#Epistemic Precision Restoration

C.2.P · Cluster C.I - Core CALs / LOGs / CHRs · Stable · Architectural (A), C.2 precision-restoration pattern · Normative unless a section is explicitly informative · Part C - Kernel Extension Specifications

#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.

Type: Architectural (A), C.2 precision-restoration pattern Status: Stable Normativity: Normative unless a section is explicitly informative

Source-expression unpacking mode. Use this mode when someone asks to make ordinary language more precise, including intake notes, seminar transcripts, external reviews, project documents, source publications, tool outputs, or other non-FPF prose. The aim is source-local clarification: recover what the sentence might mean, identify candidate kinds and relations, preserve important wording when needed, and produce one clarified phrase, candidate-set note, epistemic precision-restoration note, or transfer disposition. This mode may apply E.10, A.6.P, A.6.6, F.18, A.7, E.17, or another exact FPF pattern as a repair method, but it does not require the source text itself to become FPF-conformant.

#Keywords

• epistemic precision restoration
• source wording
• source-expression unpacking
• source-to-FPF transfer
• claim-bearing episteme
• publication face
• carrier
• PublicationUnit
• described entity
• grounding relation
• transfer disposition
• remaining admissible reader move.

#Relations

#Content

#Use this when

Source-expression unpacking mode. Use this mode when someone asks to make ordinary language more precise, including intake notes, seminar transcripts, external reviews, project documents, source publications, tool outputs, or other non-FPF prose. The aim is source-local clarification: recover what the sentence might mean, identify candidate kinds and relations, preserve important wording when needed, and produce one clarified phrase, candidate-set note, epistemic precision-restoration note, or transfer disposition. This mode may apply E.10, A.6.P, A.6.6, F.18, A.7, E.17, or another exact FPF pattern as a repair method, but it does not require the source text itself to become FPF-conformant.

FPF-transfer or conformance mode. Use this mode when episteme-publication-heavy conformant text or text being promoted into FPF current content relies on loose wording around epistemes, publications, views, publication forms, generic publication faces, governed MVPK faces, bounded publication units, carriers, records, relations, admissible uses, or pattern application. Here the output must be FPF-conformant: exact recovered FPF kind, relation record, relation phrase, tuple-like record, exact project-side FPF kind and reference, or explicit non-transfer disposition.

Use C.2.P as epistemic precision restoration for wording whose live object is an episteme/publication/source-transfer construction: source wording, source-local meaning, claim-bearing episteme, publication, view, face, carrier, publication unit, described entity, grounding relation, pattern-application wording, project-side reliance wording, or the disposition by which source expression may or may not become FPF-current wording.

E.10 governs lexical conformance of a wording use. C.2.P governs epistemic precision restoration across the epistemic and publication stack: expression, source-local meaning, recovered FPF kind stack, publication/carrier/view construction, described-entity or grounding relation, admissible reader move, and transfer or non-transfer disposition.

The practical partition is episteme/publication-slot-like, but it is not limited to named C.2.1 slots. It also includes publication constructions, carrier and face constructions, source-expression-to-current-FPF transfer, and pattern-application wording when those are used as claim-bearing or admissibility-bearing signs. Apply this pattern from E.10 only when the exact governing pattern cannot yet be selected directly because source wording, claim-bearing episteme, publication/carrier construction, project-side reliance, pattern-application wording, or transfer/non-transfer disposition is still unresolved. The pattern-local decision is not a procedure path: it selects source-expression unpacking mode or FPF-transfer/conformance mode, recovers the live episteme-publication stack, chooses recovered-by-value, quote-only, reduced-use-cue, extension-candidate, blocked-transfer, rewrite-incomplete, or not-triggered disposition, and preserves the remaining admissible reader move before any neighboring pattern governs its own invariant.

Precision-restoration pattern note. A precision-restoration pattern is an architectural pattern for a recurring complex precision problem whose wording routinely hides several live distinctions. A.6.P is relation precision restoration; C.2.P is epistemic precision restoration. C.30.P is the selected architecture/structure precision-restoration pattern when architecture or structure wording hides the live object. C.16.P is the selected characteristic/scale precision-restoration pattern when characteristic, scale, metric, score, indicator, coordinate, threshold, or comparison construction is hidden. C.16.Q is the selected quality-term precision-restoration pattern when quality/evaluative characterization is live and the found problem is not relation construction. E.10 detects the wording problem and selects the applicable path; E.10.ARCH carries the shared recovery algorithm and applicability-row architecture; neither replaces this pattern's episteme/publication/source-transfer ontology.

Ordinary-language survival. Ordinary words remain admissible until the sentence gives them FPF-kind, relation, authority, evidence, admissibility, work, gate, decision, bridge, or reliance force. Source may stay ordinary when it only means where a quote came from; view may stay ordinary when it means what the reader sees and not U.View; route may stay ordinary navigation prose; support may stay ordinary help. Repair by FPF-force-bearing sentence function, not by trigger word alone.

Not this pattern when. C.2.P is not the governing pattern for every recovered construct. General FPF lexical conformance stays under E.10; stable reusable naming under F.18; relation precision under A.6.P; A.6.B law-, admissibility-, deontic-, and effect-claim boundary splitting under A.6.B; object-description-carrier separation under A.7; view and publication discipline under E.17 and E.17.0; architecture and structural description adequacy under C.30, C.30.ASV, A.22, C.31, or the exact architecture/structure pattern; project work, evidence, gate, decision, method, action-invitation, assurance, and engineering-justification claims under their exact FPF patterns. When one of those claims is live, this pattern supplies source-expression unpacking and rewrite disposition; the exact named FPF pattern supplies its invariant.

Do not punish clarity. Prefer the clearest ordinary head that preserves kind, relation, and admissible use. Do not replace a clear plain phrase with a technical phrase unless the technical phrase blocks a live false interpretation or is needed for accepted stable FPF naming. In an ordinary case, reader help, source-pointer-only, or comparison only may be better than a more technical phrase.

