FPF Reference

    FPF Pattern-Quality Evaluation CharacteristicSpace

    E.21 · Stable · Part E - The FPF Constitution and Authoring Guides

    About this pattern

    This is a generated FPF pattern page projected from the published FPF source. It is canonical FPF content for this ID; it is not a FPF Reference product feature page.

    How to use this pattern

    Read the ID, status, type, and normativity first. Use the content for exact wording, the relations for adjacent concepts, and citations to keep active work grounded without pasting the whole specification.

    Status: Core.

    Use E.21 when one authored FPF pattern of concern must be evaluated for quality under the use required by the governing evaluation frame: ordinary practitioner use, authoring input, landing input, release input, external-review input, high-assurance reuse input, canonization input, or another explicitly requested pattern-quality use. The evaluator does not replace the required ClaimScope with an easier one. If the pattern fails the required use, the result is repairBeforeUse, holdForArchitectureDecision, or refreshNeeded; a different use needs a different evaluation frame and does not rescue the current result.

    Relations

    E.21builds onFPF Authoring Conventions and Style Guide
    E.21builds onPattern Quality Gates: Review and Refresh Profiles
    E.21builds onQ-Bundle: Authoring "-ilities" as Structured Quality Bundles
    E.21builds onMM-CHR — Measurement & Metrics Characterization
    E.21builds onA.CHR-NORM — Canonical “Characteristic” & rename (Dimension/Axis → Characteristic)
    E.21builds onCharacteristicSpace & Dynamics Hook (A.CHR-SPACE)
    E.21builds onLocal-First Unification Naming Protocol
    E.21builds onEvaluation CharacteristicSpace Construction
    E.21coordinates withImprovement-Oriented Quality Evaluation Question Framing
    E.21coordinates withQuality Improvement Loop Method
    E.21coordinates withDRR Decision-Adequacy Evaluation CharacteristicSpace
    E.21coordinates withFPF Pillar-Adequacy Evaluation CharacteristicSpace
    E.21coordinates withLEX-BUNDLE: Unified Lexical Rules for FPF
    E.21coordinates withOntology-First Plain Technical Rewriting
    E.21coordinates withRelational Precision Restoration (RPR) — Kind-Explicit Qualified Relation Discipline
    E.21coordinates withEpistemic Precision Restoration
    E.21coordinates withFirst-Practical Entry and Pattern-Use Discoverability Discipline
    E.21coordinates withExpanded Entry Disambiguation Cases
    E.21explicit referenceDRR Decision-Adequacy Evaluation CharacteristicSpace
    E.21explicit referenceFPF Pillar-Adequacy Evaluation CharacteristicSpace
    E.21explicit referenceLEX-BUNDLE: Unified Lexical Rules for FPF
    E.21explicit referenceEvaluation CharacteristicSpace Construction
    E.21explicit referenceFirst-Practical Entry and Pattern-Use Discoverability Discipline
    E.21explicit referenceExpanded Entry Disambiguation Cases
    E.21explicit referenceWording-Use Ontological Precision Restoration Architecture
    E.21explicit referenceLocal-First Unification Naming Protocol
    E.21explicit referenceOntology-First Plain Technical Rewriting
    E.21explicit referencePattern Quality Gates: Review and Refresh Profiles
    E.21explicit referenceU.Signature — Universal, law‑governed declaration
    E.21explicit referenceU.RelationSlotDiscipline - SlotKind / ValueKind / RefKind discipline for n‑ary relations (with slot‑operation lexicon)
    E.21explicit referenceRelational Precision Restoration (RPR) — Kind-Explicit Qualified Relation Discipline
    E.21explicit referenceMathematical Lens Use
    E.21explicit referenceRole-Method-Work Alignment (Contextual Enactment)
    E.21explicit referenceImprovement-Oriented Quality Evaluation Question Framing
    E.21explicit referenceQuality Improvement Loop Method
    E.21explicit referenceMM-CHR — Measurement & Metrics Characterization
    E.21explicit referenceA.CHR-NORM — Canonical “Characteristic” & rename (Dimension/Axis → Characteristic)
    E.21explicit referenceA.CSLC-KERNEL — Minimal CSLC in Kernel (Characteristic/Scale/Level/Coordinate)
    E.21explicit referenceCharacteristicSpace & Dynamics Hook (A.CHR-SPACE)
    E.21explicit referenceFPF Authoring Conventions and Style Guide
    E.21explicit referenceEpistemic Precision Restoration
    E.21explicit referenceCharacteristic and Scale Precision Restoration
    E.21explicit referenceQuality-Term Precision Restoration
    E.21explicit referenceEvidence Graph Referring (C-4)
    E.21explicit referenceTrust & Assurance Calculus (F–G–R with Congruence)
    E.21explicit referenceU.Flow.ConstraintValidity — Eulerian
    E.21explicit referenceGateProfilization: OperationalGate(profile) (GateFit core)

    Content

    Problem frame

    Use E.21 when one authored FPF pattern of concern must be evaluated for quality under the use required by the governing evaluation frame: ordinary practitioner use, authoring input, landing input, release input, external-review input, high-assurance reuse input, canonization input, or another explicitly requested pattern-quality use. The evaluator does not replace the required ClaimScope with an easier one. If the pattern fails the required use, the result is repairBeforeUse, holdForArchitectureDecision, or refreshNeeded; a different use needs a different evaluation frame and does not rescue the current result.

    Not this pattern when the evaluated object is one DRR, an FPF-level corpus object, a single wording repair, a source-use decision, or a project-side evidence, assurance, gate, release, safety, compliance, work, or decision claim. Use E.9.DA, E.2.DA, E.10 and precision-restoration neighboring patterns named by value, or the project-side pattern governing the claim for those objects.

    First useful move: recover the required scope from the governing request, E.22 frame, campaign seam, landing check, release check, or review assignment; then name the governing pattern of concern, required scope, working reader, intended use, and qualification window; then evaluate every coordinate in RequiredPatternQualityCoordinates with a value and short rationale.

    floorEvaluation changes the declared floor and expected evidence economy. It never creates a partial E.21, inactive coordinates, overlay-trigger shortcuts, narrowing to an easier use, blocker-only substitution, or a permission to skip precision-restoration discharge. Fragmentary, wrong-shaped, or weak pattern text is still evaluated under the required scope; weakness receives low coordinate values, repair status, architecture hold, or refresh status.

