Library / First Principles Framework (FPF) - Core Conceptual Specification
Jump to passage
In this reading

Link to current text

Published source confirmed at last check

Source changed 2026-10-03 11:52:20 UTC · snapshot created 2026-10-03 11:53:41 UTC · last check 2026-10-03 14:15:10 UTC

E.19:7 - Conformance Checklist

IDRequirementPurpose
CC-E19-0 (Review, result, and decision remain distinct).The reviewer performs the review; the profile or checklist declares questions; the findings, review record, or optional aggregate result state review claims; intended-reader and representation positions specify audience or form; and an authority-bearing decision uses those claims under its own rule. If an account asserts actual review, repair, or verification U.Work, the §4 actual-Work account MUST hold; a compact result may omit only an assignment identifier unused by the receiving claim. Keep any local system-role kind and separate System-classification judgment distinct, and route unresolved source role through E.10.ROLE.Prevents editorial declarations and records from acting as reviewers or authority without burdening an ordinary review with unused identities.
CC-E19-0a (MVPK face and View boundary).When a PCP inspects an MVPK face or projected publication form, apply no-new-claim and no-shadow-default to the claims and defaults actually carried by that face/form. Keep selected episteme, optional projection/construction, face, publication form, publication occurrence, rendering, and carrier distinct. Assert U.View membership only for the selected episteme when exact E.17.0 conformance independently obtains; profile selection, projection, or compliance with these two checks supplies no membership.Preserves publication/face discipline without turning E.19 into a View classifier.
CC-E19-1 (Baseline triage is mandatory).Every configured PQG review MUST include PCP-BASE for the reviewed pattern or subset. When the baseline review is complete and finds no risk requiring another profile, the review may finish after any small mechanical defect is either repaired and checked or returned as an independent finding. This is only an E.19 review boundary; it cannot support an E.21 coordinate value, PatternQualityStatus, a claim that every coordinate is 4 or every coordinate is 5, landing-quality claim, or improvement-movement claim without the complete E.21 result required for that claim.Ensures one shared triage floor without turning every review into a full audit or substitute quality measurement.
CC-E19-2 (Profile selection covers the live risks).The review scope MUST name PCP-BASE, every risk-selected PCP, the risk selecting each additional profile, and any override. It MUST consider the whole current profile set rather than only the easiest visible family. When an override affects a later admission, refresh, or other governing decision, its false-positive reason and compensating check belong in that decision basis; successful profile choices need no per-profile pass entries. An unselected profile requires no result row or durable disposition.Makes review depth repeatable without a separate record of successful checks.
CC-E19-3 (Delta-Class and actual impact for material changes).If the reviewed pattern change is Δ-2/Δ-3 under E.15’s actual-effect test, the governing change or decision result MUST carry its Delta-Class, actual dependent reach, a DRR pointer when a material content decision was selected, and the focused refresh, verification, or F.9 consequences that the changed use requires. E.19 repairs or reports a missing or false account; it does not duplicate a successful one.Keeps evolution controlled while leaving change evidence with the change decision.
CC-E19-4 (Conformance-claim coherence is enforced).Inspect-repair-verify MUST eliminate orphan and unclaimed requirements by aligning the reviewed pattern’s Conformance Checklist, deontic clauses, admissibility constraints, and Solution. Independent findings MUST identify each surviving incoherence and give concrete repair direction.Preserves the CC as the enforceable conformance check set in both review forms.
CC-E19-5 (Triage & noise discipline).The run SHOULD prioritize FPF-governed sections and deontic requirements (e.g. CC, content of deontic clauses and content of admissibility constraints, definitions, Relations, SoTA, modularity) and keep purely mechanical edits (e.g. RFC-form deontic cleanup) minimal. Template defects MUST be fixed before admission or a positive refresh conclusion but MUST NOT be used to skip semantic review.Improves semantic trust without turning review into form-only compliance.