#What goes wrong if missed

Episteme-publication-heavy text starts to build a parallel ontology. A generic publication face becomes a U.View, a file becomes an episteme, a dashboard tile becomes evidence, a pattern name becomes a procedure, a slash list becomes a group kind, or a broad word such as source hides whether the text means a pattern, a DRR, a publication, a document with named source-basis, evidence-basis, architecture-basis, or review-basis role, an exact project-side FPF kind and reference, or a relation.

The immediate cost is not only ugly terminology. Engineers and FPF authors start making action, evidence, gate, decision, or engineering-justification claims from the wrong object.

#What this buys

C.2.P gives authors and reviewers one small epistemic precision-restoration action: recover the FPF kind stack first, then write exact wording that preserves the needed distinction without adding another claim. It prevents string-replacement cleanup, keeps FPF-side and project-side episteme and publication work separate, and blocks unclear text from becoming current FPF content by author guesswork.

Successful repair condition. Epistemic precision restoration is not closed by type-correct wording alone. It is governed by E.2 Pillars, especially P-2 Didactic Primacy, together with E.12 and the register rule in E.10:6.2. It closes only when the repaired text preserves or restores one remaining admissible reader move: a usable action, a recognition reason that tells the working reader why the distinction matters, or a named exact FPF pattern application that carries the live claim. When both Tech and Plain registers are live, the Tech reading must remain recoverable and any Plain or didactic line must map back to that Tech reading. A Plain, more expressive line, or intentional didactic metaphor may stay ordinary when it carries no FPF force; when it carries ontological, evidence, causal, assurance, bridge, gate, work, decision, or admissibility force, that claim force must be recoverable through the Tech fields, exact FPF kind, recovered relation, project-side source reference, or disposition named by the repair. If a repair in a FPF-force-bearing Problem frame, Problem section, recognition text, example, or worked slice makes the text more exact but less able to show the working situation, why it matters, or what action remains, the repair is incomplete unless the exact governing FPF pattern is named for that live claim. Overread removal is only half of epistemic precision restoration; the other half is surviving admissible action under the Pillars.

Governed object in plain terms. The governed object is one episteme-publication-heavy wording use inside conformant text: the word or phrase, the sentence function it carries, the FPF kind or relation it must recover, and the admissible remaining use after recovery.

Primary working reader. The first reader is an author or reviewer of conformant FPF-style text who must repair wording without losing ontology. The downstream reader is the engineer-manager using the resulting pattern or project text in a working situation.

Anti-overread payoff question. A repair is useful only if the pattern text can answer three things in ordinary prose: what false downstream interpretation is blocked; what useful admissible action remains; and when the reader must apply the exact FPF governing pattern because evidence, gate, decision, work, assurance, bridge, release, or reliance is live. If the repair blocks an overclaim but leaves no useful action, the text is probably becoming ceremony rather than guidance.

#Problem frame

FPF already has episteme, publication, view, carrier, presentation, relation, naming, and pattern-application concepts. FPF-facing source, review, and draft prose can still introduce convenient intermediate words that survive into final pattern or architecture prose without their kind stack recovered.

The recurring situation is simple: a sentence is understandable enough to feel worth keeping, but its head kind is not recovered. If it is repaired by replacing one broad word with another broad word, the ontology gets worse while the text looks cleaner.

#Purpose and Scope

This pattern gives the current glossary and rewrite rules for terms around epistemes, publications, views, publication forms, generic publication faces, governed MVPK faces, carriers, records, and bounded publication units.

It exists because episteme-publication-heavy texts can use locally convenient heads that collapse described entity, publication unit, publication face, carrier, record, source relation, and project-side value. Those words may be useful recognition handles, but they are not safe FPF heads when they carry ontology, authority, or authority-changing meaning.

The rewrite discipline here is semantic, not lexical; in this pattern the semantic work is bounded to episteme-publication/source-transfer precision:

• do not replace one broad token with one new broad token by string substitution;
• first recover the FPF kind stack, the claim-bearing status, the publication, view, carrier, or relation construction, and any work, action, or authority crossing;
• then choose the smallest exact wording that preserves the FPF-force-bearing distinction without creating a second ontology.

This pattern uses the E.10 trigger result as its entry condition, then works in the C.2.1/E.17 epistemic-publication ontology rather than in a lexical registry.

#Problem

Without an epistemic precision-restoration discipline for episteme-publication-heavy wording:

1. broad publication words hide whether the claim is about U.Episteme, U.View, publication form, generic publication face, governed MVPK face, PublicationUnit, carrier, document with named source-basis, evidence-basis, architecture-basis, or review-basis role, review object, or exact project-side FPF kind and reference;
2. FPF pattern-application claims and project-side work-occurrence, work-plan, decision, action-invitation, method, record, carrier, or front-end claims get mixed in one sentence;
3. slash lists and heterogeneous rows become false group kinds;
4. unclear source meaning is guessed into FPF rather than blocked or promoted through an accepted FPF extension;
5. authors copy the same loose wording into DRRs, patterns, source-basis notes, or project texts.

#Forces

#Solution

Repair episteme-publication-heavy wording by epistemic precision restoration, not by dictionary replacement.