    What goes wrong if missed: pattern quality becomes taste, checklist closure, source count, review state, landing state, or length. Short patterns can pass while missing mature content; long patterns can pass while hiding the first user move; semio material can take over a non-semio pattern.

    Primary EntityOfConcern in plain terms: the quality claim of one governing FPF pattern of concern for a declared use.

    Problem

    FPF patterns need a quality evaluation that is stronger than a style checklist and lighter than a project assurance audit. Earlier review habits produced two opposite failures:

    1. Too weak. A reviewer marks a pattern "ready" because no blocker is obvious, because it landed, or because headings exist.
    2. Too heavy. A reviewer adds more warnings, evidence cards, source rows, boundary notes, and process residues until the pattern becomes harder to use.

    E.21 solves this by measuring the pattern of concern against one complete coordinate set. The coordinates ask whether the pattern is usable, coherent, current, precise, affordable, mature enough for its claim, and safe from proxy improvement.

    Forces

    ForceTension
    Comparability vs false precisionPattern versions must be comparable, but ordinal qualities cannot be averaged.
    Completeness vs affordabilityEvery coordinate is evaluated; rationale and evidence can stay compact.
    Maturity vs lengthA short pattern is mature only when selected mature-pattern ingredients are present in the body or neighboring pattern governing the claims.
    Ontology vs usabilityNames and kinds must be exact without burying the first user move.
    Semio precision vs semio-biasEpisteme and publication distinctions matter, but non-semio patterns still lead with their own EntityOfConcern.
    Open-ended improvement vs stopImprovement can continue forever, while one version needs a scoped stop condition.

    Solution

    E.21 is the FPF pattern-quality specialization of A.19.ECS. It evaluates one pattern of concern under one declared quality claim.

    There is one evaluation shape:

    1. frame the object and use;
    2. apply the ordinal scale to every required coordinate;
    3. justify each value with ShortRationale;
    4. assign PatternQualityStatus;
    5. state stop, repair, architecture hold, or refresh condition;
    6. when improvement is requested, return proposal rows without changing the coordinate result into a work plan.

    There is no separate pre-check result. If a pattern lacks frame, first move, source basis, mature comparison, or naming clarity, the relevant coordinates fall.

    Local names and kind settlement

    Local nameKind and role
    PatternQualityEvaluationAuthored quality evaluation record over one pattern of concern.
    PatternOfConcernRefFPF pattern named by value that this E.21 evaluation makes the pattern of concern: host, monolith section, edition, or pinned version. PatternOfConcern is role-relative: the same pattern can also be the pattern of concern for another role in another flow, for example when a reader selects, applies, or reviews that pattern. This row names the concern of the quality-evaluation flow, not a special kind of pattern and not a second text. The evaluated pattern also has its own primary EntityOfConcern: the subject that its Problem/Solution/guidance is about. FPF patterns are applied to situations, claims, texts, or work objects. Use governing pattern only in the typed form governing pattern for <claim/relation/boundary> when the pattern actually governs that specific item; use related pattern for a looser pattern relation; use relation only for the relation itself.
    ClaimScopeQuality claim boundary recovered from the governing frame: ordinary use, authoring input, landing input, release input, external-review input, high-assurance reuse input, canonization input, or another explicitly requested pattern-quality use. It is not chosen by the evaluator to make a failing request pass.
    WorkingReaderScopeReader role and first-use situation the pattern must serve.
    IntendedUseAction that may use the result: continue drafting, admit for declared use, repair, refresh, or compare candidates.
    QualificationWindowEdition, SoTA, related-pattern, release, time, or comparison window in which the evaluation is current.
    EvaluationEvidenceBasisEvidence loci named by value for the evaluation: pattern body version, host or monolith section, README scenario, ToC row, E.11 entry-distribution locus, I.2 expanded entry-disambiguation case when corpus-facing, card or retrieval cue when claimed, source-currentness locus when SoTA/currentness is valued, mature comparator set when maturity is valued, and worked case or absence of worked case when case coverage is valued.
    QualityEvaluationQuestionFrameRefE.22 frame when purpose, floor, trade-offs, absorption, or proposal expectation needs to be declared.
    CoordinateValueRationalesOne row for every required coordinate: Coordinate, Value, ShortRationale.
    CoordinateEvidenceRefsPer-coordinate text, case, relation, SoTA, mature comparator, projection, or review refs where the short rationale depends on evidence outside the pattern body row being discussed.
    PrecisionRestorationProfileCompact profile over five precision-restoration layers: word/head/use precision, phrase-level apparatus, repeated or distributed material, role/carrier separation, and pattern-application ontology. It collapses those layers into one scalar effect for the E.21 result, not one coordinate per defect. The profile names present or bounded issues, checked absence scope when clean, affected coordinates, and the selected restoration or governing pattern such as E.10, E.10.ARCH, F.18, F.19, or an object-specific pattern.
    DominanceSetCoordinates used to compare already evaluated candidate versions. It never changes the required coordinate set.
    PatternQualityStatusScoped pattern-quality result assigned by E.21; it is not an E.19 admission or refresh decision by itself.
    StopConditionWhy improvement may stop, continue, refresh, or hold.
    Names are local to pattern-quality evaluation unless F.18 promotes a durable name. They are not project evidence, release state, review state, or assurance.

    Evaluation record

    PatternQualityEvaluation:
  PatternOfConcernRef: <governing pattern of concern>
  ClaimScope: <declared quality claim>
  WorkingReaderScope: <reader and first-use situation>
  IntendedUse: <what may consume the result>
  QualificationWindow: <edition/source/neighbour/release/comparison window>
  EvaluationEvidenceBasis: <checked pattern, corpus, source, comparator, case, and projection loci; missing or unchecked loci named explicitly when they affect values>
  PrecisionRestorationProfile: <collapsed profile: word/head/use, phrase-apparatus, repetition/distribution, role-carrier, pattern-application; scalar effect, affected coordinates, and selected restoration or governing pattern>
  CoordinateValueRationales: <all required coordinates, values, short rationales>
  PatternQualityStatus: <status>
  StopCondition: <local stop, first repair, hold, or refresh>

    Ordinal scale, result row, and adjacent-value rationale

    ValueLabelMeaning
    0absentThe characteristic is not expressed for the declared scope.
    1namedOnlyIt is named or implied but not usable as quality evidence.
    2partiallyExpressedForDeclaredUseIt is present but incomplete, fragile, or insufficient for the declared use.
    3sufficientlyExpressedForDeclaredUseIt is usable for the declared scope, with limits visible.
    4wellExpressedForDeclaredUseIt is clear, evidenced, and bounded for the declared scope.
    5exceptionallyExpressedForDeclaredUseIt is exceptional for the declared use across reinforcing loci and cases, without hidden cost or neighbour loss.

