diff --git a/research/pr6551-retirement-audit/ACCOUNTING.md b/research/pr6551-retirement-audit/ACCOUNTING.md new file mode 100644 index 0000000000..db2ef7ec05 --- /dev/null +++ b/research/pr6551-retirement-audit/ACCOUNTING.md @@ -0,0 +1,83 @@ +# Independent accounting of the retirement tree rebalance + +At candidate PR6551 head `11ca568cd1a16776ade02b4df09ea04a784cc674`, relative to predecessor PR6545 head `775722aa94a5e85b05d1d704d7dca10cbf016f0c`, direct counts of the literal callers give + +``` +ΔT = −104 H16 + 104 H40 + 208 H41 − 208 H57, +ΔP = −104 H16 + 312 H40 − 208 H56, +Δ(T−P) = 208[(H41−H40) − (H57−H56)]. +``` + +The caller counts agree with the predecessor's claimed change. The checker does not import or execute either author's invoice, pricing, constructor, native or caller code, and does not read their RESULT.json files. The first direct event/endpoint count was derived before reading the candidate proof's complete signed invoice. The written physical semantics and literal data are shared inputs; this is independent accounting implementation, not independent derivation of the native tariff. + +## Two representations and the exact count + +For one port, a declared endpoint `q^a ∏ D_s` has positive Gaussian rank `|a|+8·#{s}` in this caller. There are no stage-zero targets. The central exponent −1 is normalized by an ordinary q² translation; zero means no central axis. Each binary whole IDLE pays for two jobs of rank eight times its stage-set cardinality, IDLE_ONE pays one, and QUERY pays one rank-one job. Each LOCAL expands into the shared native subprogram. + +The checker first traverses every literal child tree, assigning the eighteen fresh stages to leaves. It obtains internal-node sizes `{2:8,3:2,4:2,7:2}` before and `{2:8,3:2,4:2,5:1,7:1}` after. The entire root outside `children` and `child_sizes` is identical. The first two children retain sizes seven and two; C,D change from two,seven to five,four. The masks AB and CD still contain nine stages. + +The root-mask count, the two productive outputs of every internal node, and the independent retirement split give an expected histogram. It must equal a separate count of all 35 serialized endpoint rows. The checker also verifies the actual retired field 4 and fresh suffix field 34, so exchanging their targets cannot hide behind an unchanged histogram. + +Separately, all literal events are classified. The fourteen binary IDLE stage sets must equal the actual tree-node stage sets, including multiplicity. The two IDLE_ONE stage sets must equal children B and C. All seventeen binary/root thin queries and all eighteen ordered LOCAL stage calls must be present. Unrecognized event kinds fail. Q_POWER exponents must be ±2, the inherited ordinary translation case. + +Per port, excluding the identical native contribution: + +| Rank | Old productive | New productive | Old paid | New paid | +|---|---:|---:|---:|---:| +| 1 | 2 | 2 | 17 | 17 | +| 16 | 1 | 0 | 18 | 17 | +| 17 | 17 | 17 | 0 | 0 | +| 24 | 0 | 0 | 4 | 4 | +| 25 | 4 | 4 | 0 | 0 | +| 32 | 0 | 0 | 4 | 4 | +| 33 | 4 | 4 | 0 | 0 | +| 40 | 0 | 1 | 0 | 3 | +| 41 | 0 | 2 | 0 | 0 | +| 56 | 0 | 0 | 4 | 2 | +| 57 | 4 | 2 | 0 | 0 | +| 73 | 2 | 2 | 0 | 0 | +| 145 | 1 | 1 | 0 | 0 | + +The rank-40 paid coefficient three consists of two binary-node idles and the third child's additional root idle. Dropping that last job would produce a false improvement. + +The exact retained native archive contains 104 distinct ports, 770 public fields (562 productive helpers plus 208 main fields), and 3,456 MOVE events. The candidate source archive contains a native archive whose SHA256 matches the independently obtained exact PR6545 native archive, plus byte-identical copies of the predecessor caller and retirement witness; its external WITNESS.json is also identical. There are eighteen native invocations in both callers. Thus the common native paid and helper terms cancel. On the retained structural contract, both complete invoices have 13,756 productive fields and 67,096 paid jobs. Both target and paid rank mass change by −832; their difference is unchanged. Two extra Q_POWER translations are explicitly reported as ordinary events, rather than silently dropped from the event inventory. + +## Strict gain for every 0f'(56+s)`. Integrating gives `41^t−40^t > 57^t−56^t`. Multiplication by 208 proves strict positive Δ(T−P). The difference is zero at t=0 and t=1; neither endpoint is claimed strictly positive. This proof establishes a complete recursive Gaussian invoice improvement in the retained model, not the new numerical κ or physical supplier admission. + +## Evidence and controls + +`accounting.py` is standard-library Python and has seven input-sensitive rejection controls: + +- Delete the third child's extra whole idle. +- Give the fresh suffix an unearned central Gaussian factor. +- Delete a fresh stage from that suffix target. +- Delete one rank-one query. +- Delete one native invocation, invalidating cancellation. +- Replace an ordinary Q² translation with a paid odd Gaussian exponent. +- Exchange the retired/fresh field targets, preserving their histogram but violating their interface. + +Both actual sources are positive fixtures. Normal and optimized Python 3.12.14 executions passed with byte-identical JSON. Measured subprocess times were approximately 0.08 s and 0.21 s; maximum child RSS was below 35 MiB. Exact commands, status and timings are in `evidence/accounting-normal-run.json` and `evidence/accounting-optimized-run.json`; the deterministic scientific result is `RESULT.json` (the accounting subsection in this integrated package). + +From a checkout with the frozen directories staged under the indicated paths: + +```sh +python -I -B accounting.py --predecessor PATH/gaussian-root-retirement --candidate PATH/retirement-rebalanced-query-tree +python -I -O -B accounting.py --predecessor PATH/gaussian-root-retirement --candidate PATH/retirement-rebalanced-query-tree +``` + +## Exact dependencies and limits + +Runtime reads: + +- Predecessor: `CONTROLLER.json`, `WITNESS.json`. The native archive is read in memory from the candidate's exact retained `source.tar`, and its independently observed predecessor SHA256 is checked. +- Candidate: `CONTROLLER.json`, `ROOT.json`, `WITNESS.json`, `source.tar`. +- Nested old ROOT: `native-source.tar!research/p104-scaled-transport-fusion/source.tar!research/p104-third-anchor-clean-mixed-root/root-source.tar!research/mixed-projector-query-root/ROOT.json`. +- Nested native word: `native-source.tar!research/p104-scaled-transport-fusion/WORD.json.gz`. + +No archive paths are extracted. Source-pin authentication should surround this checker in the integrated package. The native archive is checked against the independently observed predecessor hash; the other reported SHA256 values identify bytes actually read and need the integrated source pins to bind GitHub heads. + +The caller identity is a separate obligation: this checker assumes the physical meaning of the serialized endpoint rows, and checks accounting/interface consistency rather than proving those endpoints from gate composition. It does