A successful rewrite satisfies these field-validity constraints:

1. the head kind and sentence function are recoverable under E.10;
2. a stable reusable name has F.18 status;
3. a relation, comparison, dependency, support, sameness, grounding, mapping, or endpoint claim has A.6.P relation precision, with admissibility and project-side support questions split into their own fields;
4. a claim-bearing episteme, exact episteme species, episteme-lane view, or exact project-side FPF kind and reference has the needed C.2.1 typing or named exact FPF claim force;
5. publication, view, face, and carrier distinctions satisfy E.17.0, E.17, and MVPK;
6. the repaired text satisfies E.2 Pillars, especially P-2 Didactic Primacy, by preserving or restoring one remaining admissible reader move: a usable action, a recognition reason that tells the working reader why the distinction matters, or a named exact FPF pattern application that carries the live claim; when both Tech and Plain registers are live, the Plain or didactic line maps back to the recovered Tech kind, relation, or exact FPF pattern application under E.10:6.2; ordinary Plain wording and intentional didactic metaphor stay light when they carry no FPF force, but ontological, evidence, causal, assurance, bridge, gate, work, decision, or admissibility force in a more expressive Plain line must be recoverable through the repaired Tech fields; FPF-force-bearing Problem frames, Problem sections, recognition texts, examples, and worked slices must still show the broad working situation and first useful move, or the rewrite is incomplete;
7. the final phrase preserves the distinction without adding another claim;
8. unrecoverable meaning, kind, register mapping, or remaining reader move fails closed.

The detailed solution below carries the glossary and rewrite rules as ordinary pattern subsections. It is not an external container: these subsections are the pattern's detailed epistemic precision-restoration guidance.

#EpistemicPrecisionRestorationRecord

For FPF-force-bearing cases, the recovery product is a compact pattern-local EpistemicPrecisionRestorationRecord or an equivalent local rewrite note. Ordinary local phrase repair may end as the repaired sentence itself when kind, relation, and admissible use are now clear and no downstream reliance, cross-context reuse, grouped-kind risk, hidden authority claim, project-side overclaim, conflict among publication, describedEntity, and project-side action claims, or contested source meaning remains live. Prefer the plain names epistemic precision-restoration note, compact epistemic precision-restoration row, or local rewrite note when durable inspection does not require the code-like field name. The recovery note is a lightweight pattern-local author or reviewer product, not a new ontology, not a dispatch table, not a durable FPF record kind, and not a mandatory heavyweight project record. It becomes a durable FPF record only if another accepted pattern or accepted DRR explicitly admits it as one. It records only the trigger, the recovered FPF kind stack, the requirement from the exact governing FPF pattern, and the final rewrite disposition that must remain inspectable after the repair.

Minimum fields when FPF-force-bearing:

Recover by claim force, not word form. For words such as source, support, status, valid, ready, approved, and used, first ask what the sentence would let the reader do or rely on: source-finding only, source availability, source use, evidence support, gate passage, decision status, readiness threshold, work permission, assurance, engineering justification, or ordinary orientation. Then fill only the field whose exact FPF kind, relation, or project-side reference is live.

Use the short form when only one field is live. Use the full record when several fields are live or when the phrase might otherwise create a grouped kind, hidden authority claim, project-side overclaim, conflict among publication, describedEntity, and project-side action claims, contested source-meaning transfer, or procedure-like ordering of pattern applications.

#General Recovery Check

Use this recovery check whenever text proposes a new term, repairs an episteme-publication-heavy term, asks for language precision, or relies on wording around PublicationUnit, describedEntity, publication, view, face, carrier, source-side relation, receiving-side relation, publication face, described entity, or bounded publication-unit status.

0. Mode selection. Decide whether the current use is source-expression unpacking mode over non-FPF prose or FPF-transfer/conformance mode. In source-expression unpacking mode, preserve source-local nuance and do not force the whole source into FPF conformance. In FPF-transfer/conformance mode, the transferred wording must satisfy E.10 and the exact governing patterns named by the recovery.

1. E.10 trigger scan and head-kind recovery. Use E.10:0.2 as the shared trigger scan. Decide what the head noun names before accepting the phrase: intension, description episteme or specification episteme, U.Episteme, U.View, publication form, generic publication face, governed MVPK face, carrier or rendering, exact project-side FPF kind and reference, A.15.1 dated U.Work occurrence, A.6.A action invitation, A.2.9 SpeechActRef, A.2.8 U.Commitment, U.Method, U.MethodDescription, document with named source-basis, evidence-basis, architecture-basis, or review-basis role, review object, or source-local ordinary sense. Apply Intension, Description, and Specification, Context, Tech and Plain, and carrier humility rules before treating a word as meaning-bearing.

2. F.18 naming pass when a stable term is being chosen. If the phrase is becoming a reusable head, fill at least the lightweight Name Card facts: Context, Kind, purpose and use-domain, local sense, candidate head families, NQD-front reasoning, sense-seed read-through, and the lexical Q tuple {SemanticFidelity, CognitiveErgonomics, MorphologicalActionFit, AliasRisk}. Do not pick a label only because it is intuitive. Do not accept a replacement label until it passes the E.10:0.2 replacement-candidate anti-umbrella rule.

3. A.6.P relation-precision pass when a phrase carries relation, comparison, or action-invitation force. Restore generic head kind first, then endpoint facets and kinds, then relation kind, slots, qualifiers, scope, time, viewpoint, and hooks for admissibility, evidence, and work. For support wording, do not stop at a substitute label: select the live support-like claim force under A.6.P first, including source-description relation, described-entity or grounding-holon grounding, base/anchor relation through A.6.6, evidence, assurance, causal-use, mathematical-lens, characteristic/measurement, admissible-use, work/enablement, or publication-companion use. If ambiguity remains, write a local Candidate-Set Note rather than debating synonyms.

4. C.2.1 episteme-slot pass when the object is claim-bearing. Name describedEntity, grounding, ClaimGraph, viewpoint and view, reference scheme, representation scheme, and bounded context as far as the claim needs. Do not use PublicationUnit or a carrier word as a substitute episteme.

5. E.17.0, E.17, MVPK publication pass when the object is published or reader-facing. Separate the underlying episteme or view, U.EpistemePublication, publication form, generic publication face, governed MVPK face, PublicationUnit, carrier or rendering, and the exact project-side FPF kind and reference when a project-side claim is live. A face, card, screen, or explanation can guide interpretation or source-finding without becoming evidence, work, gate passage, authority, or release permission. If those claims are live, fill admissibleUse and projectSideFPFRef instead of treating the generic publication face or governed MVPK face as the source value.

