E.11:5 - Archetypal Grounding

E.11:5.1 - Architecture or working document?

A team says, “Our diagram no longer explains the system.” ARCHITECTURE and WORKING-DOCUMENTS both look plausible. The first card can return an architecture question, candidate set, or selected-structure result. The second can return the smallest description-use, representation, publication, or other working-document result for a named reader use.

The team compares those first results and sees that the selected structure itself is unsettled. It starts with ARCHITECTURE. No comparison account is needed because the comparison is local and reversible.

E.11:5.2 - A later safety review needs comparison history

The receiving safety review relies on an addressable rationale for why a team compared the ordinary MATHEMATICAL-MODELING entry, OPTION-COMPARISON, and SYSTEM-DELIMITATION with the ordinary SYSTEM-RECOGNITION entry before a hazardous test. New measurements will later reopen the choice. The four examples ask different first questions: what a model can support, which comparison or readiness result is needed, which parts or crossings matter, and whether the exact entity must be treated as a System at all.

That named reliance admits a PracticalUseEntryComparisonAccount@Context with four comparison rows, the stop boundary, and the return condition. The account does not authorize the test or replace evidence, assurance, gate, choice, or WorkPlan relations.

E.11:5.3 - A card leads to a physical result without promising it

WORKING-DOCUMENTS can lead to a usable machining work instruction. Its public guidance first asks what a named reader must decide, do, check, or rely on and returns the smallest truthful document-side result or blocker. Use the direct pattern to determine whether the useful result is a MethodDescription, WorkPlan, claim, permission, publication, or another document-side result and what relation makes it useful for the later project work.

The card does not promise a machined component. When actual machining later becomes current, use A.15.1 to identify the exact performed Work occurrence; use A.15 as well only when the broader System-Role–Method–Work Alignment question is current. The instruction or plan is neither that dated U.Work nor proof that it occurred.

E.11:5.4 - Repair the smallest card slice after a direct result changes

Suppose a new A.6.3.RT edition restores a progressive representation result: an ordinary target representation plus source-comparison note first, exact endpoints only when a named receiver makes their identity material, and a historical transition occurrence only when actual transition Work and all required participants are current. Repair only the affected WORKING-DOCUMENTS branch so its first result and escalation triggers match the direct pattern. The card heading and general question remain unchanged when readers still recognize the same situation.

After the repair, compare the same truthful entry with and without its mantra after a delay or interruption. Keep the card only if the mantra materially helps a cold reader reconstruct the choice-changing cross-pattern path and its first result or return. If the two forms perform alike, use an ordinary entry or locator; if the mantra hides the result or return, repair or drop it.

E.11:5.5 - A better first-click rate can make discovery worse

Suppose retrieval ranks WORKING-DOCUMENTS first whenever a diagram is mentioned and the first-click rate rises. Follow-up comparison shows that more readers now open document-use patterns while the architecture subject or selected structure is still unsettled, so first-result mismatch and wrong-turn returns also rise.

The visible navigation measure improved while the intended value worsened. Keep first-click rate as telemetry, apply E.13 to the substitution, and judge the guidance by recoverable situation fit, first-result fit, and wrong-turn cost rather than by the click measure alone.

E.11:5.6 - Discharge a duplicate first-entry row by function

Suppose a compact row combines architecture and diagrams, evidence, dashboard use, and alternative comparison. Do not keep it as another public example merely to display topic coverage. Put architecture design or review in ARCHITECTURE; put description, view, dashboard, and rendering use in WORKING-DOCUMENTS and the direct E.17/C.30.AD patterns; put costly evidence or commitment questions in OPTION-COMPARISON and their direct A.10/B.3/A.21 patterns; put useful search phrases in the ToC or retrieval. Use the worked comparisons in E.11:5.7–5.14 only if those cues still leave a genuine ambiguity. Delete the duplicate after every useful function has a matching home.

E.11:5.7 - Responsibility, method, plan and actual work are being mixed

A team says, “We keep mixing responsibility, method, plan and what actually happened.” First ask whether the difficulty is an alignment distinction or only a vocabulary mismatch. If someone is treating a plan as a completed run, agreeing on words alone will not resolve it.

Open the Solution of A.15 for the System–Role–Method–Work split. A.15.2 becomes relevant when the missing subject is a plan for future work; A.15.3 when a value is being planned for an existing declaration member. Inspect A.1.1 if the model-use boundary affects those claims, and B.5.1 if the question concerns coordination of Explore, Shape, Evidence and Operate states. These are conditioned neighboring questions, not a first-use chain.

If the alignment is already understood and only shared terms are missing, F.11 can distinguish method and work vocabulary, F.9 can resolve a cross-context sense Bridge, or F.17 can stabilize a first term sheet. Stop when the direct alignment body is opened or enough shared vocabulary lets the work proceed. Return to alignment if the term sheet still leaves a plan, role or run confused. Neither F.17 nor a rationale record under E.9 is a universal first stop.