not replay native phase maps, verify the retirement prefix/dual-pair identity, prove arbitrary-input restoration, certify the numerical root, validate changed completion decay, implement a uniform compiler, price total ordinary work, or resolve any of the five framework assumptions. Those obligations must remain separately labelled even if this delta check passes. + +Credit Ali Romman for PR6551 and PR6545, the PR6398 mixed-root and PR6529 native-source lineage, and all inherited source notices. This independent accounting implementation and explanation were prepared for DLamdata with substantial OpenAI Codex assistance. Apache-2.0; no independent human or proof-kernel verification is claimed. diff --git a/research/pr6551-retirement-audit/DOCUMENTATION-REVIEW.md b/research/pr6551-retirement-audit/DOCUMENTATION-REVIEW.md new file mode 100644 index 0000000000..2967117a4c --- /dev/null +++ b/research/pr6551-retirement-audit/DOCUMENTATION-REVIEW.md @@ -0,0 +1,24 @@ +# Read-only documentation review + +Reviewed README.md, EXTERNAL-REVIEW.md, PRICING-SOURCE-MAP.md, ENDPOINT-REVIEW.md and pricing_binding.py against the supplied external audit text. No scientific verifier or artifact-regenerating command was run. The extended checker hash in ENDPOINT-REVIEW.md matches the current file (`3b15f6c17d2e7a941c97a29cd4269388cd07f824a4d8df16fc22a62c1515fcf8`). Publication receipts are intentionally outside this interim check. + +## Findings to resolve + +1. **Minor factual source-map correction — PRICING-SOURCE-MAP.md, paragraph beginning “The candidate's verify.py”.** It says `native_word`, `local`, `native`, and `L` all pass to both the predecessor invocation and the new price calculation. Exact candidate code passes `(native_word, old_caller, local, native, L)` to `source_candidate.run` at line 88, but only `(local, native, word, physical, L)` to `price.run` at line 136. The native word instead goes to `physical(..., native_word, L)` at line 131. Suggested replacement: “The candidate's `verify.py` `Extension.run` passes the same `local`, `native` and `L` objects to the predecessor invocation at line 88 and the new price calculation at line 136; the same `native_word` is supplied to the physical-map check at line 131.” This does not alter the cancellation conclusion. + +2. **Attribution clarity — README.md, Pins and independence final paragraph.** The sentence “The reviewer made the conceptual suffix argument's implicit invertibility premise explicit…” follows the Claude paragraph, so a reader can attribute that original internal-review finding to Claude. Earlier internal review supplied it. Change “The reviewer” to “The internal skeptical reviewer” (or name that role explicitly). Claude's six documented findings and later Codex implementations are otherwise cleanly distinguished. + +## Checks with no issue found + +- Claude Code / Claude Opus 5.5 is credited for the external audit and findings; GPT-6 Astra through Codex is credited for the lead/internal roles and follow-up implementation. No independent human or formal-kernel review is claimed. +- External results are explicitly **reported** at `13e880a734cc03f2a52c82a412a9a0e6ac2b1f0c`; unpublished scripts/logs are not represented as locally replayed evidence. The later full-fiber checker is correctly separated from that external snapshot. +- The external report's early contradictory progress wording (“caught” versus “accepts”, “passes” versus “past” the decay) is resolved using its final substantive findings: four suffix changes passed the old cut check, and the candidate ratio exceeds 31/32. +- The additional local full-endpoint coverage and four suffix rejection controls are identified separately from the old cut check; redundant guards are not falsely described as independently mutation-covered. +- Physical tariff validity remains inherited. The exact 232,960 ordinary allowance increment is presented as arithmetic under the pinned formula, not a proof of physical rates or supplier admission. +- Local-manifest integrity is distinguished from external Git-object provenance, with an explicit trusted-publication/checker boundary. The AST matching helper calls its fragments copied source contracts, not independently derived physical semantics. +- `research/20261103-open-obligation` is correctly the branch, and `research/pr6551-retirement-audit` the package directory. The external report's initial branch confusion was corrected and is not repeated. +- The old 20-file/53-file external snapshot and current 22 retained sources are distinguished. Historical receipts are labelled; this review intentionally does not flag pending final publication receipts. + +No blocking scientific or scope-overstatement issue was found in these documents. Resolve the two small factual/attribution edits before publication, then let the lead's serial validation and final receipt process complete. + +Lead resolution: both wording findings were corrected before the scientific follow-up commit. The source-map line now distinguishes the shared pricing arguments from native_word passed to the physical checker; README names the internal skeptical reviewer. No scientific code or receipt changed. diff --git a/research/pr6551-retirement-audit/ENDPOINT-REVIEW.md b/research/pr6551-retirement-audit/ENDPOINT-REVIEW.md new file mode 100644 index 0000000000..eb16187d53 --- /dev/null +++ b/research/pr6551-retirement-audit/ENDPOINT-REVIEW.md @@ -0,0 +1,58 @@ +# Independent full-endpoint extension after external review + +**Result:** both frozen callers pass independent full 35-by-35 endpoint verification in all four exact central fibers, with every stage variable formal. The previous retirement-cut and input-lifetime checks remain. The four requested changed-suffix controls all pass those previous checks but fail the new endpoint calculation with explicit coefficient counterexamples. + +This extends the review coverage; it does not change either caller, source claim, invoice, source pin or selected frontier. No predecessor constructor, source checker, lead checker or pricing routine is imported or executed. The new engine reuses only this reviewer's own Gaussian-rational, formal-stage vector interpreter. + +## Exact scope + +The retained inputs remain PR6545 `775722aa94a5e85b05d1d704d7dca10cbf016f0c` and PR6551 `11ca568cd1a16776ade02b4df09ea04a784cc674`, reviewed through contribution snapshot `13e880a734cc03f2a52c82a412a9a0e6ac2b1f0c`: + +- Old controller SHA256: `e5c873568b52aab82fc01670de3037b1ba4bedbcb15366f0e5fb38cd786714ed`. +- New controller SHA256: `419566fe8d8e84a21a14b8efbdac83df5cf1892b0de55f4f256287e9031d74da`. +- Extended checker SHA256: `3b15f6c17d2e7a941c97a29cd4269388cd07f824a4d8df16fc22a62c1515fcf8`. +- Deterministic receipt SHA256: `ff7621c206e1a553c95e41db7a13cd70877aaf4646ccab2d9f603620adcfb5a9`. + +The standalone checker reports controller hashes but does not check their expected values itself. The integrated wrapper retains its source-integrity checks before launching this checker; independent source provenance requires the optional Git-object comparison. + +## Complete endpoint argument + +For each q