5a. Precision-restoration exit after source/current recovery. If source/current, publication, carrier, face, or PublicationUnit recovery exposes architecture/structure wording, characteristic/scale wording, quality/evaluative characterization, function-like carrier wording, relation construction, controlled coarsening, naming, evidence, assurance, gate, work, decision, causal-use, release, mathematical-lens, or another exact neighboring claim, fill precisionRestorationExit with the exact receiving pattern. Do not keep the neighboring claim inside C.2.P after this pattern has recovered the source/current and publication stack.

6. Remaining admissible reader move. After the kind, relation, publication, and project-side splits are recovered, state the remaining admissible reader move in one short line: what the working reader can now do, why the distinction still matters, or which exact named FPF pattern governs the live claim. If both Tech and Plain registers are live, keep the Tech reading recoverable and make the Plain or didactic line map back to the recovered Tech kind, relation, or named exact FPF pattern application under E.10:6.2. Do not make this a heavy form for ordinary prose: a Plain line that carries no FPF force may stay ordinary; a Plain line that carries ontological, evidence, causal, assurance, bridge, gate, work, decision, or admissibility force must be recoverable through the repaired Tech fields. If the repaired wording only proves that an overclaim was removed, but leaves no usable action, recognition reason, or exact FPF pattern application for the live claim, do not classify the repair as recovered by value.

7. Authority-changing rewrite boundary. If the result would rename an accepted FPF pattern, change an accepted FPF term, or mint a reusable FPF kind, this pattern only classifies the phrase as recovered by value or as an understandable FPF extension candidate. It does not make the authority change by itself. Use the accepted source that already carries the decision by value; do not add a second decision source merely to restate the same content.

Fail closed:

• if the kind stack cannot be recovered, keep the term as plain or informative prose;
• if the relation kind cannot be recovered, keep the statement as a cue or split alternatives;
• if the publication construction cannot be recovered, do not use that publication, generic publication face, governed MVPK face, form, carrier, or rendered unit for work, evidence, gate, or authority claims;
• fill relationClaimSlice only when a relation claim is live, and fill admissibleUse plus projectSideFPFRef when an admissibility or project-side support claim is live;
• if the recovered wording is type-correct but leaves no remaining admissible reader move, recognition reason, Tech-to-Plain mapping when both registers are live, or exact FPF pattern application, or if a Plain or didactic line supplies practical force through unrecovered ontological, evidence, causal, assurance, bridge, gate, work, decision, or admissibility force, mark the rewrite incomplete or demote the phrase to quote-only wording, reduced-use cue, or blocked current transfer before using it as current pattern, architecture, DRR, or project text.

#Slash Discipline

In many standards, a slash can mark near-synonyms or parallel labels. In FPF-facing FPF episteme/publication ontology, a slash is a recovery trigger before it is a synonym marker.

Before leaving a slash expression in current prose, classify the expression as one of these cases:

• an accepted token, formal notation, file path, URL, quoted source wording, or product name where the slash is part of the carrier syntax;
• a plain-language synonym pair with no ontology, authority, evidence, or admissibility force;
• a composite-kind candidate that needs F.18 and A.6.P recovery;
• a relation claim that needs a RelationKind, a QualifiedRelationRecord, or a multi-term relation phrase with typed endpoints, slots, qualifiers, scope, time, and viewpoint;
• a tuple-like record that needs a named record kind and named slot semantics;
• a failed ontology signal where the sentence lists unlike objects because the live FPF kind, relation record, relation phrase, tuple-like record, or not-triggered disposition has not yet been recovered.

If the expression is not one of the first two safe carrier or plain-language cases, do not keep the slash as final wording. Write the recovered FPF kind, relation record, relation phrase, tuple-like record, or not-triggered disposition by value.

#Unclear Source Meaning and FPF Extension Candidates

Sometimes the problem is not a bad word but one of two different cases:

• the intended claim cannot be determined from the surrounding source, current FPF kinds, or current FPF episteme/publication ontology;
• the claim is understandable, but current FPF does not yet contain the kind, pattern, relation record, or method guidance needed to carry it.

Do not merge those cases. An unclear claim is not current architecture truth merely because deleting it feels risky, and it must not be rewritten by guessing a likely author intention. An understandable uncovered claim may be retained as a candidate FPF extension only when the problem situation, tempting overread, rejected current uses, current FPF gap, and the first user action that would improve are stated by value.

Classify the case explicitly:

• recovered by value: the text now names the exact U.Episteme, describedEntity, U.View, publication form, generic publication face, governed MVPK face, PublicationUnit, carrier relation, relation record, relation phrase, tuple-like record, FPF pattern, document with named source-basis, evidence-basis, architecture-basis, or review-basis role, exact project-side FPF kind and reference when projectSideFPFRef is live. The selected value is one live value, not the list: C.11 ChoiceResult; C.11 decision record; A.6.A action invitation; A.15 U.WorkPlan; A.15.1 dated U.Work occurrence; U.Method; U.MethodDescription; A.20 constraint or adjudication decision record; A.21 GateDecision; A.21 DecisionLogRef; A.10 evidence path; typed evidence record; B.3 assurance or engineering-justification record; typed status record whose FPF status pattern is named; carrier relation; front-end relation; or not-triggered alternative;
• understandable FPF extension candidate: the thought is clear enough to state as a candidate new or amended FPF kind, pattern, relation record, method guidance, accepted DRR content decision, or campaign-scoped content question, but it does not carry current authority, evidence, or admissibility force until an accepted architecture decision, accepted DRR, or accepted FPF pattern supplies that authority;
• quote-only source wording: the phrase may remain only as quoted source wording or provenance, with no current authority, evidence, or admissibility force;
• reduced-use cue: the phrase is kept only as a recognition cue or anti-case, not as a claim-bearing architecture decision;
• blocked current transfer: the phrase is not admissible for claim-bearing architecture, DRR, pattern, or project text until a new source, author clarification, or accepted architecture decision supplies the missing meaning, kind, or relation.
• rewrite incomplete: the repaired wording may be kind-correct, but it does not yet state a remaining admissible reader move, recognition reason, Tech-to-Plain mapping when both registers are live, or exact FPF pattern application, or a Plain or didactic line carries ontological, evidence, causal, assurance, bridge, gate, work, decision, or admissibility force that cannot be recovered from the Tech reading; continue repair or demote to a non-transfer disposition before treating the text as landed.