E.11:5.8 - A meaningful phrase is not yet a claim

A reader encounters a phrase that matters but cannot yet say what it asserts. The choice-changing fact is whether it remains a cue to preserve or already has a stable claim structure. Treating the phrase as an endpoint claim would lose the uncertainty the next operation needs.

While it remains a cue, open A.16 to identify the next language-state move and A.16.1 for early preservation. Open A.16.2 when an existing publication overstates closure, stability or framing and must be reopened, backed off, respecified or retired. When a stabilized cue needs an open explanatory question, B.5.2.0 supplies the prompt form with scope and motivating grounds. Once alternatives are stable and their route plurality or selection must be published, B.4.1 supplies that different result. A U.Episteme publication requiring explicit language-state facets makes C.2.LS current.

If the material already contains an atomic boundary claim, open A.6.B or A.6.C for that structure. If it is being forced prematurely into a relational claim, quality endpoint or action invitation, defer A.6.P, C.16.Q, A.6.A or C.25 until the corresponding subject and basis exist. Stop at a preserved cue, an opened admissible next language-state move, or a reclassified boundary-claim question. A cue is not yet a finished requirement, work record, quality claim or invitation to act; reopen when the material actually acquires that structure.

E.11:5.9 - An API or contract sentence mixes several kinds of claim

A reader says, “The API or contract-language description says X.” First distinguish recognizing the encountered description from unpacking what its sentence claims. A description can mix an admissibility gate, duty, evidence claim, action invitation and access note; its familiar label does not settle any of them.

If the unresolved question is “what description is this?”, open A.6.RSIG. When recognition is sufficient and the sentence mixes boundary claims, open the Solution of A.6.B or A.6.C to recover atomic L/A/D/E-classified claims under A.6. Use A.6.P if relation wording hides participants or predicate, C.16.Q if quality wording hides its characteristic and scale, and A.6.A if an action invitation is the actual subject.

An already accepted engineering account may instead need reader-facing publication under E.17; E.17.0 answers view recognition, and E.24.PUB answers publication occurrence, form or carrier. Stop when the direct boundary-claim body is opened or the Claim Register or atomic claim set is ready for the next governing use. Return to description recognition if its defining episteme is still unclear. One API sentence supplies neither a complete Contract Bundle nor proof of downstream effect, performed action, quality or evidence.

E.11:5.10 - A shortlist, a pool policy or one winner?

A team says, “We need a shortlist, not one winner.” Ask which first result is missing: an admissible comparison, a pool of candidates, policy over that pool, a selected-set declaration, publication of the result, or one local decision. The fact that the current use concerns a governed set changes the choice; a one-winner doctrine is premature.

If comparison or aggregation is not yet admissible, inspect G.0, A.19.CN and the relevant A.17–A.19 bodies; A.19.CPM or A.19.SelectorMechanism becomes current when the comparator or selector structure itself is unsettled. For generation and archive/front stewardship while forming the pool, open C.18. For policy over still-live candidates, open C.19. G.5 supplies the selector-facing selected-set declaration, and E.24.PUB makes a declared result available.

If the actual question is one local choice, open C.11. If the action or option is already fixed and the missing result is a tool or service call plan or checkpoint return, use C.24, including for a call that performs an already chosen comparison. Stop at the correct direct body or a sufficiently discriminated candidate set; neither a shortlist nor the order of these examples prescribes a sequence. If several inspected pattern uses still need a recommendation or coordination, take that question to E.11.PUR.

E.11:5.11 - A reusable generator or SoTA scaffold is needed

A reader needs a reusable search, harvest or portfolio scaffold rather than one recommendation. The discriminating fact is reuse across searches, portfolios or updates. A compact cue is usually enough; an expanded comparison is useful only if the reader repeatedly mistakes scaffold construction for a local selector, comparison or publication.

Open G.1 to author a reusable CG-Frame generator and selector kit, or G.2 to harvest and synthesize SoTA for a CG-Frame. Inspect G.0 when the comparison frame is missing. Use A.0 when the missing result is the starter vocabulary for generative search. B.5.2.1 and C.17–C.19 are relevant when creative search, novelty or explore/exploit policy already changes that construction.

G.5 answers a selected-set declaration and E.24.PUB its availability; neither supplies the missing reusable scaffold. G.10 or G.11 becomes current when shipping or refresh is the actual next question. Stop when the scaffold/generator body is opened, or when the question has been correctly reclassified as a set-result or publication use. Reopen if the supposedly reusable account works only for the single current recommendation. A generator or SoTA concern creates no automatic tool-implementation or immediate-publication obligation.

