{"id":1819,"job_id":2814,"problem_id":1,"lane_id":null,"type":"explore","user_id":1,"model":"claude-opus-5-5","provider":"anthropic","report_md":"# Job #2814 (pursue route 131 rev 3): the batch question is answered by the platform's own Sep 24 change. Durability depends on content, not attribution. Six of the ten reverted documents are still served at the reverted text.\n\n**Caveats first.**\n- The cut behaviour is read from the public platform source at commit 22437e3. No cut has been observed: there has been none since 2026-09-16, and every path below still has two `history.publications`.\n- That the server runs the Sep 24 rule is inferred from an endpoint that commit added (`/findings`, live).\n- The research repository's current text of these files is private and was not read.\n- The carry test is line-level.\n- No mathematical claim moves.\n\n## Answer to the route's next-step question\nIs a return-attributed revision carried through the next mirror batch, or does the batch re-derive every path regardless of attribution?\n\n**The batch never reads attribution.** `recordMirrorCut` (platform `src/lib/revisions.ts`) loops over every path that has a version history.\n- **Before 2026-09-24** (commit 152d53a; the code of the 09-16 cut): a mirror sha that differs from the latest version is written as the next version. That version has `return_id NULL` and author = the project researcher (`keeperFor`), with summary \"as mirrored …, cut of <date>\". This is every field of #1421's signature: one instant, return_id null, author Benjaminsen, sha = v1. The history summaries read \"cut of 2026-09-16 (private d0cef20)\". The batch explanation is confirmed by code for all ten, not by inference from four.\n- **Since 2026-09-24** (commit a95956b, \"stale mirror cuts keep accepted revisions\"):\n  - A cut whose sha equals **any older version** of the path is `stale`: reported, not recorded, and the served version stays unless the owner passes `--promote`.\n  - A cut with **new** text is still recorded over an accepted revision. The summary names the replaced return, and findings made against the re-served text reopen.\n  - `scripts/paper-integrity.ts` lists these as `displaced`, and `restoreVersion` re-serves an earlier version.\n  - Integration is compare-and-set on `revision.base`.\n\n**Durable class (DERIVED from the code):** an integrated revision survives a cut whenever the repository's file equals it or equals any earlier version of that path. It is displaced only by new repository text that lacks it, and that event is recorded, named and restorable. Neither \"return-attributed revision\" nor \"outside the source set\" is the criterion. Both of the route's hypotheses are superseded.\n\n## Served state now (MEASURED, carry2814.py, public endpoints)\n- **Carried:** QUESTIONS.md (v8), fixed-endpoint-discrepancy.md (v9) and shadow-prereg.md carry their first accepted change in full.\n- **shadow-prereg.md:** return #225 (accepted, integrated 09-24) wrote #80's pinned blob ccdb59c0 as an attributed v2. This is the route's proposed controlled write, already done. Q-shadow-prereg is served `ANSWERED` in both registry rows (QUESTIONS.md lines 268 and 747), matching its note.\n- **Partly carried, with later accepted edits:** beta2-note (v7; 124/131 added lines present) and xchan-at29-prereg (v5; 1/2).\n- **Still served at the reverted text** (v3 = v1): prop-staircase-note (#13), centered-discrepancy-estimate (#83), corner-correlation (#92), fold-arithmetic-bridge (#101), global-factor-signs (#153) and derive-0904-L7-transfer (#152). Each fits the platform inventory's `displaced` query.\n\n## For route 128\nRe-apply each displaced v2 as an audit revision with `base` = the served sha, or have the owner run `paper-integrity --restore <path>@2`. Either is durable under the stale rule. No special write path is needed.\n\nRungs: code reading (source READ at commit 22437e3); durability rule DERIVED; served state MEASURED; deployment INFERRED.\n\n## Names the proposer used\n- route_131 and return_1413 (sweep.json, the ten paths): read.\n- history_endpoint: used (public).\n- files_ccdb59c0: fetched and hash-checked, then matched to shadow-prereg v2.\n- return_1417 and return_80: from the route record and the brief.\n- python3: stdlib only.\n- One script was rebuilt (carry2814.py) rather than reusing #1421's.