in `1,-1,i,-i`, the checker starts with each of the 35 unit output rows and propagates it backward through every event (609 old, 611 new). Every one of the 35 resulting input coefficients is compared with its prescribed diagonal coefficient. This is 1,225 formal coefficient positions per fiber, 4,900 per caller, and 9,800 across the two actual callers. The coefficient representation is independent of the lead's complete transfer matrix: each row is processed separately after exact q specialization, with Gaussian rational coefficients and 19-component formal stage-exponent tuples. D0 is retained explicitly and required to be zero in target signatures; D1 through D18 are never numerically specialized. + +The existing Chinese remainder argument applies to the full matrices: over `A=Q(i)[D0,...,D18]`, the four factors of `q^4-1` are pairwise comaximal, so evaluation gives an injection `A[q]/(q^4-1) -> A^4`. Exact equality in all four fibers therefore establishes each complete endpoint identity in the claimed formal quotient ring. This is not random testing or a specialization of the stage operators. The written CRT argument is not a formal-kernel proof. + +The matrices still use the declared LOCAL, QUERY, IDLE, LAMBDA and ordinary scalar contracts. Physical native maps, all-width substitution, scalar-outside-tensor convention, ordinary work, numerical moments, compiler and outer obligations remain separate. Agreement with the primary replay is implementation corroboration under shared semantic and Python Fraction assumptions, not independent physical admission. + +## Independently identified changed region + +A maximal literal comparison gives exactly 358 common initial events and 67 common final events. The changed half-open slices are old `[358:542]` and new `[358:544]`. Changed diagonal roles are `20,21,22,23,24,25,26,27,34`. The 345-event retirement prefix is inside the unchanged prefix; this explains why the earlier cut-only checker alone could not detect the requested suffix errors. Identical trailing instructions do not imply identical incoming states, so the extension checks the complete end-to-end map. + +## Requested failure controls + +Each control first passes the existing cut and lifetime checks, then must raise the specifically typed `EndpointMismatch`. An input-validation or cut-marker failure is not accepted as evidence of suffix coverage. The receipt records the first exact mismatch, including its full formal polynomial, expected polynomial, q fiber and matrix coordinates. + +| Invalid change | First mismatch at q=1 | Detection | +|---|---|---| +| Flip event363 `ADD 1 22 1` to coefficient -1 | output23, input20 | Actual coefficient is `2 D10 D11`, expected zero. | +| Remove D11 from event358's five-stage IDLE | output0, input0 | Actual stage monomial differs from the full target. | +| Swap event371 LOCAL source and destination | output0, input0 | Actual diagonal coefficient differs from the required one. | +| Remove D11 from fresh coordinate34's target | output34, input34 | Literal caller retains D11; declared target omits it. | + +These four controls preserve the first345 events and pass the original retirement-cut checks. Their endpoint failures are thus direct evidence for the added scientific coverage. The controls stop at the first disagreement; they do not claim a replay of every later mismatch. + +Both actual source callers are positive fixtures. A second new positive fixture inserts an inverse half-ADD pair on arbitrary fresh input34 into the changed suffix after event371; all 4,900 endpoint coefficient positions still pass. This tests a semantically valid suffix edit instead of simply rejecting every deviation from the pinned word. The previous canceling-ADD prefix fixture and four previous cut/lifetime rejection controls remain. An additional artificial pivot-column failure was unnecessary: the retained column check remains, and all complete coefficient positions now receive independent coverage. + +## Actual commands and results + +From the intended checkout after environment activation: + +```sh +python -I -B /workspace/work/research-assignment/claude-followup/fibers/cut_four_fibers.py /workspace/work/research-assignment/checkout/research/pr6551-retirement-audit/source/pr6545/CONTROLLER.json /workspace/work/research-assignment/checkout/research/pr6551-retirement-audit/source/pr6551/CONTROLLER.json +python -I -O -B /workspace/work/research-assignment/claude-followup/fibers/cut_four_fibers.py /workspace/work/research-assignment/checkout/research/pr6551-retirement-audit/source/pr6545/CONTROLLER.json /workspace/work/research-assignment/checkout/research/pr6551-retirement-audit/source/pr6551/CONTROLLER.json +``` + +Both exit0 with `PASS_FOUR_FIBER_RETIREMENT_CUT_AND_ENDPOINTS`; normal and optimized receipts are byte-identical. CPython 3.12.14, standard library only. Normal run: 1.099399 seconds, 11904 KiB child high-water RSS. Optimized run: 1.157566 seconds, 14788 KiB. The measurement wrappers used a 120-second timeout and Python `time.monotonic`/`resource.getrusage`. Full executed interpreter paths, arguments, exit status, stderr and measurements are in [fiber-normal-run.json](evidence/followup/fiber-normal-run.json) and [fiber-optimized-run.json](evidence/followup/fiber-optimized-run.json). No scientific run failed. + +The extension was developed and run only in the assigned separate output directory. No project file, Git ref, public message or PR was changed by this agent. The lead integrated the new script/receipt; [FOLLOWUP-VALIDATION.md](FOLLOWUP-VALIDATION.md) records the serial registered package, export and publication-byte checks. The previous reviewed source/caller snapshot is unchanged, while this expanded checker snapshot requires its own evidence. + +Prepared for DLamdata with substantial GPT-6 Astra assistance through OpenAI Codex in the independent review role; Apache-2.0. Ali Romman and the retained predecessor contributors keep construction, retirement and native-interface credit. This is AI review, not human or maintainer acceptance. diff --git a/research/pr6551-retirement-audit/EXTERNAL-REVIEW.md b/research/pr6551-retirement-audit/EXTERNAL-REVIEW.md new file mode 100644 index 0000000000..e8018288d0 --- /dev/null +++ b/research/pr6551-retirement-audit/EXTERNAL-REVIEW.md @@ -0,0 +1,38 @@ +# External AI review and response + +Claude Code / Claude Opus 5.5 reviewed contribution snapshot `13e880a734cc03f2a52c82a412a9a0e6ac2b1f0c` on branch `research/20261103-open-obligation`. The contributor supplied its written audit. Claude reported no scientific error in the finite caller identity, exact CRT argument, input independence, common native-cost cancellation or complete Gaussian invoice delta under the declared semantics. It changed and pushed nothing. + +This note summarizes that supplied report; Claude's scripts and raw execution logs were not supplied, so they have not been rerun or incorporated here. The later implementation and local evidence are GPT-6 Astra / Codex work. External review of the old snapshot must not be described as review of the new checker bytes. Neither review is independent human review, formal-kernel verification or maintainer acceptance. + +## Reported external evidence + +Claude reported CPython 3.12.3 runs of the registered hook (2.2 seconds), normal/optimized direct replay, the 53-file commit/history preflight, final-SHA export integrity and both read-only Git-free cold replays with unchanged bytes. It reported direct comparison of all 20 then-retained files to the three source commits and agreement of the native archive hash. Its separate, unpublished third checker reportedly verified both full 35-by-35 endpoints in all four central fibers with formal stage variables and independently recomputed the source-defined invoice. These are attributed reported results, distinct from our committed local receipts. + +## Findings and resolutions + +1. **Changed-suffix coverage.