diff --git a/docs/P15_SPLIT.md b/docs/P15_SPLIT.md new file mode 100644 index 000000000..51dfe51db --- /dev/null +++ b/docs/P15_SPLIT.md @@ -0,0 +1,9 @@ +# P15 is not part of the Gaussian-field paper + +Scientific effect: NONE. + +The full-price theorem for a realized disjoint capacity/clutter family, with demands at least 2 and sharp factor 1/(3 - log(3e - 2)), is a combinatorics note. Demand one is an obstruction to the unrestricted extension. It is not a solution of an unrestricted prize problem, and it is not a Gaussian-field theorem. + +Read the [full public argument and discussion](https://github.com/d6g8k5htny-coder/main/issues/59), with its structural hypotheses. P15 and SIDE24 have separate mathematical scopes. + +The [public mathematics catalog](PUBLIC_MATHEMATICS.md) provides a separate P15 route alongside the Gaussian-field research. The existing proof and review links remain available there. diff --git a/docs/PUBLIC_MATHEMATICS.md b/docs/PUBLIC_MATHEMATICS.md index b16646c7c..7cc7d8db6 100644 --- a/docs/PUBLIC_MATHEMATICS.md +++ b/docs/PUBLIC_MATHEMATICS.md @@ -11,7 +11,9 @@ The full mathematical statements, proof texts, partial arguments, counterexample | Current proof index | [Math proof availability](https://github.com/d6g8k5htny-coder/Math-/blob/main/PROOF_INDEX.md), including reviewed scopes and missing arguments | | Marked-cylinder geometry | [Full proof](https://github.com/d6g8k5htny-coder/Math-/blob/1e1114f5eb8ef8cdbcde591bf74126274c250f88/imports/lifetime_parent_20260925/MARKED_CYLINDER_CAP_PROOF.md) | | Matrix cap and parent lifetime theorem | [Full proof](https://github.com/d6g8k5htny-coder/Math-/blob/1e1114f5eb8ef8cdbcde591bf74126274c250f88/imports/lifetime_parent_20260925/UNIFORM_MATRIX_CAP_AND_LIFETIME.md); [current review and reopening](https://github.com/d6g8k5htny-coder/main/issues/63) | -| SIDE24 coefficients, lifetime remainder, RN regional bounds, and P15 | [Topic guide](RESEARCH_INDEX.md); [all current candidate and review directories](https://github.com/d6g8k5htny-coder/Math-/tree/main) | +| SIDE24 coefficients, lifetime remainder, and RN regional bounds | [Topic guide](RESEARCH_INDEX.md); [all current candidate and review directories](https://github.com/d6g8k5htny-coder/Math-/tree/main) | +| Combinatorics: P15 capacity and clutter | [Scope and reading route](P15_SPLIT.md); [full public argument](https://github.com/d6g8k5htny-coder/main/issues/59). Not a Gaussian-field theorem. | +| Short-bar histogram protocol | [Simulation protocol](SIMULATION_PROTOCOL.md). A figure is a check, not a proof. The [experiment note](research-translation/20260930/EXPERIMENT.md) is the authority if the protocol and experiment note disagree. | | Recovered SIDE24 theorem chain | [Full theorem package](https://github.com/d6g8k5htny-coder/main/blob/2f7a5a9f10c9ed5f5b7792a8f2521318d9208532/research/side24/source_recovery/SIDE24_THEOREM_PACKAGE_v1.0.md), [full gap-fill supplement](https://github.com/d6g8k5htny-coder/main/blob/2f7a5a9f10c9ed5f5b7792a8f2521318d9208532/research/side24/source_recovery/SIDE24_GAP_FILL_SUPPLEMENT.md), and [source-recovery / eigenfloor provenance](https://github.com/d6g8k5htny-coder/main/blob/2f7a5a9f10c9ed5f5b7792a8f2521318d9208532/research/side24/source_recovery/README.md); exact historical V3.3 table remains unrecovered | | SARD-G chart and slicing repair | [Strict-gradient successor](https://github.com/d6g8k5htny-coder/main/blob/a1fc9581291b6f81c76c903c96a13955b51f8d54/research/q0/sard_g/SARD_G_EXECUTION_REPAIR_20260925.md), [full successor A1/A6 review and section-endpoint counterexample](../reviews/sard_g_successor_a1_a6_20260926/REVIEW.md), [earlier equality-boundary review](../reviews/sard_g_a1_a6_20260926/REVIEW.md), and [current discussion](https://github.com/d6g8k5htny-coder/main/pull/122); A1 remains AMEND as written and A6 remains conditional | | SIDE24 replay archive and arithmetic sources | [Complete 12.5 MB replay