Skip to content

Prove one-scale zero-swap helper-462 words need at least 9 shears. - #238

Draft
unpaidmoon wants to merge 1 commit into
cursor/upstream-17469fff-pin-78d7from
cursor/helper462-one-scale-rank-drop-78d7
Draft

unpaidmoon wants to merge 1 commit into
cursor/upstream-17469fff-pin-78d7from
cursor/helper462-one-scale-rank-drop-78d7

Conversation

@unpaidmoon

Copy link
Copy Markdown
Owner

Change and mathematical scope

Proof. No new κ. The selected record stays 781451/250000000.

Under the zero-swap, one-scale hypotheses of open PR CrocSwap#7030, every rational elementary word for the pinned helper-462 matcher has at least 9 shears. Subadditivity of rank(W - I) still gives only 8, and that rank is 8 on the matrix, so the stronger bound is not the same invariant. The new invariant is the elementary rank-drop depth. On the recorded 29-shear word it equals 2. The same exhaustive census on every scale row has depth at most 2, so a factorization into 8 elementary shears does not exist.

Twenty-nine remains an upper bound because that recorded word still multiplies to the matcher. Nine is not shown to be sharp. Lengths 9 through 28 stay open. This census is finite only because each rank-dropping edge has at most one coefficient. It is not a census of words that contain a non-dropping shear.

No supplier admission, no P2 ordinary bill, and no integer-coefficient claim. The integer obstruction remains the theorem of CrocSwap#6954.

Task and overlap

Base: 17469fff78c22941280f04d39dc5a1faee09c840, priority revision 8. The verifier reads certificates/selected-result.json; current-priorities.md on this base still quotes the older selected value from CrocSwap#5599.

Task card: supplier-boundary-audit, scope refined. The question is open question 2 of open PR CrocSwap#7030: whether the rank lower bound of 8 can be raised under the same hypotheses. The requested first step was to name a different invariant and evaluate it on the recorded 29-shear word.

Nearest work, and why this is distinct:

Package: research/helper462-one-scale-rank-drop.

Validation

Python 3.12, standard library only, from the repository root:

python3 -B research/helper462-one-scale-rank-drop/verify.py
python3 -I -B research/helper462-one-scale-rank-drop/verify.py
python3 -O -B research/helper462-one-scale-rank-drop/verify.py
python3 -B scripts/research_package.py --package research/helper462-one-scale-rank-drop

Proof controls use an explicit failure function, so they stay active under -O. Expected token: PASS_HELPER462_ONE_SCALE_RANK_DROP.

Checked here: selected κ is unchanged; the published 24 right-route factors multiply to S; the recorded word multiplies to S; moving its scale and returning it preserves 29 shears and the product; mutating the scale, the first shear, or dropping the first shear misses S; one shear has rank distance 1; an explicit 8-shear matrix in SL(8) peels back to I; diag(2, 1/2) has rank distance 2 and depth 0; for every scale row the minor test and the pairing test agree on every edge, the depth is the pinned value at most 2, and the tree does not reach I.

No Lean, Docker, or selected cold replay was run.

Credit

The matcher and the 24 right-route factors are from research/bit-completed-backend-api (Chris R / Clear1nsight, PR CrocSwap#4957), using the helper-462 witness of ai-nanachi CrocSwap#4760/CrocSwap#4814. The determinant identity and the balanced-scale identity are PR CrocSwap#6954. The recorded 29-shear word, the scale-moving lemma, and the rank lower bound of 8 are PR CrocSwap#7030. Prepared by unpaidmoon with substantial Cursor Grok assistance. Apache-2.0.

Open next questions

  1. Can the elementary shear lower bound be raised above 9 under the same hypotheses? A word of length 9 has one shear that does not drop rank(· - I), and that coefficient is not fixed by the rank-drop equation, so the same census does not decide it. The useful first step is to name an invariant that charges one non-dropping shear and evaluate it on this same recorded word. An explicit word shorter than 29 is open question 1 of Record a 29-shear one-scale zero-swap word for helper-462. CrocSwap/integer-mult-bounds#7030 and is not taken here.

Workflow feedback (optional)

Requiring the minor test and the pairing test to agree made a false rejection of a legal edge fail the run, instead of turning into a spurious lower bound.

Open in Web Open in Cursor 

The elementary rank-drop depth of D^{-1} S is at most 2 on every scale row, while rank(D^{-1} S - I) = 8. Length 8 is therefore impossible under the zero-swap, one-scale hypotheses. The recorded 29-shear word remains an upper bound. No new kappa.

Co-authored-by: unpaidmoon <unpaidmoon@users.noreply.github.com>
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