    Values are ordinal content evaluations. They are not U.Measures, averages, percentages, maturity-ladder steps, review votes, or landing status.

    The result-bearing coordinate row has exactly this shape:

    CoordinateValueShortRationale
    <E.21 coordinate><0..5><assigned-value basis; why the lower adjacent value would understate the evidence; why the higher adjacent value would overstate the evidence, or for 5 what evidence makes 4 too weak and what would lower/reopen>

    A two-column coordinate/value table, a narrative paragraph, a table whose comment lacks adjacent-value comparison, or a result whose value depends on unchecked external loci is not an E.21 result. It is only draft evaluation material until every coordinate has a ShortRationale row and the result names the EvaluationEvidenceBasis used for values that depend on source, comparator, corpus, projection, or worked-case evidence.

    A ShortRationale is allowed to be compact, but it is not allowed to be evidenceless. When the value depends on a source-currentness row, mature comparator, README scenario, ToC row, E.11 entry-distribution locus, I.2 expanded entry-disambiguation case, card, retrieval cue, monolith section, worked slice, near-miss, or anti-case, the rationale names that locus by value or says that the locus was missing or unchecked. "By value" means a recoverable section, row, case, checklist item, relation, source row, projection row, comparator id plus selected ingredient, or exact absent locus; a category list such as "entry, first move, boundaries, SoTA, checklist, relations" is not by-value discharge. Missing or unchecked evidence lowers the value for the coordinate that needs it; it does not create a separate "not evaluated" result.

    A 5 is not a reward for clear early wording, named neighbour relations, or a well-formed field set alone. It needs exceptional expression for the declared use: reinforcing loci, a worked or otherwise replayable slice where the coordinate demands one, and no hidden cost or neighbour loss. When the evaluator cannot say why 4 would understate the evidence, assign 4 or lower.

    When a coordinate's 5 meaning names a filled case, replayable slice, near-miss, anti-case, worked comparison, projection evidence, currentness basis, or exact-neighbour replay, absence of that evidence caps that coordinate at 4 even if the prose is otherwise strong. Do not hide the same absence only in CaseCountercaseAndTransferCoverage; lower every coordinate whose own 5 meaning needs that missing evidence. A 5 rationale names the reinforcing evidence loci that make 4 too weak.

    For MaturePatternParityAndSelectedContentSufficiency, the rationale names a mature-pattern comparison set and the selected mature ingredients being claimed. For non-epistemic patterns, include at least one mature non-epistemic comparator when one exists: work, method, role, system, control, architecture, selection, engineering-action, or another pattern whose primary EntityOfConcern is not an episteme or publication. Value 4 requires by-value discharge of selected ingredients in the body or neighboring pattern governing the claims; comparator IDs plus a generic "main ingredients are present" sentence are only value 3. The comparison is not a length target and not permission to copy semio apparatus.

    For a 4 or 5 on MaturePatternParityAndSelectedContentSufficiency, include a compact maturity-discharge payload in the rationale or CoordinateEvidenceRefs: comparator=<pattern id>; selectedIngredient=<ingredient name>; currentLocus=<section, row, case, checklist item, relation, or neighboring pattern governing the claim>; missingOrLowering=<absent or weak ingredient, if any>. A category list such as "frame, first move, neighbour relations, CC, SoTA, relations" without current loci is still value 3, even when the listed categories are plausible mature ingredients.

    Precision-restoration profile

    Before assigning the coordinate table, record one PrecisionRestorationProfile. This is not an optional scan and not a lexical grep result. It is a role-based attention discharge: the evaluator asks what work the sentence, table, section, or repeated content family is doing in the pattern of concern.

    Use this compact shape:

    PrecisionRestorationProfile:
  overallEffect: <clean | boundedLocal | lowersCoordinates | repairBeforeUse>
  wordHeadUsePrecision: <clean | E.10/E.10.ARCH/F.18/governing pattern needed | lowers coordinates>
  kindRestorationCheck: <pre-repair kind/relation/slot-or-use-position/admissible use -> proposed post-repair kind/relation/slot-or-use-position/admissible use; preserved | split | intentionally changed | blocker>
  phraseApparatus: <clean | F.19 needed | lowers coordinates>
  repetitionAndNegativeDistribution: <clean | bounded-local | lowers coordinates>
  roleCarrierSeparation: <clean | quality/development/projection carrier leakage | lowers coordinates>
  patternApplicationOntology: <clean | application relation unclear | lowers coordinates>
  checkedLoci: <sections/rows/cases/relations checked>
  affectedCoordinates: <coordinates lowered or protected>
  repairProposal: <repair, no-repair disposition with loci, or owning locus>

    This profile deliberately collapses several small subreadings into one scalar effect. The scalar is the strongest quality effect that any layer requires: clean, bounded local repair, coordinate lowering, or repair-before-use. The layers are diagnostic, not extra coordinates, checklists, or proposal quotas. A new precision-restoration symptom is classified into one of these layers or assigned to the selected restoration or governing pattern; it does not mint a new [E.21](/generated/patterns/E.21) coordinate. Details belong in the patterns that govern those objects: word/head/name problems apply [E.10](/generated/patterns/E.10), [E.10.ARCH](/generated/patterns/E.10.ARCH), or [F.18](/generated/patterns/F.18); phrase-level boilerplate and plain-technical rewriting apply [F.19](/generated/patterns/F.19); claim, relation, evidence, work, decision, assurance, publication, or pattern-application problems apply the pattern that governs that object. [E.21](/generated/patterns/E.21) consumes only the result: which coordinates fall, which stay protected, and what repair would make the quality claim true.