** The prior second engine checked the first 345 events, inside a 358-event common prefix, and only field touches thereafter. Claude supplied four suffix mutations accepted by that cut-only check. The extended second engine now checks both complete endpoints (4,900 formal coefficient positions per caller), independently of the lead matrix implementation. Each of those four mutations passes the previous cut/lifetime checks and fails an actual endpoint coefficient. A valid inserted inverse ADD pair passes. [ENDPOINT-REVIEW.md](ENDPOINT-REVIEW.md) gives exact witnesses and the CRT argument. +2. **Integrity versus provenance.** The local manifest can be consistently replaced along with retained sources. We corrected the terminology and added an optional checker using an externally supplied publication SHA and the original Git source objects. The explicit substitution experiment and trust boundary are below. +3. **Historical publication receipts.** Original committed receipts cover `28929ae5b3bb575c0bb423d26d98d6f878ab8dd3` (47 files), while the externally reviewed snapshot is `13e880a734cc03f2a52c82a412a9a0e6ac2b1f0c` (53 files). FINAL-REVIEW.md's VALIDATION hash is also historical. Those records are preserved and labelled. [FOLLOWUP-VALIDATION.md](FOLLOWUP-VALIDATION.md) identifies each new tested snapshot, unchanged scientific-content hashes and later evidence-only changes. Final exact-head checks belong in the PR's publication record; no receipt claims to include its own future commit. +4. **Redundant guards.** Several row/column/fresh-target checks do not independently trigger in the original eleven caller controls. Under the checked diagonal endpoint, untouched-field and invertibility hypotheses these are redundant; this is not evidence that every guard was separately exercised. We make no such coverage claim. The new suffix controls target a distinct scientific gap rather than manufacturing failures for redundant guards. +5. **Source tariff binding.** Both exact predecessor/candidate `pricing.py` files are now retained and directly compared with their Git objects. [PRICING-SOURCE-MAP.md](PRICING-SOURCE-MAP.md) relates the independent counts to the source rules. A non-executing AST comparison checks the reviewed contract clauses and has three failure controls. Exact ordinary arithmetic gives 254,699,248 and 254,932,208, a difference of 232,960. The physical validity of the tariff constants remains inherited. +6. **Shared assumptions and AI assistance.** The source discloses Codex assistance. The lead and both internal review roles use GPT-6 Astra through Codex. All checks, including Claude's reported checker, share the opcode meanings, source billing contract and Python rational arithmetic, so different implementations and model providers do not remove those common assumptions or guarantee independent errors. + +## Source integrity and provenance + +The portable registered replay checks source bytes against the package's own RETAINED.json and recomputes science. It is useful for accidental corruption and reproducibility, but does not establish external origin when code and manifest can be replaced together. + +`check_source_objects.py` is a separate, optional Git-dependent comparison. Supply a full publication SHA obtained through a trusted channel and make that publication and the three pinned source commits available in the object database. The checker reads the manifest from that publication commit, requires the working manifest to match, and compares every retained file with its original commit/path and both recorded hashes. It compares working retained files, not the publication commit's stored copies, code or RESULT; the separate committed-byte preflight checks those published bytes. A Git object comparison alone does not establish which remote repository supplied the object. It does not fetch or execute source code. Its own code and the supplied publication pin must be trusted; this is not a defense against an attacker replacing the checker, trusted pin and all evidence. + +```sh +python3 -I -B research/pr6551-retirement-audit/check_source_objects.py --repository . --publication-commit FULL_PUBLICATION_SHA +``` + +The three source commits are #6551 `11ca568cd1a16776ade02b4df09ea04a784cc674`, #6545 `775722aa94a5e85b05d1d704d7dca10cbf016f0c`, and #6398 `dee2663ba0918be86993fd9d63c8d64b020f4c83`. The publication record supplies the full contribution SHA; a branch name is deliberately rejected as an anchor. A network-free export does not require this optional check or these Git objects. + +[provenance-control.json](evidence/followup/provenance-control.json) records a positive comparison of all 20 files in the externally reviewed original snapshot. A temporary copy then appends harmless text to a retained proof and consistently changes its two local manifest hashes. Its portable optimized replay still passes; the new checker anchored to the original publication commit rejects the changed manifest. [direct-source-comparison.json](evidence/followup/direct-source-comparison.json) compares all 22 current retained files, including both pricing sources, with their source Git objects. This demonstrates the different trust boundaries without changing any frozen predecessor in the contribution. + +## Remaining question and credit + +Claude also emphasized why a favorable invoice does not establish the completion tail: its reported approximate completion ratio rises from 0.964631 to 0.968921, exceeding 31/32 = 0.96875. The candidate changes the decay to 97/100 and the reserve from 10^-29 to 10^-27. Those numerical/tail facts have not been independently certified here; the separately derived rational tail audit remains the next useful task. Native/all-width maps, physical ordinary rates, numerical κ, compiler and outer predicates, and all five framework assumptions remain outside this finite audit. + +Credit Claude Code / Claude Opus 5.5 for the external audit and these six findings. Credit Ali Romman and retained predecessor authors for the construction, retirement mechanism, billing model and native integration. Follow-up code, tests and packaging by DLamdata with substantial GPT-6 Astra / OpenAI Codex assistance; Apache-2.0. No new κ, supplier admission or frontier change is claimed. diff --git a/research/pr6551-retirement-audit/FINAL-REVIEW.md b/research/pr6551-retirement-audit/FINAL-REVIEW.md new file mode 100644 index 0000000000..b3afb2fd71 --- /dev/null +++ b/research/pr6551-retirement-audit/FINAL-REVIEW.md @@ -0,0 +1,55 @@ +# Final integrated internal review + +**Verdict: no unresolved blocking finding for the scoped finite caller/cut and complete Gaussian invoice-delta audit.