Skip to content

WIP: guarded seven-file source-repair draft for SM Theorem 3 - #966

Closed
amaybaum wants to merge 1 commit into
corpus/sc17-explainer-bellfrom
draft/sm-theorem3-source-repair-20261027
Closed

amaybaum wants to merge 1 commit into
corpus/sc17-explainer-bellfrom
draft/sm-theorem3-source-repair-20261027

Conversation

@amaybaum

Copy link
Copy Markdown
Owner

Scope and pins

Head: f522e62f8f02fbaec6eeecbd5a1316683470c547.

This is a concrete source-repair draft, stacked on SC-17 #963 at 81654257152d0efd2101b183aacef2e169959525 because its Explainer and book edits share the affected paragraphs. It leaves the mathematical review object #965 at d2e01a3a2670dab1a0f613d2dcfb7193e2608e1a unchanged.

The commit adds one patch generator, not applied manuscript edits. It contains exact old/new replacements for seven files and refuses to emit a patch unless all pinned full-file blobs, index entries, working copies and unique replacement contexts match. It never writes repository files or runs builds. A full checkout could not be fetched in this writer environment, so no full-file application, build or corpus closure is claimed.

Proposed changes

  • SM §4.3: replace the unrestricted microscopic equivalence with the fixed-staggered-operator result, prove the on-site converse and scalar-coefficient corollary separately, and state their distinct hypotheses. Preserve the Laplacian-diagonal remark and the independent memory/readback results.
  • ch05 and its FULL mirror: replace the second-order-center-term argument with the actual first-order on-site statement.
  • Explainer: scope the chirality sentence to the fixed first-order branch; do not silently rewrite the separate Higgs claim.
  • Lean roadmap: distinguish the existing identity from unformalized converse/coefficient uniqueness.
  • OI_Structural_Core.lean: comment-only removal of the unsupported coefficient-uniqueness gloss; line count preserved, no theorem/proof edit.
  • structural_chain_probe.py: describe the separate mass and scalar-offset tests actually executed, not a universal converse. Numeric assertions and other computation must remain AST-identical after excluding top-level documentation and print calls.

Checks actually performed

source_repair.py --self-test: 36 fragment/guard checks PASS, covering successful replacement and refusal of missing, duplicated or already-applied contexts. Normal and optimized Python outputs are byte-identical; Python compilation succeeds. These are patch-mechanics tests, not scientific evidence or a full-checkout test.

Script SHA-256: 790276035135b949ab1a3eac838d97cd6e1e0dcebf7778d2717c2bb8db713d3b.
Published Git blob 852857628422e33d48de65e30d26c59ccfd1e79c equals the locally tested file.

Remaining work / no closure

  1. Independently review WIP: SM Theorem 3 restricted converse — proof, exact controls, and scope repair #965's mathematics and the source-repair scope.
  2. Meet the existing AGENTS §A.34 requirement for separate directional kernel witnesses before publishing the new equivalence. The paper proof is not a kernel certificate. This does not create an unrelated Introduction paper: reconcile sources and integrate scoped theorem recoveries #918 blocker.
  3. Generate/apply the patch in a complete pinned checkout; inspect the diff and finish the whole-corpus propagation audit, including relevant Main/Substratum/Structure and book summaries. Search hits alone are not a completed dependency audit.
  4. Reconcile coverage, registry anchors and any guards; regenerate affected SM/Explainer/book artifacts in the same source-edit commit, or explicitly mark them stale until rebuilt. Do not leave stale distributed artifacts unflagged.
  5. Run the required checks and obtain independent acceptance on the actual source-edit SHA.

No manuscript status, coverage tier, review gate, retirement decision or merge authorization changes. The source finding remains open. The agreed #918-first sequence and explicit exact-SHA approval remain in force; this stacked draft must be reconciled after its actual dependencies land.

Related: #965, #963, #946, #918.