    The kindRestorationCheck is required whenever a precision-restoration finding or repair proposal changes wording. It records the meaning-bearing object, kind, relation, slot or use-position, admissible use, and scope before and after the proposed repair, then names the governing pattern when another pattern governs the affected kind, relation, claim, or position ([A.6.0](/generated/patterns/A.6.0)/[A.6.5](/generated/patterns/A.6.5), [A.6.P](/generated/patterns/A.6.P), [C.29](/generated/patterns/C.29), [A.15](/generated/patterns/A.15), [E.10.ARCH](/generated/patterns/E.10.ARCH), or another governing pattern). [E.21](/generated/patterns/E.21) does not restate slot discipline or mathematical-lens ontology; it only checks that the repair preserved or deliberately changed them by value. The check is a bounded complete preservation proof, not a blanket demand to formalize every sentence and not a license to do the least visible work. Complete means every field whose value can drift because of the changed wording receives one explicit disposition: not triggered/ordinary prose/no FPF-governed phrase changed with checked loci, preserved, split, intentionally changed by accepted decision, or blocker. A no-repair result is valid only as one of those dispositions with loci; "nothing to do" without that discharge is a missing repair. Expand the row only when a kind, relation, claim, slot/use-position, or admissible use can drift. A lexical replacement is not a repair when it only removes a trigger word, substitutes one umbrella for another, narrows a graph or method into a work sequence, widens a work occurrence into a method, turns a publication or evidence carrier into the object itself, or otherwise changes kind or slot/use-position without an accepted decision. If the kind or slot/use-position cannot be recovered, the profile is at least lowersCoordinates; if the proposed repair would change kind or slot/use-position and no accepted DRR or governing pattern justifies that change, the result is repairBeforeUse or holdForArchitectureDecision.

    When the profile is not clean, lower every affected coordinate named by the profile. Do not hide a present precision-restoration issue only in EntityOfConcernPrimacyAndSemioBiasResistance, and do not raise the result through related-pattern-boundary praise, projection evidence, or "correct but true" guards when the profile shows that those materials compete with the positive subject/action spine.

    RequiredPatternQualityCoordinates

    Every E.21 evaluation of an FPF pattern of concern evaluates every coordinate below.

    CoordinateWhat it evaluates
    WorkingSituationAndUseBoundaryRecognizabilityWhether the reader recognises the situation, ordinary use, non-use, harm if missed, and boundary early.
    EntityOfConcernAndClaimScopeStabilityWhether the primary EntityOfConcern and quality-claim scope stay stable across title, Problem frame, Solution, cases, checklist, relations, and status.
    ActionPathGuidanceWhether the Solution gives a usable action path after the first move is recovered.
    ClosureAndBoundedNonUseRecoverabilityWhether stop conditions, repair conditions, bounded non-use, and any governing pattern for <claim/relation/boundary> statements are recoverable.
    SemanticKindAndNameRecoverabilityWhether names, kinds, relations, qualifiers, and claim boundaries recover the same FPF interpretation.
    NeighborAuthorityAndBoundedUseFitWhether evidence, assurance, measurement, naming, work, gate, decision, publication, release, and project claims stay with the pattern that governs each claim, relation, or boundary.
    EntityOfConcernPrimacyAndSemioBiasResistanceWhether the pattern leads with its own EntityOfConcern and action move instead of letting description, publication, source, evidence, review talk, standard non-use warnings, precision-repair material, quality/projection evidence, package rationale, or cross-pattern reference apparatus take over. The PrecisionRestorationProfile supplies the collapsed diagnosis across word/head/use precision, phrase apparatus, repetition/distribution, role/carrier separation, and pattern-application ontology. This coordinate consumes that profile by lowering the value when those materials compete with the positive subject/action spine. Semio-bias is one special case when the displaced content concerns descriptions, sources, publications, notes, records, diagrams, or evidence-like artifacts.
    PracticalUseDeltaAndHarmPreventionWhether the pattern changes a real reader move, prevents a named misuse, reduces a named cost, or preserves a named boundary.
    UseAffordabilityAndApparatusProportionalityWhether ordinary first use stays affordable and heavier apparatus appears only when it buys admissible use.
    RepairLocalityAndChangeImpactPredictabilityWhether repairs have the smallest locus and predictable downstream impact.
    ProxyForValueSubstitutionResistanceWhether the evaluation asks what became worse when visible quality coordinates improved.
    ClaimJustificationTraceabilityCurrentnessAndReplayabilityWhether the claim is replayable from pinned text, scope, evidence, currentness basis, limitations, status, and stop reason.
    CaseCountercaseAndTransferCoverageWhether positive cases, near-misses, anti-cases, and transfer cases match the breadth claimed.
    MaturePatternParityAndSelectedContentSufficiencyWhether selected mature-pattern ingredients are present in the body or related patterns for this EntityOfConcern and use.
    SoTABindingAndCurrentnessWhether current best-known practice changes the pattern and has reopen/currentness discipline.
    FormalClaimLegalityAndLensFitWhether measurement, scale, comparison, formal model, simulation, causal, mathematical, QL, or learned-lens claims are legal and bounded, or correctly absent.
    FalsifiabilityAndLoweringConditionWhether coordinate values, status, and stop claims say what would raise, lower, or reopen the evaluation.
    CorpusEntryProjectionAndEcologyFitWhether README scenarios, ToC query cues, Preface cues, E.11 entry-distribution loci, I.2 expanded entry-disambiguation cases, cards, summaries, retrieval snippets, durable names, relations, and corpus ecology preserve the scoped quality result without becoming authority faces, stale echoes, or pattern content. Corpus-entry and projection evidence belongs in the E.21 result, E.19 run record, README/ToC/E.11/I.2, retrieval/card carrier, or other quality carrier unless the pattern of concern's own EntityOfConcern and user move are that projection or evaluation work.
    EvolutionFrontAndRefreshDisciplineWhether variants, fronts, archives, refresh windows, and smallest-reopen rules preserve open-ended evolution without endless polishing.

    Constraint, harm, safety, security, compliance, deontic, self-application, recursion, and high-assurance questions do not add a second coordinate family. Evaluate them through the coordinate that owns the content: related-pattern authority, traceability, formal legality, falsifiability, affordability, corpus ecology, or evolution/refresh.