These dispositions are recovery results, not a meta-governance authority over all of FPF. When recovery names another exact FPF kind, that exact FPF pattern governs that kind, its admissible use, and its conformance checks. C.2.P may identify that A.10, A.15, A.15.4, A.20, A.21, B.3, C.11, F.9, E.17.EFP, E.17.ID.CR, or another exact FPF pattern is live. It does not govern the recovered kind after that identification. C.2.P only makes the live kind, relation, and use boundary explicit enough that the right governing pattern can be applied.

No other disposition is closed. In particular, "seems to mean", "probably about", a cleaner paraphrase, or a broad umbrella replacement is not a successful recovery.

#Core Glossary

#Cross-Side Fields That Must Stay Split

These fields are current episteme-publication precision vocabulary for DRR, architecture, and pattern-drafting work. They exist to prevent one sentence from mixing FPF-side admissibility, project-side records, actual work or action, method selection, carrier access, and authority records. They are local recovery aids, not FPF kinds, not record kinds, and not a universal record ontology. Each field closes only by naming the exact FPF kind, relation record, relation phrase, exact project-side FPF kind and reference, or explicit non-transfer disposition that is live in the sentence. The same local-aid rule applies to neighboring field names such as sourceSupportPosture, explanationSourcePosture, comparativeRelationPosture, representationValiditySupportPosture, allowedUse, misuseRisk, and worldContactPolicy: they help record a local recovery or reader-use boundary, but they do not become kinds. Posture fields do not instantiate evidence, gate, assurance, work, commitment, speech act, decision, release, authority, representation kind, world-contact kind, or policy kind. Read allowedUse as a local reader-fit field under admissibleUse, not as permission, evidence support, or authority.

#Episteme, Publication, Carrier Stack

#Trigger Profile Boundary

Lexical trigger scanning and direct known-target selection are governed by E.10:0.2, E.10:0.2a, E.10:0.2b, E.10:0.2c, and E.10:0.2d.

This pattern is applicable after that scan only when the exact governing pattern cannot yet be selected directly because the sentence still confuses source wording, claim-bearing episteme, publication/carrier construction, project-side reliance, pattern-application wording, or transfer/non-transfer disposition.

When this pattern is applicable, do not restart from word taste. Keep the E.10 trigger result as input and recover source-expression unpacking mode, FPF-transfer/conformance mode, live episteme-publication stack, transfer disposition, and remaining admissible reader move.

#Current Preferred Vocabulary

Use PublicationUnit when the intended object is a bounded, human-inspected unit inside a publication. Do not use it for UI behavior, carrier behavior, front-end behavior, file identity, dashboard behavior, or export behavior; use A.7, carrier wording, front-end wording, or the exact FPF governing pattern instead.

Use the current cluster names directly: PublicationUnit Stability Discipline, Local Head Restoration, and PublicationUnit Primary Described-Entity Discipline. When the live object is a bounded unit inside a publication, use PublicationUnit; when the live object is authoring or editing work, name that work directly.

Use primary described entity, DescribedEntityRef when local wording means the described entity named by a claim-bearing episteme or episteme-lane view.

Use ordinary topic, subject, or local object only in non-normative explanatory prose where no episteme slot, publication construction or authority relation is being asserted.

Do not mint any other new reusable FPF name from this pattern alone. PublicationUnit is governed by the E.17.AUD cluster named PublicationUnit Stability Discipline; this pattern recovers bounded-publication-unit wording into that head when the object is live and points to that cluster for governance. FPF-force-bearing uses keep the nearby definition or explicit publication stack.

#F.18 And A.6.P Admission Reading For PublicationUnit

This is the F.18 and A.6.P name reading that this pattern reflects from the selected E.17.AUD cluster correction. It records why PublicationUnit is the selected bounded publication-unit head for the E.17.AUD cluster, while C.2.P remains the epistemic precision-restoration pattern.

F.18 and A.6.P admission reading:
  Context: conformant FPF authoring and review where bounded publication units must not be confused with epistemes, views, publication forms, generic publication faces, governed MVPK faces, carriers, authoring work, or review process.
  Kind: governed-object head for a bounded unit inside one publication.
  Purpose and use-domain: keep one human-inspected publication unit distinct from episteme, view, publication form, generic publication face, governed MVPK face, carrier, authoring work, and review process.
  Selected Tech label: PublicationUnit.
  Plain reading: bounded unit inside a publication that a person inspects as one unit.
  Candidate head families considered:
    - authoring-centered unit labels
    - reading-centered unit labels
    - mixed authoring-and-reading unit labels
    - PublicationReadingUnit
    - PublicationAuthoringUnit
    - PublicationUnit
    - ContentSpan
    - DocumentUnit
  F.18 result:
    - `PublicationUnit` has better SemanticFidelity than authoring-centered unit labels because the unit belongs to the publication lane, not to the authoring process.
    - `PublicationUnit` has better MorphologicalActionFit than mixed authoring-and-reading unit labels because it does not mix author, reader, and unit-boundary roles in one head.
    - `PublicationUnit` has lower AliasRisk than `content span` and `document unit` because `content` and `document` blur episteme, publication form, and carrier.
    - `PublicationUnit` still has nonzero AliasRisk because `publication` itself splits into act or occurrence of publishing, episteme publication, form, generic face, governed MVPK face, unit, and carrier; therefore FPF-force-bearing uses keep the nearby definition or explicit publication stack.
  Current status: admitted reusable FPF head for conformant episteme-publication-heavy FPF text within the declared bounded-publication-unit repair scope; use a more specific already accepted head where one governs the text under repair.

