Skip to content

spec(physics): formula_discovery -- every target is a cited measurement, every fit carries its chance - #10301

Merged
gHashTag merged 1 commit into
masterfrom
spec/formula-discovery-measured-10297
Oct 18, 2026
Merged

spec(physics): formula_discovery -- every target is a cited measurement, every fit carries its chance#10301
gHashTag merged 1 commit into
masterfrom
spec/formula-discovery-measured-10297

Conversation

@gHashTag

@gHashTag gHashTag commented Oct 18, 2026

Copy link
Copy Markdown
Owner

Closes #10297

specs/physics/formula_discovery.t27 did not compile (ln, pow, powf, given/when/then), so none of its tests had ever run. Read by eye they were circular: S1_gamma against phi^-3 itself, N1_alpha_s against its own output, PMNS2 against an unsourced 0.547.

What changed

  • Ported to current t27 and built on physics::sacred_lattice (chance, label, lattice_size, RHO_C6).
  • Every target is now the cited measurement with its uncertainty (CODATA 2022, PDG 2024, Planck 2018, HotQCD), the source written beside the constant. Asymmetric errors use the side that matters.
  • Each formula gets a verdict: REFUTED beyond 3 sigma; UNKNOWN when outside the grammar n * 3^k * pi^m * phi^p * e^q or when no measurement exists; otherwise label(resolution, rho_at(C), 19). Density is scaled by lattice size from RHO_C6 and reproduces the lab null rates at C=4 (22.5%) and C=8 (93.6%).
  • Nothing is labelled EXACT, MEASURED or SUGGESTED by fit.

What the measurements say

  • Refuted (more than 3 sigma): CKM1 theta_C (sin 8.59 deg = 0.149 vs V_us 0.2243), CKM2 V_cb (7.5e-4 vs 0.0411), PMNS4 sin^2 th12 (2.9e-4 vs 0.307).
  • 2 to 3 sigma off, kept: P10 V_ud, P12 V_td.
  • Agree, but the measurement is too coarse to tell the formula from chance (UNKNOWN): N1 alpha_s, PMNS2, PMNS3/P14, P11, P13, P15, P16, P17, P18.
  • H1 m_H/m_Z agrees to 0.09% but 1/8 is outside the grammar: UNKNOWN. PM1b matches CODATA 1/alpha within 1 sigma, also outside the grammar.
  • S1_gamma has no measurement (a theory input); N1 = S1 / 2 is the same fit twice; P10 and P11 are one expression for two quantities.

Verification

Run on the Railway lab: 23 tests pass, 0 fail, 1 invariant proved, 0 vacuous. Seals refreshed with tri seal --save.

🤖 Generated with Claude Code

@gHashTag @claude
spec(physics): formula_discovery -- every target is a cited measureme…
7a0b0fb 
…nt, every fit carries its chance (Closes #10297)

The spec did not compile (ln, pow, powf, given/when/then), so no test had
ever run. Read by eye they were circular: S1_gamma against phi^-3 itself,
N1_alpha_s against its own output, PMNS2 against an unsourced 0.547.

Ported to current t27 and rebuilt on physics::sacred_lattice. Each formula is
compared with its CODATA 2022 / PDG 2024 / Planck 2018 / HotQCD value and
uncertainty, then judged: REFUTED beyond 3 sigma, UNKNOWN when outside the
grammar or at the chance rate, never EXACT. Density is scaled by lattice size
from RHO_C6 and reproduces the lab null rates at C=4 and C=8.

Refuted: CKM1 theta_C, CKM2 V_cb, PMNS4 sin^2 th12. 23 tests pass on the
Railway lab, 0 vacuous. Seals refreshed.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@github-actions

github-actions Bot commented Oct 18, 2026

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-18 08:03:10 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 49
PRs with All Checks Green 1
READY 0
FAILING 49
PENDING 0
NO CHECKS YET 0

These columns do not partition: 0 + 49 + 0 + 0 = 49, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=0f8c66b9ed0e != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@gHashTag
gHashTag enabled auto-merge (squash) October 18, 2026 08:04
@github-actions github-actions Bot mentioned this pull request Oct 18, 2026
Merged
@gHashTag
gHashTag merged commit 4b384b6 into master Oct 18, 2026
28 of 31 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

physics: formula_discovery.t27 -- replace circular tests with measured targets and a chance label

1 participant

@gHashTag