    Coupled-flow unity/separation for pattern quality. An E.21 run evaluates a PatternOfConcernRef inside a development, refresh, or admission flow. Another flow may make the same pattern a pattern of concern for a different role, for example a practitioner selecting and using it, a reviewer applying it to another text, or a later evaluator reopening it. One TransductionGraph may join pattern development, pattern use, use-found evaluation, and repair or refresh flows through transfer, feedback, return, edition-change, or projection relations. Keep three roles distinct in each sentence: the pattern as concern of the current flow, the intended reader addressed by the pattern, and the pattern's own primary EntityOfConcern inside its Problem/Solution/guidance. E.21, E.19, handoffs, ledgers, README/ToC/E.11/I.2/retrieval checks, and landing evidence are checking operations or carriers in the development or evaluation flow. They can cause edits to the pattern, but they are not automatically user-facing content for the role addressed by the pattern. DesignRunTag stays on the subject-context, claim, work, trace, or artifact relation inside the TGA graph; it does not decide whether a pattern is current, obsolete, under development, or being used. Treat FPF pattern development as the local pilot case: quality-loop proof changes the pattern through edits, not by being copied into the pattern.

    Frequent 3/4/5 calibration points

    These rows calibrate common disagreements. They do not replace the coordinate definitions above.

    Coordinate family3 is typical when4 is typical when5 is typical when
    WorkingSituationAndUseBoundaryRecognizabilityThe use situation is recoverable but late, abstract, or missing harm/payoff/non-use detail.The situation, first move, harm, payoff, and non-use are early and clear.Early recognition is reinforced by a filled or replayable first-use slice showing that a cold practitioner can enter correctly.
    EntityOfConcernAndClaimScopeStabilityThe primary object is named but related record, evidence, lens, or project claims keep pulling the scope.The primary EntityOfConcern and claim scope stay stable, with bounded related-pattern material.Scope stability is reinforced across title, recognition text, Solution, worked or replayable case material, checklist, relations, and non-use without any local apparatus stealing attention.
    ActionPathGuidanceThe move is named but only partly executable, or the Solution mostly assigns governing loci instead of giving this pattern's own action.The first move and continuation are executable in this pattern's own subject terms; related-pattern statements are declarative, compact, and late.The action path is demonstrated by a filled worked slice or equivalent replayable evidence.
    ClosureAndBoundedNonUseRecoverabilityNon-use or related-pattern statements are present but not tied to stop, repair, or lowering conditions.Stop, repair, bounded non-use, and governing-pattern statements for specific claims, relations, or boundaries are recoverable for declared use.A worked stop, overturn, or non-use case shows how closure changes status or the next applicable pattern relation.
    NeighborAuthorityAndBoundedUseFitRelated patterns are named but some authority split remains generic, future-pattern-like, ambiguous, role-nicknamed, or too early in the Solution.Related patterns named by value and limited declarative relations are clear enough for declared use and do not replace the pattern's own content.Related-pattern authority is replayable across examples, relations, and overread cases, with pattern application and authority kept explicit.
    EntityOfConcernPrimacyAndSemioBiasResistanceThe pattern is about its object but one or more precision-restoration layers lead or leak into the pattern in a developer/reviewer/evaluator role.The pattern leads with its own object and action path; auxiliary material is compact, declarative, and late; role/carrier/locus/flow/status words are used only when they add a real kind, relation, evidence value, or user move; quality/projection evidence about the pattern stays outside the pattern.The primary object and action spine are first recoverable across recognition text, Solution, cases, and checks even when auxiliary material is present, and any precision-restoration or quality/projection material is in its proper carrier rather than in the pattern.
    PracticalUseDeltaAndHarmPreventionThe prevented harm is named but not demonstrated.The pattern changes a recoverable move and blocks named misuse for declared use.A worked or near-miss case shows the practical delta, cost of the missed pattern, and prevented harm.
    UseAffordabilityAndApparatusProportionalityThe first move exists but apparatus is heavy for ordinary readers.Ordinary first use is affordable and heavier apparatus opens only when useful.A minimal first-use example shows the thin path works before heavy apparatus.
    RepairLocalityAndChangeImpactPredictabilityRepair conditions or related-pattern relations are named but downstream impact is not shown.Repairs have local loci and predictable impact for declared use.A worked repair or downstream-impact slice shows the smallest locus and changed related-pattern relation.
    ProxyForValueSubstitutionResistanceProxy risks are named but "what got worse" is not applied.The pattern blocks visible proxy substitutions and asks what worsened.A proxy-failure case shows a visible improvement damaging intended value, and the pattern prevents that stop.
    ClaimJustificationTraceabilityCurrentnessAndReplayabilityFields or sources exist but replayability/currentness basis is incomplete.The claim can be replayed from pinned text, evidence, currentness basis, status, and stop reason.A filled evidence/currentness slice shows how the claim is replayed and when it reopens.
    CaseCountercaseAndTransferCoverageArchetypes are listed, but no filled worked case or near-miss exercises the claim.At least one filled worked case plus a near-miss or anti-case covers the declared use.Heterogeneous cases, countercases, and transfer slices cover the breadth claimed.
    MaturePatternParityAndSelectedContentSufficiencyMature comparators are named or implied, but selected mature ingredients are not discharged by value.Mature comparators are named and selected ingredients are discharged by value in the body or related patterns named by value.Mature parity is shown across reinforcing body sections, related patterns, omissions, cases, and lowering conditions without copying irrelevant apparatus.
    SoTABindingAndCurrentnessSources are relevant and not decorative, but currentness, source-use status, or reopen conditions are compact or incomplete.Load-bearing sources state adopt/adapt/reject, content mutation, currentness window, and reopen condition.The pattern compares current best-known practice against popular, official, or lineage alternatives and carries the resulting source decisions into solution, cases, boundaries, and refresh.
    FormalClaimLegalityAndLensFitFormal, scale, lens, or measurement terms are bounded but not exercised.Formal/lens/measurement claims are legal, bounded, and governed by related patterns governing the pattern makes those claims.A worked formal/lens/scale comparison shows what is preserved, lost, admissible, and not proved.
    FalsifiabilityAndLoweringConditionStop, waiver, or non-use fields exist, but lowering and reopen triggers for the main claims are mostly implicit.The pattern states explicit lowering/reopen triggers for its main claims; named fields alone do not reach 4 unless they say what evidence change lowers, overturns, rejects, or reopens the claim.Worked lowering or overturn cases show how values, status, or use change.
    CorpusEntryProjectionAndEcologyFitHost text is coherent, but README, ToC, E.11, I.2, card, retrieval, monolith, or projection evidence is absent for a corpus-facing claim, or that evidence is placed anywhere in the pattern as method, note, appendix, relation, rationale, or quality-status content about the pattern.Corpus-facing entry/projection loci are named and aligned enough for the declared use, and their evidence stays in the evaluation/result/projection carrier rather than entering the pattern.Retrieval, stale-projection, cold-reader, or projection-update evidence shows corpus ecology stays aligned after change without leaking into the pattern.
    EvolutionFrontAndRefreshDisciplineReopen is delegated to related patterns or implied by source-return.The smallest reopen locus, source/currentness trigger, or variant/front condition is explicit.Variant/front/archive or ongoing refresh discipline is replayable for the declared use.