#Epistemic Precision Restoration After E.10

Lexical trigger rewrite rules are governed by E.10:0.2b, E.10:0.2c, and E.10:0.2d.

Use this pattern after those rules only when one of these remains unresolved:

• source-expression unpacking mode versus FPF-transfer/conformance mode;
• source-local meaning versus current FPF wording;
• claim-bearing episteme versus publication, view, face, carrier, or publication unit;
• described entity, grounding relation, or source cue versus project-side evidence, work, gate, decision, assurance, method, action, release, or engineering justification;
• declarative FPF pattern application versus project work or control flow;
• transfer disposition: recovered by value, quote-only wording, reduced-use cue, understandable FPF extension candidate, blocked current transfer, rewrite incomplete, or not triggered;
• remaining admissible reader move or Tech-to-Plain mapping after epistemic precision repair.

When none of these remains unresolved, apply the exact governing pattern selected by E.10 directly.

#Rewrite Execution Modes

Use the smallest sufficient mode that preserves the distinction. The template is an epistemic precision device, not a form to fill for every ordinary wording cleanup.

#Local prose cleanup

Use this mode when the phrase under repair is non-normative local prose and does not carry ontology, authority, review scope, release posture, admissibility, or a reusable name.

Action: rewrite directly or leave it unchanged. No table row is required.

#Compact epistemic precision-restoration row

Use a compact row for ordinary architecture and source-basis or review-basis document cleanup where a sufficient FPF kind, relation record, relation phrase, or tuple-like record can be recovered without minting a new FPF head.

Compact epistemic precision-restoration row:
  file path, if live:
  FPF pattern, if live:
  pattern section, if live:
  sentence reference:
  phrase under repair:
  live sentence function:
  selected exact FPF kind or exact project-side FPF kind:
  `relationClaimSlice` triggered? yes or no
  relation problem, if triggered:
  admissibleUse triggered? yes or no
  projectSideFPFRef triggered? yes or no
  relation claim? yes or no
  if relation claim:
    RelationKind:
    endpoint, slot, qualifier notes:
    admissibilityTargetKind:
    admissibilityTargetRef:
  if local current-context unpacking:
    FPF-side kind, reference, or relation:
    exact project-side FPF kind, if live:
    exact project-side reference, if live:
    notTriggeredReason:
  replacement:
  remaining admissible reader move:
  distinction disposition: preserved, split, intentionally retired, still missing

#Full epistemic precision-restoration check

Use the full check when the wording may change ontology, introduce or retire a reusable head, change a claim-bearing pattern or document with named source-basis, evidence-basis, architecture-basis, or review-basis role, or resolve a contested source-meaning problem.

Epistemic precision-restoration check:
  file path, if live:
  FPF pattern, if live:
  pattern section, if live:
  sentence reference:
  phrase under repair:
  sentence function:
  distinction carried:
  E.10 head kind and Intension, Description, and Specification reading:
  F.18 naming status: no stable term, reuse, MintNew sketch, DocumentLegacy
  F.18 candidate head families, if naming is live:
  F.18 lexical Q result, if naming is live:
    SemanticFidelity:
    CognitiveErgonomics:
    MorphologicalActionFit:
    AliasRisk:
  A.6.P trigger? yes or no
  A.6.P selected relation kind, slots, qualifiers, if live:
  claim-bearing episteme live? yes or no
  FPF kind stack:
  describedEntity, grounding, ClaimGraph, viewpoint slots triggered:
  E.17 and MVPK publication form, generic face, governed MVPK face, view, carrier split:
  PublicationUnit typing, if any:
  FPF-side or project-side sentence:
  `relationClaimSlice` triggered? yes or no
  relation problem, if triggered:
  admissibleUse triggered? yes or no
  projectSideFPFRef triggered? yes or no
  relation claim? yes or no
  if relation claim:
    RelationKind:
    QualifiedRelationRecord slots:
    admissibilityTargetKind:
    admissibilityTargetRef:
  if local current-context unpacking:
    FPF-side kind, reference, or relation:
    exact project-side FPF kind, if live:
    exact project-side reference, if live:
    notTriggeredReason:
  rejectedOverread, if live:
  project-side record, work, action, method, carrier crossing:
  heterogeneous-list classification: one live kind, relation stack, tuple-like record, alternative cases, failed ontology, not triggered
  pattern application, project work, decision distinction:
  chosen rewrite:
  remaining admissible reader move:
  distinction disposition: preserved, split, intentionally retired, still missing
  unrecovered wording retained? no, yes, with scope and reason:
  transfer disposition: recovered by value, extension candidate, quote-only, reduced-use cue, blocked transfer, rewrite incomplete, not triggered

#Epistemic Precision-Restoration Note

Use an epistemic precision-restoration note only when wording carries ontology, authority, evidence, or admissibility force. The note records the original phrase, recovered FPF kind or relation, exact reference when live, project-side FPF kind and reference when live, remaining admissible reader move, and disposition: recovered by value, extension candidate, quote-only, reduced-use cue, blocked transfer, rewrite incomplete, or not triggered.