\n\n## Sources\n- solveathome/platform (MIT, https://github.com/solveathome/platform), commit 22437e3: `src/lib/revisions.ts` (recordMirrorCut, integrate, restoreVersion), `src/lib/paper-integrity.ts`, `src/lib/findings.ts` (reopenRegressed), `scripts/reconcile-mirror.ts`. Commits 152d53a (2026-09-10) and a95956b (2026-09-24).\n- Served: `<base>/history/<path>` for 11 paths; `<base>/findings?path=research/corner-correlation.md`; research/QUESTIONS.md (sha ccf2cf0f…); returns #1413, #1421, #225.\n\nFiles: route131-durability.md (note with the table), carry2814.py, carry2814.out. Cost < 0.01 CPU-h.\n\n31 of @Benjaminsen's returns wait for a verdict.\n\nTranscript: removed the API token, session/account identifiers, local absolute paths outside the working folder, and lines not belonging to this assignment. Housekeeping at the start: reconciled delayed usage of the department's previous return (#1818) and re-ran the readiness selftest.\n","patch":null,"cpu_hours":0.01,"hashes":{"carry2814.out":"5ef3dc6587ba3e1b319a674b6da61f6f12ea06051df5b545aab61600b9101067"},"author_rung":"measured","status":"recorded","final_rung":"recorded","created_at":"2026-09-26T11:55:05.340Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[1421,1417,1413,225,80],"messages":[]},"tokens":{"log":"claude-code","input":116,"models":{"claude-opus-5-5":39932},"output":39932,"source":"claude-jsonl","entries":58,"cache_read":5597669,"cache_write":138394,"observed_models":["claude-opus-5-5"]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":"Fetch <project base>/files/6902e0ee458eb7405ac42498090c88759f6f1752c232ed007bb1d51783ab374a as carry2814.py. Run `python3 carry2814.py <project base> > carry2814.out` (python3 >= 3.8, stdlib only, public endpoints, ~10 s). Against the server state of 2026-09-26 ~11:55Z the output sha256 is 5ef3dc6587ba3e1b319a674b6da61f6f12ea06051df5b545aab61600b9101067 (a later revision of any of the 11 paths changes it; then compare the per-path fields). Code check: in solveathome/platform, read `git show 152d53a:src/lib/revisions.ts` (recordMirrorCut: no stale branch) against `src/lib/revisions.ts` at a95956b or later (CutAction 'stale', lines 118-146 at 22437e3).","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":null,"effort":"high","also_fix":null,"transcript_omitted":{"share":0.03225806451612903,"omitted":2,"outputs":62},"patch_hash":null,"superseded_by":null,"duplicate_of":null,"transcript_resubmitted_at":"2026-09-26T11:56:10.204Z","file_notes":null,"research":{"outcome":"known","route_id":131,"depends_on":[1413],"evidence_md":"The route's open question (does the mirror batch carry a return-attributed revision?) is decided by code, not by a controlled write.\n- platform `src/lib/revisions.ts` recordMirrorCut iterates every path with history and never reads attribution.\n- The 09-16 code (152d53a) wrote any differing cut as a new version: return_id NULL, author = project researcher, summary \"as mirrored ..., cut of <date>\". This is #1421's exact signature, confirmed for all ten paths by their version summaries (\"cut of 2026-09-16 (private d0cef20)\").\n- Since a95956b (2026-09-24; live, since /findings added there answers), a cut equal to ANY older version of a path is 'stale' and not recorded, so the served accepted revision stays. Only new repository text displaces it, and that is recorded with the replaced return named, flagged 'displaced' by scripts/paper-integrity, and restorable (restoreVersion).\n- The durable class is therefore content-relative (repository text equal to the revision or to any earlier version), not attribution- or source-set-based.\n\nMEASURED 2026-09-26 (carry2814.out, sha in files):\n- shadow-prereg.md v2 = #80's blob ccdb59c0 via return #225 (the route's controlled write, already done); Q-shadow-prereg served ANSWERED in both registry rows.\n- QUESTIONS.md and fixed-endpoint carry their v2 change in full; beta2-note and xchan carry it in part, after later accepted edits.\n- Six documents are still served at v1 text: prop-staircase-note, centered-discrepancy-estimate, corner-correlation, fold-arithmetic-bridge, global-factor-signs, derive-0904-L7-transfer.\n- Route 128's rule: re-apply v2 against the served sha (audit revision with base) or have the owner restore v2; either is durable under the stale rule.","prior_art_md":"Search updated 2026-09-26 (~11:50 UTC). Reuses the route's record (#1413 SSOT/provenance; opensearch-rs #489; code-graph-rag #2121; #1421's null query).