Skip to content

SUB-241B-OBS: native V3 round taking up Lemma 24.1B (control plane, candidate F) - #987

Closed
amaybaum wants to merge 3 commits into
mainfrom
v3/sub-241b-obs
Closed

amaybaum wants to merge 3 commits into
mainfrom
v3/sub-241b-obs

Conversation

@amaybaum

@amaybaum amaybaum commented Oct 28, 2026 •

Copy link
Copy Markdown
Owner

This is the guiding draft PR for native V3 round SUB-241B-OBS (AGENTS.md §A.39). The planning issue is #985 and the sibling round is #986.

State. This is a candidate control plane at a926233afc1acf4d5975942bc4f2f9029337711a. It is the third of three single-parent commits from D = 0d917665d32287e5d0ce649012175f2c7bf75b09 (main; verify run 40535620231 green): db19ccbf, 02c77a11, a926233a.

  • F is not designated.
  • No workflow_dispatch has been run.
  • Independent review is pending.

The owner approved drafting on 2026-10-28. F designation, the exact-head dispatch at F and E, and any merge each need separate owner authorization. The owner's standing rule is that this PR is not merged into main.

Content. One file: verification/programmes/substratum/round-sub-241b-observables-v3/preregistration.md.

  • v3 blocks:
    • round SUB-241B-OBS, kind non-sealing;
    • record directory and receipt verification/receipts/SUB-241B-OBS.json (AM);
    • execution M: verification/ROADMAP.md and verification/README.md.
  • Provenance. The legacy freeze lemma-24-1b-framework-data/preregistration.md (blob f614b666, execution base 445701c7, Lemma 24.1B control plane: does the framework's hypothesis fix the block-word traces? #631) is quoted verbatim and adopted by citation. The legacy directory is byte-for-byte untouched.
  • Evidence base. The nine frozen pinned blobs at 445701c7. SC-44 (SC-44 (draft): retire false semigroup-transfer lemma; state verified block-word rigidity #980), SC-45 (SC-45 (draft research): reproducible fixed-interface counterexample to Theorem 24(i) #981) and the current manuscripts are not substituted. Drift at D is recorded only as a start-state note, and a separate non-verdict section records how the findings bear on the corrected theory.
  • Recorded discrepancies.
    • X1: WD1's four statements versus "three".
    • X2–X5: the result-note sections, the §A.37 mechanics, the base wording and the shortened pin paths.
    • Readings R1–R6.
  • Independence. Checks N4 and N14 enforce independence from SUB-241C-SPAN and from any 24.1C object.
  • Round-local controls. The controls text is frozen inside the preregistration and pinned by blob, following BG-2. The file itself is committed at stage 1, after F.

Local checks at the head:

  • v3_verifier self-test, corpus and --receipts: OK.
  • v3_receipt self-test: OK.
  • legacy_records_check: 303 records intact.
  • Release gate: every step except lean-axioms, which needs the Lean toolchain.
  • Mock execution (no refs) from candidate F:
    • checks T1–T7 and N1–N17 pass;
    • the countercontrol fails at T6;
    • a mock reconciliation and receipt give --verify-round → HOLDS.

These local runs are design evidence, not check-run attestations.

Pushing. The push started no governed workflow. Of the path-filtered workflows that can fire on any branch, the two that did are read-only checks.


Generated by Claude Code


Generated by Claude Code

Native V3 round taking up the frozen, never-executed Lemma 24.1B round
(legacy blob f614b66, merged by #631 as 445701c). Planning issue #985.

Adds the preregistration only: the v3-round and v3-governed-paths
blocks, the legacy freeze quoted verbatim and adopted unamended, the
evidence base (the nine pins located at 445701c), the start-state
drift note, independence from SUB-241C-SPAN, the recorded discrepancies
and readings, the non-verdict section's limits, the frozen result-note
form, the frozen controls.py text and blob, the invariant/checkpoint
table, stages and outcomes, and the preconditions of F.

F is not designated; nothing is executed.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y87wExth9RoUb3dtbumhTb
…e F, WIP)

Applies the review of db19ccb (REVISE), control-plane only:
- B1: record that the pull request is not merged into main and its
  section A.42 consequence in The protocol; the records-hold invariant
  row and non-verdict item 3 follow it; decisions 5 and 6 cited from
  the #985 comment 6463778359.
- B3(i): T5 admits no file in the record directory beyond the
  control-plane files at F, result.md and controls.py.
- B3(ii): one reviewer-checked invariant row for the freeze's
  remaining requirements; R6 states strength per determination.
- B4: T7 requires exactly the three frozen spans, bodies equal.
- O1-O3, O6-O8: WD3 token mutation, N14 scans blocks, N16 ignores list
  and quote markers, start-state act 14 note, HALTED record commit.
Controls blob 63e9e9c2 -> ddf80e63; self-test 53 -> 59 cases.
B2 is held for the owner's choice. F is not designated.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y87wExth9RoUb3dtbumhTb
…ndidate F, WIP)

Applies the review of 02c77a1 (ACCEPT WITH CHANGES), control-plane only:
- C1: the governed-path, invariant and drift-note requirements are this
  control plane's own, not decisions quoted from #985; the encoding
  table says which rows are which, and the drift note moves to row 7.
- C2: the reviewer-checked row names the WD2/WD3 promotion rule at the
  freeze's :276-278.
No change to the controls (blob ddf80e63). B2 and O5 are held for the
owner. F is not designated.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y87wExth9RoUb3dtbumhTb

amaybaum commented Nov 1, 2026

Copy link
Copy Markdown
Owner Author

Candidate-F state (read 2026-11-01). F is not designated, and no workflow_dispatch has run on v3/sub-241b-obs. The head is a926233a, the third of three single-parent commits from D = 0d917665, each titled "candidate F, WIP". No release of this candidate state is recorded here.

Owner decision 17 retires this PR to #985 under §A.43 item 6. That needs the owner's explicit release of the candidate state, recorded on this PR, and the preservation checks listed on #985. This comment neither releases the candidate state nor closes the PR.


Generated by Claude Code

amaybaum commented Nov 1, 2026

Copy link
Copy Markdown
Owner Author

Owner decision recorded: candidate-F release for retirement

The owner (amaybaum) gave this decision in the coordinating session on 2026-11-01, just before 17:35Z. It is quoted verbatim:

I release the candidate-F administrative reservation on SUB-241B-OBS at a926233a for retirement to issue #985. No F designation, scientific execution or merge is authorized. Preserve the branch, commits, frozen sources and dependent evidence. Closure without merge is permitted only after the preservation checks pass.

The owner also stated: "These decisions release the administrative reservations. They do not authorize automatic closure or change the other research holds."


Generated by Claude Code

amaybaum commented Nov 1, 2026

Copy link
Copy Markdown
Owner Author

Preservation checks for retirement to #985 (read 2026-11-01, read-only). One check fails, so closure is not yet permitted.

The preservation results were independently re-checked: both verdicts confirmed.


Generated by Claude Code

amaybaum commented Nov 1, 2026

Copy link
Copy Markdown
Owner Author

Closing without merge, on the owner's decision (amaybaum, 2026-11-01): "Approve fallback. Record B2, O5 and O4 as unrecovered review obligations on #985. Any restart must independently review the entire control plane. After an independent preservation re-check, #987 may close without merge."


Generated by Claude Code

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants