Six app-source comments are framed as 「§09 asks for X and we cannot deliver it」 — §09 stopped asking on 2026-09-09 #76

The defect

The 2026-09-09 maintainer ruling put 各段时长, per-type 「超 SLA」 and 各台阶平均停留 outside the V1.0 surface, and PR #40 (296fab8) wrote that into DESIGN.md §09. Six comments in src/ still describe the world as it was before that ruling — either quoting a §09 list that has since changed, or framing a limitation as a demand from §09 that the code cannot meet when §09 no longer makes the demand.

Their substance is correct. Every one of them explains something true about why a number is not delivered, and §09 now cites the same reasons. It is the framing that inverted: 「§09 要 X，我们做不到」 became false at the moment §09 stopped asking. That is why this is a reframe and ⛔ never a deletion.

What to change

Reframe each so it describes what the code delivers and why that is the right thing to deliver, citing §09 as the authority that says so — rather than as a demand being failed.

• Item 1 is the one with a concrete reader cost: a header that names a brick the file does not build. Prefer removing the quoted list and leaving a pointer, which is what PR Two app-source comments quote a DESIGN.md §09 that no longer says any of it — the stale-pointer pattern, pointing the other way #74 did for legal.dashboard.ts's header and for the same reason — a copy drifts, a pointer cannot. Read that header on main first and follow its shape.
• Items 2–6 keep every measurement and every technical reason (ADR-0021, the semantic layer taking no SQL or expressions, Field.datetime stored as ISO text so AVG() returns an average year, sys_approval_request being empty under the standard demo data so F5's dwell times have no source at all). ⛔ Do not delete any of it — §09 itself now carries these arguments, so the comments and the document should read as agreeing, not as one complaining about the other.
• Where a comment cites a platform tracker (objectstack-ai/objectstack#16737), keep it. ⚠️ And keep the point §09 makes and the comments should too: the platform's in-flight fix refuses the wrong path rather than providing date arithmetic, so durations remain uncomputable afterwards — they merely stop producing a plausible fake number.

Constraints

• ⛔ Comments only. No metadata key, no widget, no dataset field or dimension, no type, filter or options. If an edit touches one, stop.
• ⛔ Do not touch DESIGN.md — §09 is correct; this card brings the code's framing into line with it, never the reverse.
• ⛔ Do not re-promise an authored stage order anywhere (objectstack-ai/objectstack#17344), and ⛔ do not reopen the stage_funnel id (settled in PR Draw 「Pipeline by Stage」 as a ranked bar, not a funnel #57, reaffirmed in Plot 「Pipeline by Stage」 over the six in-flight stages, not seven #62).
• ⚠️ These files ARE inside the gates' surface — src/**/* is in tsconfig's include and in the metadata graph. ⛔ Do not copy the "no gate reads this file" framing from §06 F14 says an archived contract stays editable 「除 notes 外」, but clm_contract has no notes field — §03 gives it summary #44/docs: the position census disagrees with itself in three places — README says seven over a table of six, DESIGN.md §11 says both 7 and 8 #58/DESIGN.md §09 still specifies 各阶段合同数**漏斗** — the app stopped shipping a funnel in #48 and narrowed what it plots in #59 #63; it would be false. If you want an ablation, the honest one is PR Two app-source comments quote a DESIGN.md §09 that no longer says any of it — the stale-pointer pattern, pointing the other way #74's shape: prove a syntax error in one of these files is caught, and note that comments are invisible to tsc, which is exactly why this drift went unnoticed.

Acceptance

• All four gates 0, exit codes captured before any pipe — and say plainly that they do cover these files.
• Every claim in the rewritten comments traced to §09 as merged (main, post-30fd863) and to the code beside it, in a claim → reading table.
• Verify the edits landed by before/after spelling counts — ⛔ never by an editor's exit code.
• Re-run the sweep. PR Two app-source comments quote a DESIGN.md §09 that no longer says any of it — the stale-pointer pattern, pointing the other way #74 found these six by grepping src/ for comments quoting a DESIGN.md § list rather than pointing at one. Run it again on your tree and report what it returns, including anything the six above missed.
• No changeset (this repo has no changeset gate). No browser run owed.

Explicitly out of scope, decided and recorded

PR #74 also found six comments that quote §09 or §05 accurately today: finance.dashboard.ts:4–5 · dashboards/index.ts:1 · contract.dataset.ts:4 · obligation.dataset.ts:4 · payment.dataset.ts:4 · pages/contract_detail.page.ts:132. Each was re-checked and each still matches word for word.

⛔ Do not convert them. They are the same latent trap, but rewriting six correct comments is churn, and the argument for a pointer is strongest exactly where a copy has already lost the race. If one of them drifts, that is a card then. Named here so the decision is on the record and the next dev does not re-discover them as unfiled work.

Provenance

Found by the dev on #71 while running the sweep PR #74's card mandated, reported as out-of-scope with the stale/accurate split already drawn, and correctly not ridden into that PR — a reframe of six comments is not a rider on a two-comment card. The split and the reasoning above are the dev's; the decision to file the first group and decline the second is the PM seat's.

Related

#71 / PR #74 (where this was found; the same fix applied to legal.dashboard.ts's header) · #63 / PR #70 (§09's own correction) · PR #40 / 296fab8 (the ruling that inverted the framing) · #31 (pm:blocked on objectstack#16737) · #26 · #54