archive](https://github.com/d6g8k5htny-coder/main/blob/2992819828dfe8a23e682816a8c10eab5861f94d/research/side24/source_recovery/custody/SIDE24_V3_4_AUDIT_REPLAY_FULL_ARCHIVE.zip), [both exact arithmetic scripts](https://github.com/d6g8k5htny-coder/main/tree/2992819828dfe8a23e682816a8c10eab5861f94d/research/side24/source_recovery/custody/arithmetic_ledgers), and [source identities / missing-object record](https://github.com/d6g8k5htny-coder/main/blob/2992819828dfe8a23e682816a8c10eab5861f94d/research/side24/source_recovery/README.md); public PR126 source, with historical V3.3 table and exact Module I extraction still unrecovered | diff --git a/docs/READER_SPLIT.md b/docs/READER_SPLIT.md new file mode 100644 index 000000000..7a3ac8b96 --- /dev/null +++ b/docs/READER_SPLIT.md @@ -0,0 +1,9 @@ +# Reader path split + +Scientific effect: NONE. + +Start with the [public mathematics catalog](PUBLIC_MATHEMATICS.md). It provides separate routes to Gaussian-field research and the [P15 combinatorics note](P15_SPLIT.md). + +The [short-bar histogram protocol](SIMULATION_PROTOCOL.md) summarizes the proposed numerical check. The [experiment note](research-translation/20260930/EXPERIMENT.md) is the authority if the two disagree. + +This page is a navigation aid. It does not accept a theorem and does not close pairing. diff --git a/docs/SIMULATION_PROTOCOL.md b/docs/SIMULATION_PROTOCOL.md new file mode 100644 index 000000000..6a55e274f --- /dev/null +++ b/docs/SIMULATION_PROTOCOL.md @@ -0,0 +1,35 @@ +# Bargmann-Fock short-bar histogram + +Scientific effect: NONE. A figure is a check, not a proof. This summary does not replace the [experiment note](research-translation/20260930/EXPERIMENT.md). If they disagree, the experiment note is the authority. + +## Generator + +Variance-one periodized Bargmann-Fock on the flat torus of side 24, dimension 2. Covariance is the normalized image sum + +K_24(z) = sum_n exp(-|z+24n|^2/2) / sum_n exp(-|24n|^2/2). + +A different width, a minimum-image kernel, or a differently normalized truncation is a different field. The SIDE24 enclosure does not apply to it. + +## Filtration + +Cubical superlevel sets on a periodic cubical complex: opposite faces are glued. An ordinary square with boundary is not the torus. Record strictly positive finite nonessential H0 bars only. Count zero-length and tied pairs separately. Do not put them in the logarithm. + +## What is being estimated + +Let mu(A) = 24^{-2} E[number of those bars with lifetime in A]. The candidate law is about the density dmu/dell ~ c ell^{-1/3}. The count of bars shorter than ell is the cumulative mu((0, ell]), which is order ell^{2/3} if the density law holds. + +Normalize a bin (a, b) by the number of fields, the area 24^2, and the bin width. Or compare the bin count with the integrated prediction (3c/2)(b^{2/3} - a^{2/3}). Raw counts are not comparable to c. + +## Fit + +Fit log(density) against log(ell). The candidate slope is -1/3. Do not fit against (-1/3) log(ell) and then expect slope -1/3; that transformed predictor has slope 1 if the law holds. + +A lifetime is a field value. Grid spacing is a length. Convert the mesh to a filtration error before comparing it with a bin. A window that is large against the grid and small against the correlation length is not, by itself, a resolved window. + +## What a mismatch means + +A stable slope or prefactor discrepancy is a diagnostic. Check sign, volume and amplitude normalization, the exact covariance, essential and zero bars, spectral cutoff, spatial approximation, and bin integration. Report sampling uncertainty across fields separately from spatial and spectral error. If those errors cannot be controlled, write inconclusive at the tested resolutions. A histogram mismatch does not by itself isolate failure of elder pairing or of the continuum asymptotic. + +## Reading path + +Return to the [public mathematics catalog](PUBLIC_MATHEMATICS.md) for the Gaussian-field proof sources and their stated scopes. The [P15 combinatorics note](P15_SPLIT.md) has a separate reading route.