CC-E21-0 | Keep the exact checked pattern edition, characteristic space and specification, evaluation configuration, admitted evaluator U.System and ordinary assessing action, coordinate-result claims, aggregate result episteme, and any admission or refresh decision distinct. Do not invent a system-role kind, assignment, Method, Work, or A.6.1 application for the ordinary form. When actual assessment U.Work is asserted, item 5 MUST hold: every precise evaluator-performer has an A.13 core, and A.15.1 independently admits the Work from its performance history, enacted Method, extent, and containing-System relation under the exact boundary and qualification window. F.6 MUST be added only when the evaluation account also needs precise assignment-bound attribution. A compact account may omit an unused identifier only when every consumed relation remains recoverable. The Work assertion alone establishes neither an application nor a result. Keep the Work, any returned value or direct evaluation-result relation, and the C.2.1 result episteme separate; connect them only through an exact A.6.1 result binding or a separately declared direct evaluation-result relation that actually obtains. When an exact application is asserted, the compact A.6.1 rule in E.21:4 MUST hold and its required application-and-binding account MUST be recoverable. Keep witnesses or evidence use, optional record, status use, assurance, publication, currentness, and later repair separately recoverable when claimed. |
CC-E21-0a | Constitute each coordinate value as a quality ascription about the same exact checked pattern edition with recoverable ReferenceScheme, characteristic, scale value, evaluation rule or probe, scope, use, and window, ordinary assessing action or exact assessment application when asserted, rationale, and evidence locus. Keep all required coordinate claims in one non-arithmetic PatternQualityQBundle payload carried by a separate C.2.1 aggregate result episteme; evaluator identity, viewpoint, witness presence, record placement, or bundle membership supplies neither value nor grounding by itself. |
CC-E21-1 | Recover ClaimScope from the governing evaluation question: the current request, an E.22 frame, an accepted decision or content source named by value, a landing or release check, a review request, or another actual quality-use request. Then name PatternOfConcernRef, ClaimScope, WorkingReaderScope, IntendedUse, QualificationWindow, and EvaluationEvidenceBasis. |
CC-E21-2 | Evaluate the full RequiredPatternQualityCoordinates set. |
CC-E21-2a | Before assigning coordinate values, apply F.19 and record one PrecisionRestorationProfile under E.21:4.3a with overallEffect, checkedLoci, and affectedCoordinates. Add issue-bearing detail when it changes the result; ordinary repairs use F.19’s local revalidation, and a risk of changed FPF-governed meaning opens kindRestorationCheck. |
CC-E21-3 | Use the result-bearing three-column table: coordinate, value, and ShortRationale; a two-column coordinate-and-value table is not an E.21 result. |
CC-E21-4 | Let floorEvaluation change floor and evidence cost only, not the coordinate set. |
CC-E21-5 | Assign values from checked pattern content and named content evidence, not review, landing, popularity, praise, or absence of prior use. |
CC-E21-6 | For corpus-facing values, name the checked README, ToC, E.11, card, retrieval, monolith, or projection loci. Lower a value for an established absence under its anchor; leave it unestablished when the necessary check or basis is unavailable, following §4.3. |
CC-E21-6a | Keep corpus-projection; README, ToC, E.11 alignment; retrieval or cold-reader evidence; monolith-parity; PatternQualityStatus; developer, reviewer, and executor correspondence; and other quality evidence out of the pattern unless the pattern’s own EntityOfConcern and user-facing action are that evaluation or projection work. Part E patterns may define or guide FPF-pattern authoring, review, evaluation, entry, or publication as their subject matter; that does not license rationale or instructions about developing the same pattern version. Test what the sentence is doing, not whether it contains a listed word. If such material appears anywhere in the pattern, including notes, appendices, Relations, Rationale, SoTA-Echoing, examples, tables, conformance rows, or any other host or monolith pattern section, as development, review, projection, or quality-status content about the pattern, lower CorpusEntryProjectionAndEcologyFit, EntityOfConcernPrimacyAndSemioBiasResistance, and the affected action or usability coordinates. |
CC-E21-7 | For any 5, name the reinforcing evidence loci required by that coordinate’s 5 meaning; otherwise lower the coordinate to 4 or below. |
CC-E21-8 | For MaturePatternParityAndSelectedContentSufficiency = 4 or 5, include a compact maturity-discharge payload: comparator id, selected ingredient, current locus, and missing or lowering item if any; category lists without loci cap the coordinate at 3. |
CC-E21-9 | Invoke the canonical definition and positive comparison contract in E.8:11 for every positive SoTA judgement; use F.1 only to inspect whether its source cut can support that comparison. For SoTABindingAndCurrentness = 4 or 5, include one complete E.8:11 comparison payload by value. A relevance/currentness table plus adopt/adapt/reject labels is below the ordinary floor when that comparison is absent. |
CC-E21-9a | Treat source identity and currentness as supporting traceability only. Official, popular, maintained, canonical, highly cited, recent, or academically praised status supplies zero positive evidence for bestKnownLine; a registry or publisher check cannot raise SoTABindingAndCurrentness. An official or widespread source can still fill bestKnownLine when its substantive answer independently wins the required comparison. An original FPF or DPF construction can also be selected under E.8:11; own authorship neither raises nor lowers its value, and a prior external publication with the same answer is not required. A value 5 additionally names the replayable comparison and reinforcing loci that make 4 too weak rather than adding bibliography, prevalence, or freshness. |
CC-E21-10 | Keep 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 defines, constrains, or tests the claim when the evaluated pattern makes those claims. |
CC-E21-11 | State floor satisfaction, next usable action, stop or repair, and lowering or reopen conditions. Add a non-use boundary only for an independently grounded reading that a plausible intended reader could make here. |
CC-E21-12 | Keep coordinate rationale separate from improvement proposal rows. |
CC-E21-13 | Keep quality results within their declared receiving use; any further evidence, assurance, gate, work, safety, compliance, release, or publication claim requires its direct relation. |
CC-E21-14 | Do not raise a pattern with a bad PrecisionRestorationProfile through related-pattern-boundary, projection, or quality-result 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 and usability coordinates; do not offset that loss with generic related-pattern-boundary praise or correct corpus projection evidence. |
CC-E21-15 | Keep ordinal values as ordinal content-evaluation result claims, not repair targets. Below-floor values require findings or repair. Values at or above the floor receive proposal rows only for concrete non-dominated content opportunities when improvement is requested; a non-5 value is not automatically a defect. No proposal may raise a value by adding quality proof, guards, relation catalogues, or process evidence that worsens use, affordability, locality, ecology, or the pattern’s own subject kind and positive action guidance. The aggregate no-proposal or stop disposition in E.21:4.7 must name checked loci and why no substantive content improvement remains. |
CC-E21-16 | When E.19:4.3.3 use-value replay evidence is current, evaluate the full existing coordinate set once and keep only materially affected uses whose outcomes can differ distinguishable during assessment. Carry into the durable result only replay distinctions that affect a coordinate rationale, cap, or status; do not create a second clean-use ledger. Apply the required-failure caps and repairBeforeUse effects in E.21:4.5.1; keep optional absence non-blocking by itself while denying unsupported breadth, transfer, or value-5 claims. Do not average outcomes, substitute an E.19 label for an ordinal value or status, compensate a failed required use with unrelated strengths, or infer values from a successful label alone. |