#Ordinary Completion and Reopen Boundary

A C.2.P application is complete for ordinary pattern-authoring use when the smallest sufficient product is present:

1. E.10 trigger result is kept as input and the text does not restart from word taste;
2. the wording is either left ordinary, repaired locally, expressed as a compact epistemic precision-restoration row, or escalated to the full check because the live claim requires it;
3. the recovered episteme, publication, view, face, carrier, publication unit, described entity, grounding relation, project-side reference, or transfer disposition is named by value;
4. every relation-like slice that remains live is assigned to A.6.P or its retained specialization, rather than being hidden inside this pattern;
5. the remaining admissible reader move survives in ordinary prose or the wording is explicitly demoted to quote-only wording, reduced-use cue, blocked current transfer, rewrite incomplete, or not triggered.

Use the lowest sufficient product. A clean sentence is enough when one sentence fixes the claim force. Use a compact row when the reader must inspect one recovered kind, relation, or disposition later. Use the full check only when several fields are live, when source-to-FPF transfer is contested, when a durable name may be minted, or when a publication/carrier/project-side overread would otherwise survive.

This pattern can be applied to its own wording at the same lowest sufficient mode. If C.2.P text itself blurs a source expression, publication construction, pattern application, relation slice, or project-side reliance claim, repair that local wording here; do not create a recursive pattern-quality apparatus.

Reopen or lower a prior C.2.P repair when one of these content discoveries appears:

• the replacement head is another umbrella word such as support, surface, route, kind, object, record, map, or mapping without exact FPF kind and boundary;
• the repaired wording is type-correct but no longer tells the working reader what action, non-use, or neighboring-pattern application remains;
• a neighboring pattern is now the true governing pattern for the live evidence, assurance, gate, work, decision, publication, architecture, structure, relation, or naming claim;
• an entry cue, ToC row, summary, dashboard, retrieval snippet, or source-basis note preserves the pre-repair broad interpretation after the pattern body was repaired;
• repeated use shows that authors are filling the full check where a local sentence or compact row would suffice.

The ordinary stop condition is local: once the current sentence or bounded publication unit preserves kind, relation, transfer disposition, and remaining admissible reader move, stop. Do not keep improving wording merely because a more elaborate record could be filled.

#Archetypal Grounding

#Boundary and Anti-Cases

#Transfer Coverage

C.2.P is intentionally narrow but must transfer across three recurrent publication situations:

• FPF-side drafting: pattern text, DRR text, source-basis notes, review-basis notes, and pattern-host prose;
• project-side publication: dashboards, explanations, cards, documents, front-ends, rendered files, and generated summaries used around evidence, work, gates, decisions, assurance, or methods;
• external source transfer: seminar fragments, papers, reviews, standards, and tool outputs being clarified before possible FPF use.

In all three situations the same invariant holds: recover the source/current distinction, claim-bearing episteme, publication/carrier construction, relation-like slice, exact neighboring pattern, and remaining admissible reader move before accepting the wording as current FPF text.

#Bias-Annotation

#Conformance Checklist

#Current Scan Boundary

Lexical trigger scanning is governed by E.10:0.2, E.10:0.2a, E.10:0.2b, E.10:0.2c, and E.10:0.2d.

C.2.P conformance begins only when the E.10 result is epistemic precision restoration required or combined precision restoration required, or when non-FPF source text is being unpacked before possible FPF transfer.

Do not copy the E.10 trigger list into this pattern as a second registry. Use the E.10 result as input and recover source-expression unpacking mode, FPF-transfer/conformance mode, live episteme-publication stack, transfer disposition, and remaining admissible reader move. When the relation-bearing slice is live, A.6.P remains a separate required precision-restoration pattern.

#Common Anti-Patterns and How to Avoid Them

#Consequences

#Operating Consequence

For new episteme-publication precision prose:

• start from FPF kinds and relations, not from familiar publication nouns and document nouns;
• use PublicationUnit for bounded publication units;
• use describedEntity only when the episteme slot is live;
• keep publication form, generic publication face, governed MVPK face, view, carrier, document with named source-basis, evidence-basis, architecture-basis, or review-basis role, review object, and exact project-side FPF kind and reference separate;
• name relationClaimSlice, admissibleUse, and projectSideFPFRef separately when more than one is live;
• classify heterogeneous kind lists before writing a sentence that depends on them;
• say that FPF patterns are applied in problem situations, not called or routed as procedures;
• leave accepted FPF names untouched unless a separate accepted naming decision authorizes a rename.

Operationally, each rewrite should:

• separate FPF-side episteme and publication context from project-side episteme and publication context whenever both are present;
• name relationClaimSlice, admissibleUse, and projectSideFPFRef separately when a publication, display, cue, or explanation is treated as evidence, gate, constraint, adjudication, decision support, work permission, assurance, or engineering justification;
• classify heterogeneous lists before naming them: one live kind, relation stack, tuple-like record, alternative cases, not-triggered alternatives, or failed ontology;
• say that FPF patterns are applied in problem situations, while project records, publications, views, carriers, and actions are worked with in project practice;
• avoid strength metaphors unless the characteristic, scale, threshold, evidence class, or admissibility relation is named.

For cleanup of existing conformant texts:

• do not do a global string replacement;
• classify each unclear term occurrence by the smallest sufficient rewrite mode;
• use the full epistemic precision-restoration check only when ontology, reusable naming, FPF pattern text, or source-bearing project text is live;
• do not rename accepted FPF patterns from this pattern alone.

#Rationale

FPF already contains the relevant ontology. The recurring defect was not lack of concepts but ad hoc wording that bypassed them: source, target, surface, object, host, route, supported use, and similar terms packed several FPF kinds and relations into one convenient phrase.

The correct repair is therefore not a new umbrella. It is a disciplined recovery action: use E.2, E.10, F.18, A.6.P, A.7, C.2.1, E.17.0, E.17, and MVPK together until the sentence says what object, relation, publication, view, carrier, record, work, action, or pattern application it means.