Keep #965 at its review pin. This adds a patch generator, not applied
manuscript edits. Full source blobs are pinned; altered/missing/duplicate
contexts refuse before any patch is emitted. The generator never writes
the checkout. Local 36 fragment/guard tests pass normally and under -O
with identical output; Python compilation passes. Full-checkout application,
corpus-wide propagation, builds, kernel witnesses and independent review
remain outstanding. No merge or scientific-status upgrade.
amaybaum pushed a commit that referenced this pull request Oct 27, 2026
Applies the independent review of 97d5b37 (accept with changes).

- Family qualifier (blocking). F = K T0^-1 Pi_- + T0 Pi_+ is local,
  anticommutes with eps and squares to K for every (C, alpha), so
  chirality constrains C only within Lemma 3a's family. The alpha = 1
  center-free mass statement now carries "within Lemma 3a's family" and
  "for d >= 2" in the SM Remark, the ROADMAP, the track README and the
  script verdict. New check G6 verifies F exactly at d = 2 and 3.
- Proposition 3b states the family hypothesis, lambda != 0,
  characteristic 0 and even extents >= 4.
- Lemma 3a: extents >= 4 for the coefficient identification.
- Theorem 3: C = +-m^2 with a0 = 1.
- Remark: "this lemma" is named as Lemma 3a or Theorem 3.
- SM §4.2, "the two roles of the dynamics": the Susskind factorization
  is massless at alpha = 1/d and massive at the unit coefficient within
  the family. Added to the dependents table.
- "Scope in the chain": the closing sentence becomes a plain status
  statement with the same content, which SC-17's ch05 :95 cites.

The ledger re-affirms SM:L-3a, SM:T-3 and SM:P-3b at their new
fingerprints, with levels unchanged. The track output is regenerated
and byte-identical under -O. SM .tex/.pdf rebuilt. The two #966 fixes
are listed as pending the owner's route decision and are not applied.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Y87wExth9RoUb3dtbumhTb
amaybaum pushed a commit that referenced this pull request Oct 27, 2026
Owner decision (owner, 2026-10-27): SC-32's theorem correction is the
route; #965 is credited as the first step of Proposition 3b; #966's two
documentation fixes are adopted; #966's SM replacement is not adopted.
SM:T-3 is downgraded K2 -> K1, and no upgrade is made without a
corresponding machine-checked theorem.

- OI_Structural_Core.lean:413, center_anticommutator docstring: "zero
  iff the center terms vanish" -> "an identity, not coefficient
  uniqueness" (#966's text). Comment only: with comments and docstrings
  stripped the code is byte-identical, and the file has 551 lines before
  and after.
- structural_chain_probe.py, C3 docstring and print, take #966's text.
  The AST is identical excluding the module docstring and print calls.
  The output under python3 -I differs only in the C3 print line.
- Track README: records the owner decisions with author and time, and
  the precise proof coverage. Staggered.factorization covers unit
  coefficients and abstract generators. center_anticommutator is an
  identity with {D, eps} = 0 as a hypothesis. SM:L-3a and SM:P-3b have
  no kernel witness of the statement. The fixes are credited to #966.
- LEDGER.json: the SM:T-3 note states the same coverage and the owner
  decision. Level K1 and the ID are kept, and no other row changes.

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

Copy link
Copy Markdown
Owner Author

Closing without merge: the owner chose SM Theorem 3 Route A (Workstream D audit on #946; owner decision of 2026-10-31).

This PR's generator pins an old base and refuses to run on current main. Here is where each of its intended changes stands:

  • Absorbed on main: the first-order SM Theorem 3 statement and the M3.1 docstring.
  • Carried into draft Workstream D: consolidated minimal source corrections #1019, commit 9a6a2f18 (Route A):
    • the scope sentence and the separate-bridge qualification (SM:302–308);
    • verification/lean/ROADMAP.md:193-196 and structural_chain_probe.py (no longer "⟺");
    • the M3.2 docstring in OI_Structural_Core.lean;
    • ch05:95/FULL:1422;
    • Explainer:826.
  • Obsolete as written: its book and Explainer hunks targeted SC-17 text that is not on main. The underlying defect is handled in Workstream D: consolidated minimal source corrections #1019 (D-3/D-4).

The branch and history are kept as provenance.


Generated by Claude Code

@amaybaum amaybaum closed this Oct 31, 2026
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.

1 participant