for a bounded wording judgement when it resolves the question from the permitted public inputs and no actual response is required. Obtain actual public-only recovery when:
- the intended conclusion claims that a reader actually recovered the instruction;
- plausible reconstructions still differ in an action-changing participant, operation, result or condition, and the reader's response can change the repair or use decision;
- it remains materially uncertain whether the publication supplies the needed instruction or the evaluator supplies it from private knowledge or test-only explanation; or
- demonstrated misses undermine the sufficiency of the earlier expert reading for the affected instruction family.
A visible wording defect can be repaired directly. Select any reading still needed for the repaired text and intended conclusion.
Give a cold reader the publication, its stated prerequisites and task inputs, withholding private rationale and suggested answers. Preserve the initial recovery before supplying further explanation; qualify subsequent source returns or help under C.2.8:4.3.
Limit a recovery conclusion to the natural text units and relations examined under its actual evidence basis. A whole-publication conclusion needs coverage of the whole publication, including relations needed to read its parts together. Successful application probes cover their tested uses; they leave unexamined text outside that result.
A whole expert reading combined with local actual readings retains that mixed basis.
Reuse a result for the same or explicitly unaffected text, recovery question, reader preparation, access, assistance and budget when those conditions still apply. A local revalidation by a reader familiar with the earlier wording retains that qualification.
State a material limit in the receiving judgement or result already needed by the work. When a required reading is unavailable, identify the missing basis and leave only the dependent conclusion open.
Correct recovery of what the text asserts and the subject-matter warrant or practical adequacy of that assertion remain separate questions under C.2.8.
Result and local revalidation
Keep wording that already serves the reader. Where a change is needed, return the repaired text or identify the meaning that still needs to be established. Do not require a separate record for each facet of the reading, whether a form, card, table, progress row, or written answer.
After changing words or syntax, reread the changed sentence with the passages needed to determine its referents, predicate, participants, contrast, modality, support, action, and result. Revisit a directly affected instruction wherever it appears.
The earlier semantic verdict does not transfer to new wording. Reuse a judgement about an unchanged passage when the repair leaves its meaning and use unaffected. A local edit does not require a new reading of the whole document.
When a named high-risk or disputed decision needs inspectable evidence, show the before text, repaired text, live values that had to survive, and any unresolved blocker. Use the receiving decision's existing comparison or review result. The optional fields below can structure that result when its receiver needs them; ordinary corrections need no separate form.
Before finishing this reading, can the recipient recover the intended work and carry it out at the reach the passage promises? For unresolved wording, consult E.10, E.10.ARCH, E.10.ROLE, A.6.F, F.18 or the definition needed. If following the text or using its result would answer a different question, use C.39.QR:4.1 within this reading to recover the intended work and check what the available result supplies. When the recipient needs to choose composition scales or levels of model abstraction, use C.39.SA through F.19:4.3.1.1. That section also identifies the relevant subject method for a disputed relation or functional layer. If a narrower way of arguing or explaining has displaced the promised one, return the missing step through C.39.GA. Follow the repaired step only into the conclusions it affects. Finish when ordinary reading and the available grounds suffice for the assigned use.
Result form
Use this optional form when the receiving decision needs the corresponding inspectable detail.
Pattern-prose specialization
When the repaired prose is an FPF pattern, apply the same method with one purpose test:
Does this sentence help the pattern's intended user recognize and perform the pattern, or does it record development, review, projection, landing, quality, or source-management evidence about this version?
If it records evidence about the pattern version, keep that evidence outside the pattern unless the pattern's own primary EntityOfConcern is that evaluation or projection object. The evidence can cause edits to the pattern; it is not automatically pattern content.
Pattern prose keeps:
- the pattern's own primary
EntityOfConcern; - the first useful move;
- the practical delta and cost of missing it;
- a local boundary that passes F.19:4's full independent-ground, plausible-reader, contribution, and smallest-clear-correction test; and
- short references to related patterns after the pattern's own content is visible.
Public architectural reasons stay with the explanation when they help the intended user understand, select, combine, or adapt the Methods and profiles. Apply E.8:4.2.3 to distinguish those reasons from the current version's development history.
Pattern prose moves out:
- rationale for the current draft's placement or promotion as a development decision;
- correspondence about producing or reviewing the draft rather than using the pattern;
- quality, projection, monolith-parity, landing, and source-management evidence; and
- repeated boundary doctrine already carried by another pattern.