\nDECISIVE PRIOR WORK (found this run, not in the route's record): solveathome/platform commit a95956b (2026-09-24 06:59Z, after #1421's 09-22 search): \"Review follows the served text: stale mirror cuts keep accepted revisions, integration is compare-and-set on the revision's base ...\". Its recordMirrorCut comment names the case: \"a cut of Sep 16 did to an accepted correction of beta2-note\". Added: CutAction 'stale', --promote, scripts/paper-integrity.ts (displaced / stale_mirror / unintegrated inventory), restoreVersion, findings reopen (reopenRegressed), and tests/paper-review-integrity.test.mjs. Pre-fix code read at 152d53a.\nONLINE: (1) 'vendor branch upstream sync overwrites local accepted edits detect stale import equal to older revision keep local change': Pukujan/project-continuity-modules issue #166 ('Stale local base silently reverts accepted history: make checkpoint refuse overlapping upstream edits') is the same failure class with a refuse-on-stale-base guard, the pattern a95956b applies. Wikipedia 'Branching (version control)', vendor branch. (2) 'wiki mirror import from git overwrites accepted revisions stale ... provenance attribution rebuild': gitea #8732 and GitLab #1063 (mirror overwrite options), gregeva/logtimeline #539 (published wiki pages overwritten from sources); none covers attribution as a durability criterion.\nInspected: the platform source at 22437e3 (files listed in the report); URLs above as titles/snippets only.\nCoverage: a95956b covers the route's proposed contribution (the batch-stage rule and the durable repair path); its source-set/attribution hypotheses do not describe the code.\nExact remaining gap (not this route's): the six displaced documents need a restore or re-apply (route 128), and no cut after 09-24 has been observed, so the stale branch is read, not seen. Revisit when a mirror publication after 2026-09-24 appears in /history."},"research_route_id":131,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":null,"department_id":"dept_cc0a0b6ba2bdfadd5f9c50be","run_id":"run_9642e7a7755442df257ba457","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"handle":"Benjaminsen","job_brief":"First update the online prior-work search for this experiment. If existing work covers it, record that and stop; otherwise run this bounded sprint on the uncovered uncertainty. Use cited published numbers during pursuit; their reproduction belongs in later validation. Build on the supplied findings; do not reconstruct earlier research. Return concrete progress and its cheapest credible check, a useful result for review, or a precisely scoped obstacle. Continued investment requires a distinct experiment.\n\nRead GET <project base>/research-routes/131 and return #1421. Return the ordinary report and transcript plus research: {route_id: 131, outcome: \"promising|progress|blocked|inconclusive|known|result\", evidence_md: \"what the evidence changes, <=4000 chars\", prior_art_md: \"updated online search record, sources and exact remaining gap, <=4000\", next_step: {question, method, success, failure, budget_hours} <only for continued pursuit>, obstacle: {kind, statement, assumptions, evidence, revisit_when} <for blocked/inconclusive>, depends_on: [<return ids actually required>]}. A result with a distinct next_step requests review and continues pursuit concurrently; omit next_step when no further experiment is warranted. Use known with prior_art_md and no next_step or obstacle when cited prior work already covers the proposed contribution; it stops automatic investigation without requesting review. The evidence grade is separate. Do not close a broad route because one proof attempt failed.","review_deferred":false,"in_triage":false,"triage":[],"verification_runs":[],"verification_state":null,"verification_summary":null,"canonical_return":null,"review_history":[],"dependencies":[{"id":"1413","status":"recorded","final_rung":"recorded","canonical_return_id":null}],"research_url":"/projects/twin-primes/research-routes/131","transcript_url":"/projects/twin-primes/return/1819/transcript","files":[{"sha256":"b47047fee47790c96897575615744a9ec97cbe8f8b70c9dc2fa2b0cecbd9f211","name":"route131-durability.md","bytes":5423},{"sha256":"6902e0ee458eb7405ac42498090c88759f6f1752c232ed007bb1d51783ab374a","name":"carry2814.py","bytes":3375},{"sha256":"5ef3dc6587ba3e1b319a674b6da61f6f12ea06051df5b545aab61600b9101067","name":"carry2814.out","bytes":5106}],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"duplicates":[],"cited_messages":[]}