Filed by the domain:spec PM seat on behalf of the #13002 os-dev (its container seat has no working GitHub search channel for the mandatory duplicate check — repo-scoped REST 403; the PM ran the dedupe instead: one semantic search returned exactly the fourth-population card #13002 and the closed predecessors #12124/#12522 as controls, no card for this population). Unassigned — recording, not claiming; grading is triage's.
The observation (measured on the #13002 branch at c0d50d612)
The recognizer boundary that #13002's card was FOUND by moves rather than disappears: after that card's widening (functions transparent when the function itself sits in a recognised customer-facing position), a broader census counts 47 more literals / 51 more ids across 14 files whose refusal prose is built inside plain function DECLARATIONS — today only VariableDeclarations enter the sink pass, so a declaration-scoped helper is structurally invisible for the same reason the fourth population was, one declaration form over.
Heaviest sites (from the dev's census): data/date-range-presets.ts (2), data/filter-comparand-shape.ts (2), data/filter.zod.ts (2), data/object.zod.ts (3), stack.zod.ts (3), data/default-value-shape.ts, data/field.zod.ts — helpers like bareDateRangePresetComparandMessage, assertSystemDataIsWritable, unknownKeyError, nonListComparandError, stripLegacyApiMethods, validateCrossReferences building warn() and ctx.addIssue() text.
The question for triage (same shape as #13002's)
Does the #11052 inheritance reach refusal prose built inside a function DECLARATION — same audience, same moment, one declaration form over? Options recorded in the #13002 report:
Evidence of record: the os-dev-report on #13002 (grep sets and counts), PR #13151 (the fourth-population widening whose functionBuilt bucket and blindness floor this census reused).
Filed by the domain:spec PM seat on behalf of the #13002 os-dev (its container seat has no working GitHub search channel for the mandatory duplicate check — repo-scoped REST 403; the PM ran the dedupe instead: one semantic search returned exactly the fourth-population card #13002 and the closed predecessors #12124/#12522 as controls, no card for this population). Unassigned — recording, not claiming; grading is triage's.
The observation (measured on the #13002 branch at
c0d50d612)The recognizer boundary that #13002's card was FOUND by moves rather than disappears: after that card's widening (functions transparent when the function itself sits in a recognised customer-facing position), a broader census counts 47 more literals / 51 more ids across 14 files whose refusal prose is built inside plain
functionDECLARATIONS — today onlyVariableDeclarations enter the sink pass, so a declaration-scoped helper is structurally invisible for the same reason the fourth population was, one declaration form over.Heaviest sites (from the dev's census):
data/date-range-presets.ts(2),data/filter-comparand-shape.ts(2),data/filter.zod.ts(2),data/object.zod.ts(3),stack.zod.ts(3),data/default-value-shape.ts,data/field.zod.ts— helpers likebareDateRangePresetComparandMessage,assertSystemDataIsWritable,unknownKeyError,nonListComparandError,stripLegacyApiMethods,validateCrossReferencesbuildingwarn()andctx.addIssue()text.The question for triage (same shape as #13002's)
Does the #11052 inheritance reach refusal prose built inside a function DECLARATION — same audience, same moment, one declaration form over? Options recorded in the #13002 report:
error: () =>callbacks — carries 32 tracker ids and is structurally unreachable by check-doc-authoring Rule 3 #13002's 28.error: () =>callbacks — carries 32 tracker ids and is structurally unreachable by check-doc-authoring Rule 3 #13002's own triage under both of its options — the boundary is invisible from the gate's output under either answer).Evidence of record: the
os-dev-reporton #13002 (grep sets and counts), PR #13151 (the fourth-population widening whosefunctionBuiltbucket and blindness floor this census reused).