** One source-navigation issue was found and resolved before publication. This is separate AI review, not human review, formal verification, supplier admission or certification of PR6551's numerical kappa. + +## Reviewed bytes + +The following SHA256 values identify the integrated files reviewed after the navigation and measured-memory corrections. The package root is `research/pr6551-retirement-audit/`. VALIDATION.md was read in the final evidence-only follow-up and is included below. + +| File | SHA256 | +|---|---| +| `caller.py` | `667dfc0b99b194dd4213f44596f8a7de4916022517a63da589fc2f201a8fcd2b` | +| `accounting.py` | `c032bab18b8fe8763fec40958a8f6ae09d7e973755005347cec4e74f86128605` | +| `cut_four_fibers.py` | `49b0be279b33c4b346cff8ebc4b10ed97f63e1efe0fcb0ffac9cc2f11dc9defa` | +| `verify.py` | `a36e1b376308ffaaba752fa5230261ecd2ac13bceb6a4e989f172b144a9e78e7` | +| `README.md` | `15e15154c0cf6cfa1a172b259d36f710bb99f912ac1f66976a5d258781348a83` | +| `REVIEW.md` | `983393592d3249232b79752a5f08606abf5b7bb24c2ac943a84bba7925960c4a` | +| `TASK.json` | `7d6ce0fb843322092a12816f837fe0ae58452459c330fa810e004b0c6d1ab000` | +| `VERIFY.json` | `2e6eaba0ad815003e17381e1a992045bed0b10551acb114de2ffd77d018138c2` | +| `RESULT.json` | `f872cfe53b831e904cae956abb81138efb1ce735ace664fcfdca53592da98364` | +| `RETAINED.json` | `5568ad3ee8390e07c40400a12c27dd8666331d0d1bc87da7a9e71de7cd59ca31` | +| `VALIDATION.md` | `7b3e9fd1ebf6f2bc8e4897b935437b8ea56ae170f7bb77fd5bce258fe320fa41` | + +## Integration and source authenticity + +The wrapper authenticates every retained input by both SHA256 and Git-blob framing before running an audit module. It also requires the exact retained-file inventory. It never imports a retained predecessor Python module. It imports only the new caller/accounting modules under a temporary bytecode-cache namespace; the separate cut script runs in an isolated Python subprocess with its own temporary cache namespace. The normal and optimized manifests invoke identical scientific checks, and the receipt comparison is enabled by default. No direct scientific replay was run concurrently with the lead's registered replay during this final review. + +I independently compared all 20 retained source entries. Sixteen direct PR6545/PR6551 objects matched `git show SOURCE_COMMIT:SOURCE_PATH` byte for byte with `GIT_NO_REPLACE_OBJECTS=1`. The four PR6398 original commits are unavailable in the local Git object database (read returned status128); instead, all four copied inspection sources matched their members through the exact, already authenticated PR6551 source archive chain. The copies are consequently backed by the pinned PR6551 archive without needing side-branch history at replay time. This final check does not newly establish the original PR6398 commit mapping independently of the retained predecessor metadata. The recorded original pins and retention chain disclose that distinction. + +The complete caller uses exact Q(i) coefficients and symbolic stages, checks both 35-by-35 endpoint maps, and verifies pivot/fresh row-and-column isolation plus access lifetimes. The explicit zero check before dropping serialized stage D0 prevents an accidental loss of a real stage variable. The second cut engine uses all four exact central fibers and the written Chinese remainder argument; it is not a random numerical test. The accounting engine derives target ranks and paid event counts from literal source data and independently traverses binary trees. Its same-native cancellation is conditional on the declared native interface; the unchanged native archive and 18 ordered invocations are checked, not mathematically replayed. + +The receipt contains the advertised 11 primary-caller, 7 accounting and 4 cut failure controls. All primary result fields examined agree with README scope. In particular the literal ordinary counts are ADD295, LAMBDA34, SCALE86, SWAP126 for both sources, and Q_POWER17 versus19. This is a complete Gaussian invoice delta with two extra ordinary events counted. It is not a proof of the total ordinary-operation tariff, nor a complete price/outer proof. The README clearly says so and retains every framework assumption. + +The 20-file input record, default receipt comparison, no-executed-predecessor claim and self-contained replay design are consistent. Wrapper source authentication is an integrity check against declared pins, not a claim of independent upstream Git availability for every historical source. Scientific correctness still rests on the exact fixed-source semantics and written arguments described in REVIEW.md. + +## Finding and resolution + +The first integrated TASK and README directed the next reviewer to a "retained PR6551 pricing.py", but that file was not in the copied candidate inventory; the nested archive contains predecessor pricing and the candidate declares adaptations. The lead corrected both next-step texts to link the exact public PR6551 pricing.py at `11ca568cd1a16776ade02b4df09ea04a784cc674`. I read back both replacements and confirmed the links. The README also corrected its measured-memory statement to the recorded child high-water below64MiB (63,820KiB). I read back that wording and the new VALIDATION.md, which records successful local evidence while leaving committed-byte/export checks pending. No checker, retained source or RESULT bytes changed. This resolved a navigation/claim-accuracy issue, not a scientific counterexample. + +The earlier proof review supplied the implicit invertibility of the complementary cut block in the Laurent stage ring and emphasized that ordinary coefficients stay outside tensor powers. Those qualifications remain in integrated REVIEW.md. Neither finding is left as an unacknowledged proof gap. + +## Actual final-review operations + +After sourcing `/workspace/.integer-mult-env/activate` in the intended checkout, I used Python hashlib to record the eleven file hashes; read the integrated checker, manifests, receipt and prose with `cat` and Python JSON parsing; counted receipt controls and ordinary-event inventories; and ran the following read-only source comparison for each declared input: + +```text +git show : +``` + +Those subprocesses set `GIT_NO_REPLACE_OBJECTS=1`, used captured output and a ten-second timeout. Sixteen returned0 and matched; four missing historical objects returned128. Python `tarfile` then read, without filesystem extraction, the nested archive members for the four PR6398 files and compared all bytes successfully. The retained-input inventory supplies the exact paths, hashes and archive chain needed to repeat these comparisons. No network request, Git mutation, package edit, wrapper replay or artifact-regenerating suite was performed by this final review. + +## Remaining