Because E.2 governs all normative FPF patterns, epistemic precision is not a value apart from P-2 Didactic Primacy. An epistemic precision restoration may be stricter than the original wording, but if it turns FPF-force-bearing reader-facing problem text into a kind inventory with no working situation or first useful move, it has not landed the FPF repair. The remedy is not expressive license and not metaphor removal; the remedy is admissible recognition wording whose claim force remains recoverable through the Tech reading or a named exact FPF pattern application.

The detailed rules remain in ordinary pattern sections, so the pattern is usable as FPF guidance rather than as an external glossary container.

#SoTA-Echoing

C.2.P does not claim to replace semiotics, terminology science, document engineering, or ontology engineering. Its live claim is narrower: episteme-publication-heavy conformant text must recover accepted FPF kinds and relations before it is rewritten, so that episteme, publication, view, carrier, naming, relation, and project-side records are not replaced by ad hoc words.

Full external SoTA comparison is therefore not the governing evidence mode for this architectural precision-restoration pattern. A reduced external practice set is still required because the pattern governs terminology drift and epistemic precision restoration. The reduced set is admitted for the recovery discipline; it does not create a new ontology and does not outrank the FPF patterns named below.

External-practice boundary. External traditions are admitted only through the exact local FPF invariant they sharpen. Object-oriented modeling and OWL-style ontology modeling do not become the default repair for vague FPF wording. Architecture-description standards help keep views, viewpoints, concerns, and descriptions explicit. Explainability and NLP faithfulness work helps prevent explanation laundering. RAG evaluation helps separate retrieval support from answer trust. Quality-diversity and multi-objective search help avoid premature scalarization in candidate selection. None of these traditions becomes FPF ontology, FPF authority, or a universal pattern-quality benchmark.

#Internal FPF Governing Patterns

The current FPF corpus already has explicit governing patterns for this discipline:

• E.10 supplies the head-kind, term, morphology, register, and forbidden-umbrella discipline.
• E.10.D2 gives the "thing vs words vs rules" discipline and the carrier humility rule.
• F.18 gives the local-first naming protocol: Context, Kind, purpose and use-domain, local sense, candidate head families, NQD-front, semantic read-through, and lexical Q components before one label becomes a reusable head.
• A.6.P gives the relation-precision restoration method: restore generic head kind, build candidate sets for endpoint kinds and relation kinds, select kind-explicit slots and qualifiers, then allow guardrailed wording.
• C.2.1 gives the episteme slot graph and describedEntity discipline.
• A.7 keeps Object, Description, and Carrier distinct.
• E.17.0, E.17 distinguish views, viewpoints, MVPK faces, publication forms, and publication projections.
• A.15.4 is a good current pattern example of keeping encountered publication, display, or cue items distinct from the exact project-side FPF kind and reference that makes work or reliance admissible.
• A.16, A.16.0, A.19, B.2.5, C.27, and A.3.3 provide the movement, control, temporal stack used when episteme-publication prose talks about route, trajectory, movement, cadence, or dynamics.
• E.19 already treats terminology and sentence-level precision restoration as required review checks, not editorial polish.
• A.6.A carries action-invitation discipline when a publication, representation, or cue invites an action without itself becoming authority, evidence, gate passage, or work completion.
• C.11 carries decision-making and decision-record discipline when the live question is a decision rather than generic action.
• A.15 and A.15.4 split role, method, work-plan, and actual-work alignment from work-relevant source restoration, so episteme-publication prose must not let A.15 become a universal episteme-publication governing pattern.
• E.9 is the campaign DRR pattern for campaign-level content decisions; E.11 is only for entry-discoverability situations and must not organize an episteme/publication repair by default.

The internal FPF governing patterns remain primary:

This reduced external-practice set changes the Solution in one practical way: an epistemic precision restoration cannot close merely because the replacement wording sounds cleaner. It closes only when the FPF kind, relation, admissible use, and any exact governing-pattern application is recoverable by value; otherwise the wording is blocked, quote-only, or becomes a candidate for a separate FPF-kind decision.

#E.19 Review Profile Carry-Through

Run PCP-TERM when the repair changes episteme-publication-heavy naming, umbrella words, slash compounds, trigger-word replacements, or relation wording.

Run PCP-BRIDGE when the repair imports terms, claims, norms, or authority expectations across contexts, disciplines, reference schemes, publication forms, or project record kinds.

Run PCP-ENTRY only when the repair changes which FPF pattern a working reader should apply in a problem situation. Do not use PCP-ENTRY as a substitute for the epistemic precision restoration itself.

Run PCP-PRAG when the repair changes reader action, practical payoff, or what the working reader can safely do next. A type-correct but inert repair fails this line even when every recovered kind is technically right.

When the repair changes evidence, proof, witness, grounding, explanation, gate, release, or engineering-justification claims, apply the exact governing FPF pattern for that claim (A.10, E.17.EFP, A.20, A.21, B.3, or another exact pattern when live) and select the live E.19 profile by the changed pattern claim. Do not mint a local evidence-review profile inside C.2.P.

#Relations

• Builds on: E.2 Pillars, especially P-2 Didactic Primacy; E.10, E.10.ARCH, A.7, F.18, A.6.P, C.2.1, E.17.0, E.17, MVPK, and A.6.A.
• Coordinates with: E.6, E.7, E.8, E.9, E.12, E.19, A.10, A.15, A.15.4, B.3, A.20, A.21, A.6.F, A.6.3.CSC, A.6.3.CR, A.6.3.RT, C.30.P, C.16.P, C.16.Q, E.17.EFP, and E.17.ID.CR.
• Does not replace: E.10 general lexical rules, F.18 naming protocol, A.6.P relation precision, or local episteme/publication patterns. It tells authors when those patterns must be applied to episteme-publication-heavy wording.

#C.2.P:End

Last Updated: 2026-05-31 — this section last modified in upstream FPF commit 16cd3138 (github.com/ailev/FPF)