CC-E19-6 (Review form and findings completeness).The review MUST choose one form from E.19:4.1 and inspect every independently answerable in-scope question even after the first defect, blocker, or negative conclusion. Inspect-repair-verify ends with every in-scope defect repaired and focused verification performed; independent review ends with one complete set of actionable defects and blockers plus concrete repair direction. A question that reveals no defect gets no durable pass entry. Early stop is allowed only when the remaining questions cannot be judged truthfully or safely, and then the unexamined scope and cause MUST be named.Prevents both first-defect stopping and a third, report-producing review form.
CC-E19-7 (Recognition text, assurance text, and self-containment).Admission or refresh runs for new and substantially revised patterns MUST check that a first-reading recognition text appears early enough for the intended reader, that the heavier assurance text remains visibly second rather than becoming the first real point of entry, and that the assurance text does not silently shift the recognition-text claim. The run MUST check for a recognisable working situation, what goes wrong if the pattern is missed, what the pattern buys, the first admissible action-guiding move the user should take, and an ordinary not this pattern when boundary; for any FPF-governed typed declaration or modeling lens, the run MUST confirm that a short user-facing statement exposes the primary EntityOfConcern, relation record, or claim record and the minimal lens that keeps it reviewable; the run MUST also check that the primary EntityOfConcern, relation record, or claim record keeps one stable kind across title, opening function, declaration function, worked slices, and related-pattern or companion guidance named by value rather than drifting between the named primary EntityOfConcern, an act, a work-result record, and carrier-placement labels. When a broader umbrella name and a narrower operative branch are both used, the run MUST check that the recognition text makes that stack explicit enough to identify the umbrella, the active branch, the primary EntityOfConcern, the move, and the wider work or process that still remains outside. The recognition text MUST start from a recognisable problem-owning domain or practice moment whenever that can be done without loss of precision, rather than opening first with internal package architecture or taxonomy language. Early FPF-governed technical terms MUST receive nearby pairwise plain glosses; transform-like families MUST carry concrete worked slices plus ordinary-vs-FPF-governed wording guidance where needed; and any SoTA-Echoing used as explanatory grounding MUST state a short practitioner or manager implication plus visible linkage to the worked cases or boundary slices it disciplines. If SoTA or practice tradition has FPF-governed use, the run MUST check that primary-EntityOfConcern choice, narrowed-branch choice, and practical payoff remain answerable to the relevant domain or practice rather than only to internal package architecture. If a pattern claims universal or transdisciplinary usefulness, the run MUST check that this breadth is already demonstrated in the recognition text through heterogeneous situations adequate to the claimed breadth, with F.16 preferred as the example-matrix template.Prevents architecturally correct but reader-opaque patterns and keeps broad claims from appearing only late in the assurance text.
CC-E19-7a (Precise-language repair cannot leave inert recognition).If admission or refresh includes a precise-language repair, apply F.19 to the changed span and use E.10 only for compact FPF routing. Check that the intended reader can recover why the distinction matters, the reader use, and the pattern or rule contribution that carries any formal claim. Keep Plain or didactic wording ordinary when it adds no such claim; otherwise map it back to the repaired Tech reading. Add an exact assertion, predicate, ClaimGraph, or displayed identity only when it distinguishes truth, action, stop, or named reliance. If the Tech reading asserts dated Work, apply CC-E19-0. In inspect-repair-verify, restore any harmed working situation or first useful move; in independent review, record the exact harm.Prevents type-correct cleanup from destroying practical guidance or forcing unused formal apparatus.
CC-E19-8 (Whole-span precise-language repair).Apply F.19 to the complete natural span, including predicates inside negation or modality, required operands and relational complements, referents, subject-predicate compatibility, coordination, lists, modifiers, guards, and governing-claim order. Use compact E.10 cues to locate candidates and open E.10.ARCH or an exact subject pattern only while an FPF kind or relation remains unresolved. After a wording or syntax change, reread the changed sentence and only its meaning-dependent neighbors. Inspect-repair-verify leaves repaired wording and focused verification; independent review records only a failed repair or blocker.Keeps precise language semantic and usable without duplicating F.19 as a phrase-by-phrase account.
CC-E19-9 (Package-form, concrete pattern contribution, and package-relation function-word discipline).Use the package-form and relation cues in E.10:0.2 to check whether the text preserves the actual package form, concrete pattern contribution, and package relation. If a repair introduces or retains a head already occupied elsewhere in FPF, verify intentional reuse or repair/report the collision.Keeps concrete pattern contributions, package relations, review functions, and package forms legible without recording successful collision checks.