    For EntityOfConcernPrimacyAndSemioBiasResistance, do not compensate a bad PrecisionRestorationProfile with NeighborAuthorityAndBoundedUseFit or CorpusEntryProjectionAndEcologyFit. This is a role-based evaluation, not a lexical search: ask what role the sentence plays. Material about developing, reviewing, projecting, landing, evaluating, or proving this pattern's quality belongs in the carrier that owns that work, not in the pattern. Related-pattern statements named by value can be true and still damage the pattern of concern when they appear before the pattern's own EntityOfConcern and action spine are recoverable. If the opening Problem frame or Solution starts with precision-restoration material before the pattern's own subject and move, this coordinate is at most 2; if a positive action exists but the reader must traverse that material across sections to find it, it is at most 3. Compact related-pattern statements belong in Relations or short late boundary rows and must preserve kind. Local boundary prose is admissible only when it states a documented local confusion and local stop condition not already carried by the owning pattern for that specific distinction or claim boundary. Also lower ActionPathGuidance, WorkingSituationAndUseBoundaryRecognizability, PracticalUseDeltaAndHarmPrevention, and UseAffordabilityAndApparatusProportionality when the profile shows that precision-restoration issues displace first-use content. If the declared use is Stable, landing-input, release-input, external-review-ready, or another corpus-facing use, the evaluation must use evidence for corpus entry and projection coordinates. A host-only body evaluation can still evaluate the pattern body, but it cannot silently turn missing README, ToC, E.11, I.2, card, retrieval, monolith, or projection evidence into a high CorpusEntryProjectionAndEcologyFit value.

    Status and stop condition

    StatusMeaning
    admissibleForDeclaredUseEvery coordinate meets the declared floor for the scoped use, and bounded non-use is stated.
    repairBeforeUseOne or more coordinate floors fail for the declared use.
    holdForArchitectureDecisionThe defect is not local prose; EntityOfConcern, neighbour authority, split, merge, or placement must be decided.
    refreshNeededA SoTA, neighbour, terminology, retrieval, telemetry, use-scope, or corpus change invalidates a previous evaluation.

    Default floor is 4 wellExpressedForDeclaredUse on every coordinate for ordinary practitioner use, authoring-input use, landing-input use, Stable, external-review-ready, release-input, canonization-input, stop-improving claims, and ordinary improvement-loop use. A diagnostic or exploratory request still measures every coordinate and reports values; it does not create an admissible-use shortcut. If the assignment asks for corpus-facing, landing-input, Stable, release, or external-review use, the evaluator measures that required use and returns repairBeforeUse, holdForArchitectureDecision, or refreshNeeded when the floor is missed.

    An all-5 result is a local exceptional result under the declared scope and qualification window. It is not a permanent end of development. E.23 can reopen improvement when use, source, comparison set, front, affordability, or payoff changes.

    Compact result form

    An E.21 result uses this result-bearing form:

    E.21 result:
  Pattern of concern: <PatternOfConcernRef>
  Declared scope/use/reader/window: <ClaimScope, IntendedUse, WorkingReaderScope, QualificationWindow>
  Evidence basis checked: <EvaluationEvidenceBasis>
  Status: <PatternQualityStatus>
    PrecisionRestorationProfileOverallEffectKindRestorationCheckLociAffectedCoordinatesRepairProposal
    <word/head/use / phrase-apparatus / repetition-distribution / role-carrier / pattern-application profile>`<cleanboundedLocallowersCoordinatesrepairBeforeUse>`<pre/post kind, relation, slot/use-position, and not-triggered/ordinary/preserved/split/changed/blocker disposition>
    CoordinateValueShortRationale
    <all RequiredPatternQualityCoordinates rows><0..5><assigned-value basis; why not lower; why not higher or what would lower/reopen>
    First repair or stop: <repair | hold | local stop>
Reopen if: <smallest changed locus or condition>

    Status is not assigned from a two-column table, a prose summary, a checklist count, an [E.19](/generated/patterns/E.19) pass/fail row, a table missing ShortRationale, a result missing the required PrecisionRestorationProfile, or a result missing the evidence basis needed for the values it claims. Such material can support a later evaluation, but it is not the [E.21](/generated/patterns/E.21) result. Conversely, an [E.21](/generated/patterns/E.21) status is a pattern-quality status, not a release crossing: [E.19](/generated/patterns/E.19) or the release named by value/admission process still checks gate-specific carry-through, projection, monolith, packaging, authority, and non-overread conditions.

    Finding and proposal rows

    E.21 finding:
  Pattern of concern: <PatternOfConcernRef>
  Coordinate or status affected: <coordinate | status | stop>
  Pattern locus: <section, row, example, relation, source row, projection>
  Value or status effect: <value/status/floor/stop impact>
  Correction direction: <what should change>
  Closure test: <what changed pattern text would show>

    When [E.22](/generated/patterns/E.22), [E.23](/generated/patterns/E.23), returned-finding absorption, or exceptionalImprovementEvaluation asks for improvements, add proposal rows for every below-target coordinate, status weakness, stop-condition weakness, or open question that can be improved within the declared scope. One proposal may cover several coordinates only when it names all affected coordinates and the shared repair.

    Worked slices

    Names named by value, no first move. A pattern has precise Tech names and current source rows but no first user move. WorkingSituation..., ActionPathGuidance, and PracticalUseDelta... fall; source currentness does not rescue ordinary use.

    Short architecture pattern. A compact pattern has a triage form but no worked slice and no mature-pattern comparison. It can be useful as local expert reference material, but MaturePatternParity... and CaseCountercase... stay below exceptional until selected mature content is present.