limits + +The lead's recorded local evidence reports the registered normal/optimized verification and additional optimized direct replay passed; I read those exit records and checked the logs contain the integrated result marker. I did not rerun them. Committed-byte and export/cold checks remain lead publication steps at this review snapshot; their success is not inferred here. Native all-width maps, physical ordinary rates, numerical crossing, changed completion tail, compiler and outer predicates remain outside this audit. Independent implementations share source opcode meanings, physical rank interpretation and Python rational arithmetic. These dependencies and the five inherited assumptions remain disclosed. A future source/checker/receipt change invalidates the corresponding reviewed hashes and needs a targeted recheck. + +Prepared for DLamdata with substantial OpenAI Codex assistance in the separate skeptical-review role; Apache-2.0. Predecessor construction credit remains unchanged. diff --git a/research/pr6551-retirement-audit/FOLLOWUP-REVIEW.md b/research/pr6551-retirement-audit/FOLLOWUP-REVIEW.md new file mode 100644 index 0000000000..038496c89d --- /dev/null +++ b/research/pr6551-retirement-audit/FOLLOWUP-REVIEW.md @@ -0,0 +1,70 @@ +# Independent review of the expanded audit integration + +**Verdict: no unresolved blocking finding within the finite caller, cut, source-contract binding and Gaussian invoice-delta scope.** This verdict was derived from code, source objects and receipts before reading any response to the external review. The existing source/native assumptions are unchanged. This is separate AI review, not independent human review, formal verification, a new numerical kappa certificate or supplier admission. + +## Snapshot + +Reviewed contribution baseline: `13e880a734cc03f2a52c82a412a9a0e6ac2b1f0c`. Construction sources remain PR6551 `11ca568cd1a16776ade02b4df09ea04a784cc674`, PR6545 `775722aa94a5e85b05d1d704d7dca10cbf016f0c`, and PR6398 `dee2663ba0918be86993fd9d63c8d64b020f4c83`. The following working-file hashes identify the reviewed expanded code and receipts; a new publication commit must bind them. + +| File | SHA256 | +|---|---| +| `check_source_objects.py` | `793b5eba68ff87deb137c2b07f24b434790538fca76701bb80ab271088420a9f` | +| `pricing_binding.py` | `a0362e2bfb29ef38f914e8d6d89784a6078a0e24a8f7cc2ea07385d071c113f9` | +| `verify.py` | `ec190aa7e13c2d73ec871e9f2cc4ffdfe031ee7ec4722538f648127d786d5bdb` | +| `cut_four_fibers.py` | `3b15f6c17d2e7a941c97a29cd4269388cd07f824a4d8df16fc22a62c1515fcf8` | +| `caller.py` | `667dfc0b99b194dd4213f44596f8a7de4916022517a63da589fc2f201a8fcd2b` | +| `accounting.py` | `c032bab18b8fe8763fec40958a8f6ae09d7e973755005347cec4e74f86128605` | +| `RESULT.json` | `8c6a7c6fc8872b8de465ebd3101428c9344aed3d93bf82a1a930f183802c8c13` | +| `RETAINED.json` | `4dadc39c1e4b65f74ab1874107c64ebe58ba10018062f74eaad8c4b551381cb8` | + +The package root is `research/pr6551-retirement-audit/`. Documentation and publication evidence were still being updated by the lead and are not included in this code-snapshot verdict. + +## Full independent endpoint extension + +The integrated four-fiber script has exactly the same hash as the independently developed and tested extension. The integrated `four_fiber_cut` receipt is exactly equal, as parsed JSON, to the independently sealed normal/optimized receipt. It retains all earlier cut/lifetime checks and adds every coefficient of both complete 35 by 35 endpoints at q=`1,-1,i,-i`, with formal stage variables. The CRT justification supplies exact coverage of the full q quotient ring, rather than probabilistic testing. Both source callers and a valid inverse half-ADD suffix edit pass. + +The four new suffix mutants are required to pass the original cut checks and to raise `EndpointMismatch` from the complete transfer comparison. This avoids confusing source-schema rejection with actual suffix coverage. Explicit first disagreement polynomials are retained. Maximal common prefix 358 and suffix 67, the changed event slices, and nine changed diagonal-role records are computed from literal data. The earlier shared 345-event cut is not presented as sufficient evidence for the changed suffix. + +The independent full-caller engine remains unchanged and uses a different polynomial organization. Shared assumptions include the declared opcode meanings, Gaussian physical rank interpretation, factorwise tensor convention and Python rational arithmetic. These engines do not independently establish physical native maps or all-width implementation. + +## Pricing-source binding and ordinary arithmetic + +`pricing_binding.py` parses, but never executes, the two exact pinned pricing sources. Its copied AST tariff clauses are explicitly labeled a **source contract**, not an independently derived physical law. The independent accounting implementation remains separate. The source hashes bind the entire file, while the AST checks identify the precise clauses that were reviewed. The changed-source, query-tariff and raw-tariff controls reject; the two actual source files are positive fixtures. + +The wrapper recomputes the declared ordinary expression using the actually replayed 609/611 event counts and requires exact values 254699248/254932208. This establishes arithmetic under the copied source formula. It does not prove the physical 32-per-field rate, local 10^7 allowance, routing costs or ultimate 10^9 bound. The receipt key `declared_ordinary_bounds` and the pricing receipt's explicit scope keep that distinction visible. The complete Gaussian delta continues to cancel the exact same 18 native invocations under the shared native contract; the two extra ordinary translations are neither omitted nor declared free. + +I ran the pricing binding directly under isolated optimized Python. It passed and produced exactly the integrated pricing receipt, so no stale clause-binding receipt was found. The aggregate now contains 29 internal rejection controls: 11 primary caller, 7 accounting, 4 retained cut, 4 full-endpoint and 3 pricing. External corruption/provenance controls are separate evidence and should not be silently included in that count. + +## Integrity and the external trust anchor + +`verify.py` now correctly names its portable source check **integrity**. It checks both hashes and the exact source-file inventory before importing only the new audit modules. It gives those imports a fresh bytecode-cache namespace and launches the separate endpoint checker with its own temporary cache. The default compares the freshly computed aggregate with RESULT; `--emit` is explicit. No source pricing code, constructor or predecessor scientific verifier executes. + +`check_source_objects.py` adds a separate, optional provenance operation. It requires an externally supplied full publication commit, reads that commit's RETAINED manifest with replacement objects disabled, requires the working manifest to equal it, and compares working retained inputs with each declared source commit/path and blob identity. It never fetches. This is an appropriate narrow source-object check