Archetypal Grounding
These cases show repairs and situations in which ordinary wording should remain.
Illustrative case — expert preparation and public wording. A production-planning guide says, “Use their exact result for the plan; it is not a completion certificate.” A private briefing identifies the result as the forecasting team's demand estimate in Report R for 1–7 June under scenario S. The planner must compare that demand with available production capacity; a changed period or scenario requires an applicable estimate before comparison.
An expert who knows the briefing can reconstruct this operation. The public sentence still leaves the result, comparison operands and change condition unstated, and the case supplies no reason for its reader to expect a completion certificate.
Repair: “Compare the demand estimate in the forecasting team's Report R with available production capacity for 1–7 June under scenario S. If the period or scenario changes, obtain an estimate for the applicable period and scenario before comparing.”
The repair makes the result, operands, applicability and action order public. It removes the empty intensifier and unsupported certificate alternative. The expert's reconstruction from private preparation and recovery from this repaired public instruction have different reading conditions.
Bias-Annotation
F.19 favors direct technical prose that its reader can use. The protected value is kind-preserving clarity, not brevity by itself. A longer rewrite is better when it restores a participant, relation, boundary, or operational detail that the declared use needs.
Conformance checklist
These questions guide one connected reading; they do not require separate recorded answers. KindPreservationCheck names the comparison required of every rewrite. Its separate result form is optional under F.19:4.1.
Common anti-patterns and how to avoid them
Consequences
Technical prose becomes easier to trust and use because every asserted relation has supported participants, every retained guard answers a plausible question, and lists serve a visible proposition or action.
A repaired passage can give later authors an example of direct wording. It shows guards and lists where they serve the reader, without presenting defensive denials, false symmetry or exhaustive-looking catalogues as marks of precision.
A routine edit needs a reread of the changed wording and its meaning-dependent context. If the meaning remains disputed or recovery depends on a reader’s response, the additional work follows §§4.5–4.6. Ordinary correction produces repaired text; only a named high-risk or disputed decision needs comparison evidence.
Rationale
Precise plain language has two obligations. The sentence must be semantically complete enough to recover its predicates, participants, referents, kinds, and operational detail. Every additional structure must also earn its place by changing understanding or use for the intended reader.
Either obligation alone is insufficient: a fully typed sentence can still be noise, and a short sentence can still hide its object.
Recover the governing message and relations, remove unsupported structure and displaced apparatus, then write the shortest ordinary sentence that preserves the live meaning. Use E.10 for cues and unresolved FPF wording. F.19 supplies the ordinary connected reading. Attention management remains outside the language pattern.
SoTA-Echoing
SoTA here means the best current contribution to the stated practice question, not the newest or most formal publication. A source's official status does not by itself make it SoTA.
Bounded choice for ordinary technical prose. Compare the audience-sensitive whole-span reading with cue-led sentence revision: find listed suspect expressions, improve their wording, and retain true ontological distinctions. In the denial case in section 5, a cue-led revision can leave “The evidence does not notice the error and does not begin a new cycle” when those verbs are outside the selector and local truth is taken as sufficient.
For this comparison, suppose the surrounding paragraph already tells the reader to check or revise the first distinction on receiving new evidence. With the same paragraph available, F.19 recovers that stated action, finds no independently grounded reader mistake that the denial prevents, and deletes the denial. The positive action stays; no new action is inferred from the deleted sentence.
Use one bounded reread of the same paragraph or short instruction as the comparison allowance. On the ordinary sentence “The diagram shows the dependency”, both approaches retain the wording; F.19 expressly preserves its recoverable metonymy. On the operational-detail case, applying either approach with meaning preservation rejects “Cook until ready”: it discards “five minutes after simmer begins”.
The additional F.19 move is to examine the contribution and participants of an unflagged proposition, rather than taking a clear and locally true sentence as sufficient. Adapt the audience-sensitive line in F.19:4 steps 1–7 and CC-F19-9/CC-F19-12/CC-F19-17; keep lexical cues for recall. The accepted trade-off is a contextual judgement about each claim, including unflagged claims, instead of a fully mechanical vocabulary check.
These are qualitative case comparisons, not measured timing or a controlled-language compliance test. Reopen the choice if that judgement repeatedly rejects useful prose or a lighter method catches the same defects while preserving the same uses.
Relations
F.19:End
Last Updated: 2026-11-22 — upstream FPF commit 7a93d8f8 (github.com/ailev/FPF)