    Precision-restoration profile in a non-semio pattern. A pattern about architecture, work, system levels, method, P2W, or another non-semio EntityOfConcern tries to introduce the subject through a catalog of other claim kinds or objects that are outside its own subject. That catalog is unbounded because every EoC is outside infinitely many other EoCs. If copied boundary doctrine leads the Problem frame or Solution, EntityOfConcernPrimacyAndSemioBiasResistance falls to 2 or 3 even when every individual boundary is true. Repair by leading with this pattern's own EntityOfConcern and action spine, and replace copied boundary doctrine with one governing pattern id or one governing pattern for <claim/relation/boundary> statement unless a documented local confusion needs an local stop condition not already carried there. If the same doctrine is spread across Problem frame, Solution, anti-patterns, checklist, and Relations, classify the aggregate under the profile's repetition/distribution layer and repair the distribution, not just each local sentence.

    Reference apparatus before Solution content. A pattern's first Solution paragraph assigns other patterns or related-pattern mappings before it unfolds the ontology, method, norm, worked action, or other positive solution for the pattern of concern's own EntityOfConcern. Even if the related pattern id is correct, ActionPathGuidance, EntityOfConcernPrimacyAndSemioBiasResistance, PracticalUseDeltaAndHarmPrevention, and sometimes NeighborAuthorityAndBoundedUseFit fall. Repair by moving discoverability to README, ToC, E.11, I.2, or retrieval/projection carriers, moving compact pattern-id or governing pattern for <claim/relation/boundary> statements to Relations or a late boundary row, moving architecture-placement rationale to DRR or architecture documents, and rewriting the Solution to answer "what do I do with this pattern's EoC?" before any statement about another pattern.

    Overformalized precision. A pattern uses correct FPF kinds, slots, references, and governing-pattern pointers so densely that the working reader cannot recover the first useful move, practical delta, or generalizing insight without doing an internal audit. Precision is then present but not usable. Lower UseAffordabilityAndApparatusProportionality, WorkingSituationAndUseBoundaryRecognizability, and sometimes ActionPathGuidance. Repair by keeping the ontology named by value only where it carries a live FPF-governed claim, moving restoration evidence to the evaluation or DRR carrier, and adding a short worked slice or plain recognition sentence that preserves the same kind without extra apparatus.

    QualityCarrierLeakage in the pattern. The pattern says that corpus projection, README/ToC/E.11/I.2 alignment, retrieval/cold-reader evidence, monolith parity, external-review readiness, landing evidence, PatternQualityStatus, all-4/all-5 posture, or another quality-carrier result is what the user should do with the pattern's EntityOfConcern, or records developer/reviewer/executor correspondence as if it were pattern content. The defect is not limited to Problem frame, Solution, examples, or checklist; notes, appendices, Relations, Rationale, SoTA-Echoing, tables, and conformance rows are also parts of the pattern in hosts and the monolith. That evidence may be required for E.21, E.19, landing, or retrieval carriers, but it is not automatically a user action in the pattern of concern. Lower EntityOfConcernPrimacyAndSemioBiasResistance, ActionPathGuidance, UseAffordabilityAndApparatusProportionality, and CorpusEntryProjectionAndEcologyFit when this evidence enters the pattern. Repair by moving the evidence to the E.21 result, E.19 run record, README/ToC/E.11/I.2, card/retrieval/projection carrier, or release/landing evidence carrier, and keeping in the pattern only the user-facing move or boundary that follows from that evidence.

    Quality table without rationale. A result gives values but no adjacent-value rationale. Values are unsupported. Add ShortRationale or lower.

    Goodharted improvement. A rewrite improves source refs and proof sketches but becomes hard to use. Re-evaluate affordability, repair locality, proxy-for-value, and corpus ecology before stopping.

    Conformance checklist

    CheckRequirement
    CC-E21-1Recover ClaimScope from the governing request, E.22 frame, campaign seam, landing check, release check, or review assignment; then name PatternOfConcernRef, ClaimScope, WorkingReaderScope, IntendedUse, QualificationWindow, and EvaluationEvidenceBasis.
    CC-E21-2Evaluate the full RequiredPatternQualityCoordinates set.
    CC-E21-2aBefore assigning coordinate values, record one PrecisionRestorationProfile with word/head/use, phrase-apparatus, repetition/distribution, role-carrier, and pattern-application layers. A missing, grouped, or memory-only profile makes the E.21 result incomplete.
    CC-E21-3Use the result-bearing three-column table: coordinate, value, and ShortRationale; a two-column coordinate/value table is not an E.21 result.
    CC-E21-4Let floorEvaluation change floor and evidence cost only, not the coordinate set.
    CC-E21-5Assign values from checked pattern content and named content evidence, not review, landing, popularity, praise, or absence of prior use.
    CC-E21-6For corpus-facing values, name the checked README, ToC, E.11, I.2, card, retrieval, monolith, or projection loci, or lower the affected coordinate when those loci are missing or unchecked.
    CC-E21-6aKeep corpus-projection, README/ToC/E.11/I.2 alignment, retrieval/cold-reader, monolith-parity, PatternQualityStatus, developer/reviewer/executor correspondence, and other quality-carrier evidence out of the pattern unless the pattern's own EntityOfConcern and user move are that evaluation/projection work. This is a role test, not a word-list test. If such material appears anywhere in the pattern, including notes, appendices, Relations, Rationale, SoTA-Echoing, examples, tables, conformance rows, or any other host/monolith pattern section, as development, review, projection, or quality-status content about the pattern, lower CorpusEntryProjectionAndEcologyFit, EntityOfConcernPrimacyAndSemioBiasResistance, and the affected action/usability coordinates.
    CC-E21-7For any 5, name the reinforcing evidence loci required by that coordinate's 5 meaning; otherwise lower the coordinate to 4 or below.
    CC-E21-8For MaturePatternParityAndSelectedContentSufficiency = 4 or 5, include a compact maturity-discharge payload: comparator id, selected ingredient, current locus, and missing/lowering item if any; category lists without loci cap the coordinate at 3.
    CC-E21-9Make SoTA rows adopt, adapt, or reject current practice and change the pattern.
    CC-E21-10Keep measurement, score, scale, formal, causal, mathematical, QL, simulation, representation, or learned-lens claims under C.16, A.17, A.18, A.19, or the pattern that governs the claim when the evaluated pattern makes those claims.
    CC-E21-11State floor satisfaction, remaining bounded non-use, and lowering or reopen conditions in any stop claim.
    CC-E21-12Keep coordinate rationale separate from improvement proposal rows.
    CC-E21-13Keep quality results out of project evidence, assurance, gate, work, safety/compliance, release, and publication truth claims.
    CC-E21-14Do not raise a pattern with a bad PrecisionRestorationProfile through related-pattern-boundary, projection, or quality-carrier praise. When the profile shows defects before the pattern of concern's primary subject action is recoverable, or enough volume to compete with the Solution, lower EntityOfConcernPrimacyAndSemioBiasResistance and the affected action/usability coordinates; do not offset that loss with generic related-pattern-boundary praise or correct corpus projection evidence.