when the publication pin and checker are obtained through a trusted channel. + +A material boundary is explicit: that optional check anchors the manifest and **working retained source files**. It does not by itself validate all source copies, audit code or RESULT stored in the publication commit, establish a GitHub repository's ownership of an object, or authenticate its own executable. The separate committed-byte preflight and trusted acquisition of the checker/publication pin remain necessary. I notified the lead to preserve this distinction in the documentation; no change to the code was needed. + +The recorded positive 20-file check and consistent-substitution rejection use the old 13e880a publication manifest. That is a meaningful control of the external-anchor boundary: a joint harmless source/manifest substitution can pass portable integrity while failing the anchored comparison. It is not a recorded pass of the expanded 22-file manifest at a new publication commit. The lead must run the tool again with the new final publication pin before claiming that new pass. + +Independently of that future publication step, I compared every current retained source directly to its exact original Git object with `--no-replace-objects`: all 22 match. In contrast to the first review, PR6398's original commit is now available locally and all four of its copies matched directly. The earlier retention-chain fallback is superseded for this source-comparison observation only; it did not change any source bytes. + +## Actual review operations + +All operations ran from the intended checkout after sourcing `/workspace/.integer-mult-env/activate`. I read the expanded scripts and current aggregate/input manifests, recorded the eight hashes above, inspected the source-comparison/provenance-control evidence, compared the integrated CRT JSON with the independently sealed JSON, and queried each exact source using: + +```text +git --no-replace-objects show SOURCE_COMMIT:SOURCE_PATH +``` + +Each source command had a ten-second timeout and captured output; all 22 returned 0 and matched the corresponding copied bytes. I also ran: + +```sh +python -I -O -B research/pr6551-retirement-audit/pricing_binding.py research/pr6551-retirement-audit/source/pr6545/pricing.py research/pr6551-retirement-audit/source/pr6551/pricing.py +``` + +It returned 0 in 0.182039 seconds; all three controls passed and the JSON equaled `RESULT.json`'s pricing section. Exact arguments and comparison outcomes are in the companion reviewer check record. No full registered replay, exporter, artifact regeneration, Git mutation, public write or nested agent was run during this final integration review. No additional source dependency was installed. + +## Remaining publication and scientific work + +The lead's serialized registered replay, export/cold replay, committed-byte check and new-publication source-object check must bind the integrated snapshot before publication. Code, source or receipt changes invalidate the corresponding hashes above and require targeted rechecking. Docs-only corrections do not establish or invalidate a new scientific theorem by themselves. + +No numerical crossing, changed completion tail, compiler, outer theorem, physical native/operator tariff, all-width realization, erasure, precision or tape claim is added here. The five inherited framework assumptions remain. Construction and retirement credit remain with Ali Romman and the retained predecessor authors; new audit work is prepared for DLamdata with substantial OpenAI Codex assistance, Apache-2.0. diff --git a/research/pr6551-retirement-audit/FOLLOWUP-VALIDATION.md b/research/pr6551-retirement-audit/FOLLOWUP-VALIDATION.md new file mode 100644 index 0000000000..94908085a3 --- /dev/null +++ b/research/pr6551-retirement-audit/FOLLOWUP-VALIDATION.md @@ -0,0 +1,48 @@ +# Verification after Claude's review + +This follow-up keeps the exact #6551/#6545 construction pins and mathematical claim of original contribution `13e880a734cc03f2a52c82a412a9a0e6ac2b1f0c`. It expands the second engine to both full endpoints, retains and binds the two pricing sources, and separates local integrity from optional Git-object provenance. [EXTERNAL-REVIEW.md](EXTERNAL-REVIEW.md) distinguishes Claude Code / Claude Opus 5.5's reported review of the original snapshot from the new GPT-6 Astra / Codex implementation and internal review. + +## Current scientific evidence + +CPython 3.12.14, Linux x86_64, standard library only. Actual commands from the activated contribution checkout: + +```sh +python3 -B scripts/research_package.py --package research/pr6551-retirement-audit +python3 -I -B research/pr6551-retirement-audit/verify.py +python3 -I -O -B research/pr6551-retirement-audit/verify.py +``` + +All pass. The registered normal/optimized pair took 3.702 seconds; additional direct runs took 1.609 and 1.778 seconds and reproduced RESULT.json byte-for-byte. RESULT SHA256 is `8c6a7c6fc8872b8de465ebd3101428c9344aed3d93bf82a1a930f183802c8c13`. The measuring process's cumulative child high-water mark was 95,784 KiB. Every subprocess had a 120-second outer limit, with the integrated four-fiber child additionally bounded at 30 seconds. Exact commands, timestamps/source context, elapsed measurements and logs are in [evidence/followup](evidence/followup). + +The integrated receipt contains 26 semantic failures (11 primary caller, seven accounting, four cut, four changed-suffix) plus three source-contract failures. New suffix controls pass the earlier cut/lifetime checks before failing specific endpoint coefficients; a valid inverse-pair suffix edit passes. These distinguish useful coverage from blanket rejection of changed files. Three external optimized corruption controls also reject modified caller bytes, native archive bytes and expected receipt; [external-corruptions.json](evidence/followup/external-corruptions.json) records exact error strings. These controls demonstrate integrity and recomputation, not origin authentication. + +All 22 retained inputs match their exact original Git objects, including the four #6398 files and two newly retained pricing sources. [direct-source-comparison.json](evidence/followup/direct-source-comparison.json) records the byte comparison. The optional anchored checker additionally passes all 20 original inputs under the original contribution pin and rejects a consistent manifest/source substitution accepted by the portable old replay; [provenance-control.json](evidence/followup/provenance-control.json) records that deliberately different trust boundary. New-publication comparison is part of the committed-byte check below. + +[scientific-inventory.json](evidence/followup/scientific-inventory.json) records the hashes of scientific programs, frozen inputs, RESULT and argument/review-boundary documents. It excludes subsequent publication receipts and validation records to avoid self-referential hashes; it is an integrity record, not an external trust anchor. Three reviewed wording fixes after the first integrated run did not change programs, inputs or RESULT. Original logs and FINAL-REVIEW.md remain historical