CC-E19-10 (Reader-fit discipline).Check the reviewed pattern or subset for the intended FPF user, an explicit primary reader/concern/viewpoint when several readers are served, and separation of user guidance from package-development, review, evaluation, projection, integration, or release reasoning about the same pattern version. Part E patterns may govern authoring or review as their declared subject matter, but that does not admit development correspondence about the current version. Repair each leak or return its exact locus as a finding; sections with no leak need no scan recital.Keeps reviews from accepting conceptually correct but reader-confused patterns.
CC-E19-10a (Quality/projection carrier leakage).Check whether pattern prose, including Relations, Rationale, SoTA-Echoing, worked slices, examples, tables, and the Conformance Checklist, contains corpus projection, retrieval/cold-reader evidence, publication parity, integration evidence, PatternQualityStatus, all-4/all-5 posture, or development correspondence about that pattern version. This is a sentence-function check, not a lexical search. Move such material to the applicable E.21 result, E.19 findings, README/ToC/E.11, projection, publication, integration, or release result and retain only the pattern’s admissible user-facing move or boundary.Prevents quality and projection proof from becoming pattern prose.
CC-E19-11 (Precision before relaxation).If remediation preserves or introduces a Plain, didactic, or coarsened restatement of a repaired FPF-governed sentence, the run MUST keep a more precise upstream interpretation recoverable and must not let the softened form become the only wording with authority-reference claim kind or admissible-use boundary.Keeps later readability aids subordinate to an explicit more precise interpretation.
CC-E19-12 (Integration impact is checked).Before publication or integration of a new or substantially revised subset, inspect related patterns and the concrete constraints or tests they supply, companion notes, Relations entries, and affected published sections. Repair each in-scope mismatch or return it as a finding and name any genuinely outside boundary. Successful synchronization remains in the changed sources and governing publication, integration, or release result.Prevents an isolated local improvement without duplicating synchronization evidence.
CC-E19-13 (Usability and proxy-to-value are checked).For a new or substantially revised subset, check recognition versus assurance text, first-minute situation, practical payoff, ordinary boundary, worked slices, primary reader/viewpoint, and the applicable E.8, E.12, E.13, E.14, E.17.*, F.16, or local-equivalent questions. Repair or report a usability defect. If a score, coordinate, benchmark, projection signal, or all-5 posture is used as value evidence, the governing E.13 result—not an E.19 pass account—must carry intended value, proxy use, gains, losses, minimally viable value slice, and reopen condition.Prevents visible review success from replacing practical value.
CC-E19-14 (Scenario, anti-case, and utility fit are checked when applicable).When the domain has a relevant scenario pack, anti-case corpus, pilot bank, utility tree, fitness catalog, or analogous common source, use its applicable cases and qualities. Repair a failing case or return the exact failure, missing source, or out-of-scope boundary as a finding; do not record cases that revealed no defect merely to prove consultation.Keeps common validation sources active without a separate consultation record.
CC-E19-15 (Packaging, concrete pattern contribution, package relation, and shipping fit are checked).Before a publication or integration claim, inspect the relevant package form, the definition, constraint, test, or other pattern contribution actually used, package relation, publication function and authority reference, and the actual publication and integration facts. Repair or report any mismatch. The governing publication, integration, or release result carries the successful state claim; E.19 does not repeat it.Keeps shipping claims truthful without a second state account.
CC-E19-16 (Domain-tightened profile depth is applied).When a domain-specific depth note such as semio FIT-* applies, use it to tighten the selected PCP questions. Repair or report any defect it reveals; do not add positive or not-found recitals to an E.19 result.Keeps domain-specific depth operative rather than optional folklore or extra reporting.