    Common anti-patterns and repairs

    Anti-patternRepair
    Score illusion. Pattern quality = 87/100.Use ordinal coordinate values; no arithmetic aggregation.
    Two-column table. Coordinate/value table has no rationale.Add ShortRationale for every coordinate.
    Floor as omission. A floor evaluation omits maturity, SoTA, formal, corpus, or evolution coordinates.Keep floor low if needed; evaluate all coordinates.
    Scope laundering. A landing-input, corpus-facing, Stable, release, or external-review request is reported under an easier use, local-only use, diagnostic pass, or evaluator-selected use.Re-evaluate under the governing scope; if it fails, return repairBeforeUse, holdForArchitectureDecision, or refreshNeeded with the missed coordinates and repairs.
    Administrative proxy. "4 because landed" or "3 because not externally reviewed".Evaluate pattern content.
    Comparator-free or locus-free maturity. MaturePatternParity... = 4 by impression, comparator IDs only, or category list such as "frame, first move, checklist, SoTA, relations".Name mature comparison patterns and use the maturity-discharge payload: comparator, selected ingredient, current locus, and missing/lowering item. Without that payload, cap at 3.
    Omission account as maturity. A note explaining absence raises the value.Add content to body/neighboring pattern governing the claim, lower value, or mark the current request repairBeforeUse.
    Semio-biased maturity. Non-semio pattern is judged by episteme/publication exemplars only.Include non-epistemic mature comparators and score action on the primary EntityOfConcern.
    Quality-carrier leakage. Corpus projection, retrieval evidence, README/ToC/E.11/I.2 alignment, monolith parity, PatternQualityStatus, developer/reviewer/executor correspondence, or other quality evidence is written anywhere in the pattern as method, problem, note, appendix, relation, rationale, or status content about the pattern.Move the evidence to the E.21 result, E.19 run record, README/ToC/E.11/I.2, card/retrieval/projection carrier, or release/landing evidence carrier; keep only the user move or boundary that the evidence justifies.
    Apparatus overwrap. A simple FPF claim is wrapped in extra role, carrier, locus, flow, state, status, text-state, package, or process words, such as live pattern text, current object, active record, field when live, or route-like pattern talk where no real state/use-position is named, so the reader sees a bureaucratic apparatus instead of the object, relation, action, or boundary.Apply F.19; record the scalar effect in PrecisionRestorationProfile, then lower the affected coordinates or name the completed repair.
    Apparatus maximalism. Every pattern gets evidence cards, telemetry, archives, and companions.Keep evidence compact unless it changes value, status, stop, or candidate comparison.
    Quality veto theatre. "Not ready" has no E.21 coordinate named by value, evidence, status effect, and repair.
    Rewrite as an E.21 finding or remove the veto.

    Consequences

    BenefitTrade-off or mitigation
    Pattern quality becomes inspectable without a fake score.Authors must name scope and all coordinate values.
    Compact evidence remains possible.The coordinate table is still complete.
    Maturity claims become harder to fake.Mature-pattern comparison adds cost where maturity or corpus-facing use is claimed.
    Semio-bias becomes visible.Semio distinctions remain auxiliary unless they are the pattern's own EntityOfConcern.
    Stop decisions become less taste-based.Open-ended improvement remains possible through E.23 when a stronger aim is requested.

    Rationale

    E.21 keeps the measuring device simple: one object kind, one ordinal scale, one required coordinate set, one status set, and one stop condition. The evaluation never asks whether a coordinate is active. It asks what value the current pattern and its named evidence basis earn under the declared use.

    The mature-pattern parity coordinate is deliberately strict because recent short patterns looked formally clean while lacking the worked slices, source carry-through, lowering conditions, and transfer coverage present in mature FPF patterns. The repair is not "make everything long"; it is "carry the selected mature ingredients that the declared use needs."

    SoTA-Echoing

    ClaimSource-use dispositionConcrete E.21 effect
    Feedback connects desired state, current state, and next action.Adopt from feedback-for-learning lineage such as Sadler and Hattie/Timperley.ShortRationale and proposal rows are separated: value now, next improvement when requested.
    Questions and metrics derive from the goal.Adopt from GQM-style measurement discipline.Scope, reader, use, and window precede coordinate values.
    Multi-criteria improvement needs explicit trade-offs.Adopt from MCDA, Pareto, ATAM, and current QD/OEE lines.Dominance comparisons and protected trade-offs replace one-score closure.
    Proxy optimization can make intended value worse.Adopt from Goodhart/proxy-risk lines.ProxyForValueSubstitutionResistance and stop condition ask what got worse.
    Evaluation results are not governance, safety, or compliance proof.Adopt as non-overread boundary from current evaluation-governance practice.Neighbour authority and status boundaries keep project claims outside E.21.

    Relations

    NeighbourRelation
    A.19.ECSConstructs or repairs the general evaluation CharacteristicSpace; E.21 is one specialization.
    E.8.ECSPFPublishes an evaluation CharacteristicSpace as an FPF pattern when that form is selected.
    E.8Authors the pattern body whose quality E.21 evaluates.
    E.19Runs admission and refresh review profiles; it can consume or request E.21, but it does not assign E.21 coordinate values or replace the required pattern-quality table.
    E.22Frames purpose, floor, trade-offs, and proposal expectation before an evaluation.
    E.23Runs repeated improvement using E.21 values and stop meanings for pattern versions.
    E.9.DAEvaluates upstream DRR decision adequacy when pattern-quality defects trace to decisions.
    C.16, A.17, A.18, A.19Govern scale, coordinate, and measurement legality.
    F.18, E.10, A.6.P, C.2.P, C.16.P, C.16.QGovern naming and wording-use precision when quality defects are lexical or ontological.
    A.10, B.3, A.20, A.21, A.15Govern project evidence, assurance, local CV state, gates, and work authority.
    E.11 and I.2Govern entry-distribution and expanded entry-disambiguation cues; E.21 supplies only the scoped quality result.

    E.21:End

    Last Updated: 2026-06-05 — this section last modified in upstream FPF commit eaafd3a4 (github.com/ailev/FPF)