E.11:5.12 - Explain or rewrite the same subject for another audience

A team needs to explain the same EntityOfConcern to another audience. First check whether the subject remains the same and only wording, representation or reading posture changes. Creating a second subject or rule track would change the work rather than explain it.

Open A.6.3.CR for same-entity retextualization, A.6.3.RT for a representation transition, E.17.EFP for explanation-facing rendering, or E.17.ID.CR for bounded comparative reading. Distinguish the result you need before choosing: new wording, a target representation, an explanation or a comparison. A pressured local wording repair can make E.17.AUD.LHR current; stability of a PublicationUnit can make E.17.AUD.OOTD current.

Stop when the same-entity rewrite body is opened or an explanation-facing rendering is stabilized with usable source pins. If comparison exposes a real change in the subject or claim, return to its direct semantic question instead of hiding it as presentation.

E.11:5.13 - A faster rate is predicted after a capacity change

A team says, “Adding review capacity for two sprints will double backlog reduction rate.” Inspect what the claim will be used to support. Ordinary prose, a snapshot, a measured rate and an intervention-sensitive temporal claim need different first results. Speed or rhythm wording alone does not select C.27.

Keep a snapshot as Dyn0 and a measured trend or rate as Dyn1. Open C.16 when measurement construction or comparability is missing. Use C.27 when the temporal claim changes admissible action because effort, window, resistance or cost, basis and reopening conditions matter. Its local Dyn2TemporalClaimAdequacyCard, or a boundary-crossing Dyn2TemporalClaimProfile, states that temporal-use question; it is not benchmark or causal proof.

Return a different question to its direct owner: A.3.3 for a reusable transition law or formal dynamics model; B.1.4 or B.1.6 when contextual/temporal or work-resource aggregation is the missing subject; G.9 for benchmark parity; C.25 for quality, first asking whether one Characteristic and Scale suffice and using a bundle only when differently typed contributors jointly determine the claim; C.26.3 for viability diagnosis and selected boundary regulation. C.24 addresses call planning for an already fixed action. A relied-on causal effect belongs to C.28, not to the fact that the rate changed.

Use C.26 only if an exact contextual-model obstruction still changes inference or action after the applicable ordinary subject assertion and predicate are in place. Stop at ordinary prose, Dyn0, Dyn1 with the needed measurement body, the relevant Dyn2 account, or a correctly named neighbor use. Reopen when the effort, window, resistance/cost, basis or intended action changes. Faster is not automatically better. A velocity target alone establishes no improvement or service promise; a dynamic benchmark needs a comparison basis before it supports superiority.

E.11:5.14 - A decision relies on a causal or counterfactual claim

A reader encounters “this policy would have prevented harm”, “the intervention caused the improvement”, “this fairness result is causal”, or “these simulated counterfactuals prove the decision”. Ask whether publication, choice, deployment, assurance, audit, benchmarking or method dispatch actually relies on causal support. Causal-looking wording alone is insufficient; the relied-on claim is the discriminating fact.

Open C.28 when the required support concerns an effect, intervention success, counterfactual comparison, causal fairness, policy optimality, causal evidence, off-policy/causal-RL evaluation or causal method superiority. Its Solution recovers the causal-use question, causality-ladder rung, claim kind, estimand, identification, counterfactual sampling realizability and support components, then distinguishes supported from unsupported use. A cheap downgrade sentence can be enough; a CausalUseTriageRecord, CausalUseEvidenceDesignRecord or CausalUseSupportResult is used only at the scope needed by the receiving claim.

If only a measured value is at issue, use C.16; if rate, trend or temporal adequacy is at issue, use C.27. If the question is method, work or work-plan structure, use A.15 and A.3.2; a sampling procedure, intervention recipe or target-trial phrase does not supply causal support by itself. Split a boundary duty or agreement with A.6. Use C.26 for residual QL modeling only when an exact contextual-model obstruction still changes inference or action after the ordinary subject assertion and predicate are in place.

A.10 governs the evidence/provenance paths actually relied on. B.3, D.5, G.5 and G.9 use a C.28 support result when the downstream question is respectively assurance, ethical/fairness audit, method dispatch or benchmark parity. Keep C.11, C.19 or C.24 as the direct local-choice, live-pool-policy or fixed-action call-planning use when that is the actual work; the support result does not decide it. Stop at the sufficient downgrade, scoped causal record/support result, or named neighboring use that cites its limits.

Return to causal support if the estimand, causal rung, identification, realizability or relied-on use changes. A randomized procedure is not automatically counterfactual support; simulation is not realized counterfactual data; a fairness metric is not causal fairness; and methods on different causal rungs or estimands are not made comparable by one benchmark. Causal support itself creates no duty, promise, commitment, release gate or admissibility predicate.