CC-E19-17 (Companion-material retention is justified).When a new or refreshed pattern subset keeps a long-lived companion, profile, check sheet, pattern-local companion row, review harness, or analogous selected non-pattern FPF kind-reference pair, the retention basis MUST make its companion function explicit: companion use question, concrete pattern contribution or selected non-pattern FPF kind-reference pair served, admissible companion-only use, one real breakage if absent, and retention, accepted-source-material-only, or removal condition when no such breakage exists. Use the governing retention or design decision; no separate E.19 pass account is required.Prevents companion material from remaining by inertia or becoming hidden authority after the pattern body already carries the usable guidance.
CC-E19-18 (Substantive solution and locus adequacy is checked).A new, refreshed, or materially repaired subset MUST receive a pattern-specific substantive adequacy check unless the change is purely mechanical. Check whether it still solves the stated problem, assigns claims to the correct governing loci, preserves kind boundaries and selected companion/projection functions, keeps SoTA grounding current enough, remains usable without excess apparatus, and worsens no content relation. Repair each in-scope failure or return it as a finding and name any needed wider boundary. Questions that reveal no defect need no separate account.Prevents clean checklists and terminology from hiding wrong content.
CC-E19-19 (Accepted-decision carry-through is checked).When the reviewed pattern, subset, or current change is claimed to implement an accepted DRR, repair findings, intake material, architecture source material, or other accepted source named by value, inspect each applicable decision against the reviewed loci or the named concrete pattern contribution, claim, companion, result, or source that carries it. Require exact predicate or defining ClaimGraph identity only when the decision or named reliance needs it. Repair or report partial, missing, wrongly rejected, wrongly routed, or wrongly classified carry-through. The accepted source remains the decision source; do not duplicate decisions expressed sufficiently, inherited unchanged, correctly absent, or outside the subset. Keep E.17.ID.CR units, PublicationUnit, publication forms/faces, source materials, and project-side review relations in their governing kinds.Prevents accepted decisions from disappearing without making E.19 their second authority.
CC-E19-20 (Project-side reuse has its own governing result).Reuse outside FPF pattern-quality review requires a project-side governing result that names the project claim, relation, required evidence or assurance, and the exact contribution taken from E.19. The E.19 result remains scoped to the reviewed FPF pattern edition or subset.Keeps pattern review and project-side decisions under their respective rules.
CC-E19-21 (Precise-language distribution is preserved).When the reviewed change repairs language or edits its rules, check the selected distribution: F.19 owns the common whole-span semantic and pragmatic repair; E.10 supplies compact cues and exact routing; E.10.ARCH opens only for unresolved ontological recovery; exact subject patterns define or test the recovered object or relation; affected patterns keep only thin cues unless the recovered subject is their own EntityOfConcern. A review fails this row when an affected pattern grows a rival normal-pass algorithm, mandatory counterreading field, duplicate trigger registry, or check-internal attention mechanics.Prevents pattern admission or refresh from duplicating the language method or reintroducing ungrounded guards.
CC-E19-22 (EntityOfConcern and precise-language triage is applied).For the changed span and affected pattern texts, apply the review-specific continuity questions in §4.2.1 and the common F.19 reading. Compare admissible use before and after independently of whether an explicit guard is warranted. Add formal identities only when truth, a live distinction, or named reliance needs them; apply A.3.1 and A.3.2 before claiming Method or MethodDescription, and CC-E19-0 to dated Work. Use E.21 PrecisionRestorationProfile when that evaluation is active.Prevents a type-correct rewrite from changing referent, path, use, or consumer behavior while keeping deep restoration auxiliary to the pattern claim.
CC-E19-23 (Pattern-edition use-value replay preserves distinct outcomes).When E.8:4.1.2 selects a material edition change, judge separately only the affected use probes and changed wording groups whose result can differ. Apply F.19 to each group: compare its subject, predicate, participants, modal force, referents, operands, contribution, information order, and applicability boundary under its governing rule. Check widening and narrowing against that rule and the accepted change decision independently of reader-plausibility; use the plausible-reader test for explicit-guard contribution. Do not invent an alien case or guard. Replay the positive Solution and each coordination or enumeration member whose membership or contribution can differ. Reuse prior results only when object, editions, scope, and assurance question match. Run this once on the stable candidate before acceptance or landing; do not create per-keystroke review, a second ledger, or positive recitals.Prevents broad-use preservation from hiding semantic drift while keeping bounded edits cheap.