snapshots and are not relabelled as review of the new checker. + +## Committed-byte and export procedure + +The publication process runs these commands on the exact committed scientific snapshot and repeats them on the final evidence commit: + +```sh +python3 -B scripts/check_research_commit.py --package research/pr6551-retirement-audit --check-source-history --check-modes +python3 -I -B research/pr6551-retirement-audit/check_source_objects.py --repository . --publication-commit FULL_TESTED_COMMIT +python3 -B scripts/export_research_package.py export --package research/pr6551-retirement-audit --ref FULL_TESTED_COMMIT --output /tmp/retirement-audit.tar +python3 -B scripts/export_research_package.py check /tmp/retirement-audit.tar +``` + +After integrity checking, extraction occurs outside the checkout. Every exported file is made read-only, Git metadata is absent, PATH is empty, and both direct isolated Python modes run from the exported root. Full file hashes and modes are compared before/after. The frozen scientific inventory must still match. An evidence commit can retain receipts for its parent but cannot embed its own commit identifier; each retained receipt must therefore name its actual tested commit, and the final exact-head check is reported separately in the PR publication record. No claim of final-head evidence is inferred from a parent's receipt alone. + +## Review, continuity and limits + +The final code review and documentation review are separate AI roles using the same GPT-6 Astra model family. Findings and resolutions are preserved in FOLLOWUP-REVIEW.md and DOCUMENTATION-REVIEW.md. The external Claude review applies to the earlier snapshot only. Shared opcode, physical tariff and Python rational-arithmetic assumptions remain; agreement is not a proof of those assumptions. + +Canonical upstream was refreshed at the time recorded in [context-refresh.json](evidence/followup/context-refresh.json) and remained `7c49fe32c55de02c2c873655c804b16e4c3fef0a`, priority revision 8. All three predecessor heads were unchanged. The bounded live overlap refresh found no competing exact audit; its stated partial coverage is not evidence of novelty. + +No native/all-width physical replay, physical ordinary-rate proof, numerical crossing, completion-tail, compiler or outer replay, selected-construction replay, historical full suite, Lean or Docker run was needed or performed. All five framework assumptions remain. Remote CI and draft/maintainer status must be reported separately. Next useful work remains a separately derived rational audit of the changed completion-tail parameters. + +## Recorded committed snapshot + +The expanded scientific snapshot is `092009c8548a520c9913eb905f351c7ce2a4b8cd`. Its committed-byte/history/mode preflight passed on all 75 package files. The optional publication-anchored source check passed on all 22 retained inputs. Export integrity passed; both isolated normal/optimized cold replays reproduced RESULT exactly, and all 76 exported file hashes and modes stayed unchanged. All 37 scientific-inventory entries matched. See [publication-checks.json](evidence/followup/publication-checks.json), [export-inventory.json](evidence/followup/export-inventory.json) and the corresponding `.txt` logs in [evidence/followup](evidence/followup). + +The following evidence commit adds this paragraph and those receipt/log files only. It leaves all scientific-inventory entries unchanged. Its own exact-head preflight, source comparison, export integrity and read-only replays are repeated after committing, and their actual SHA/count/results are reported in the PR body. The 75-file receipt above remains explicitly a receipt for `092009c8548a520c9913eb905f351c7ce2a4b8cd`; it is not relabelled as a receipt for the evidence commit. diff --git a/research/pr6551-retirement-audit/LICENSE b/research/pr6551-retirement-audit/LICENSE new file mode 100644 index 0000000000..261eeb9e9f --- /dev/null +++ b/research/pr6551-retirement-audit/LICENSE @@ -0,0 +1,201 @@ + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/research/pr6551-retirement-audit/NOTICE b/research/pr6551-retirement-audit/NOTICE new file mode 100644 index 0000000000..b2ec07d122 --- /dev/null +++ b/research/pr6551-retirement-audit/NOTICE @@ -0,0 +1,6 @@ +Independent PR6551 retirement-cut and invoice audit +Prepared for DLamdata with substantial OpenAI Codex assistance, including separate native AI accounting and skeptical-review roles. Apache-2.0. + +This package adds independent finite checks and a scoped review; it does not claim the predecessor construction or its gain as new. +Ali Romman retains credit for PR6551, PR6545, PR6398 and integration; Cinna Davis and the PR6348 binary/physical caller lineage, PR5599/5602/5611/5615 handoff/basis work, PR6529 native source and PR6524 replay services retain their credit. +Exact source NOTICE files and the nested source archive retain the complete contributor and licensing notices, including Jacob Alan Sussman, DreamingOfClouds, Rohan Arun, Erwan L. and all preceding authors. diff --git a/research/pr6551-retirement-audit/OVERLAP.json b/research/pr6551-retirement-audit/OVERLAP.json new file mode 100644 index 0000000000..4a73df7c3c --- /dev/null +++ b/research/pr6551-retirement-audit/OVERLAP.json @@ -0,0 +1,149 @@ +{ + "searched_at": "2026-11-03T08:33:26.041140+00:00", + "terms": [ + "retirement-rebalanced-query-tree", + "6551", + "6545" + ], + "coverage": { + "local": { + "head": "7c49fe32c55de02c2c873655c804b16e4c3fef0a", + "records": 8917, + "head_records": 8917, + "complete_head_inventory": true, + "unsearched_head_paths": [] + }, + "open_prs": { + "fetched_at": "2026-11-03T08:31:20.413465+00:00", + "complete_pagination": true, + "records": 504 + }, + "discussions": { + "fetched_at": "2026-11-03T08:33:25.542311+00:00", + "complete_pagination": true, + "records": 6491 + } + }, + "warnings": [], + "limits": "Remote PR scope is open PRs including drafts. Closed work is searched through local READMEs, TASK metadata and ledger only. Local coverage compares searched working-tree notes with regular note paths at coverage.local.head; it does not certify HEAD contents or historical coverage. No code diffs or arbitrary package files are searched. Recheck relevant heads before submission.", + "relevant_matches": [ + { + "number": 1166, + "title": "Handoffs and requests for independent checks", + "kind": "discussion-comment", + "url": "https://github.com/CrocSwap/integer-mult-bounds/discussions/1166#discussioncomment-19367661", + "updated_at": "2026-11-03T06:09:28Z", + "matched_terms": [ + "retirement-rebalanced-query-tree", + "6551", + "6545" + ], + "score": 3, + "excerpt": "HANDOFF / new complete conditional kappa improvement: [PR6551](https://github.com/CrocSwap/integer-mult-bounds/pull/6551), head `11ca568cd1a16776ade02b4df09ea04a784cc674`.\n\nKappa rises from PR6545's .00329394 to **.00329492**, an incremental **0.0297516%** gain. Keep its published retirement and PR6529 native word; change child sizes (7,2,2,7) to (7,2,5,4). The exact complete signed gain is `208*(H41-H40-H57+H56)`, positive for every power tariff with 0