Response to the independent review (REVIEW-SR-ASTROPHYSICS-2026-07)
CORRECTION (2026-07-14). The post-revision re-review demonstrated that several Done labels below (M3, M6, M8, M10, M11, M12) were premature: the intended corrections had been applied at some sites while the old claims survived verbatim at others (Parts 1/2/4 reconnection language, V1’s two-radius sentence, the README’s part count and inversion, Part 8’s spiral count). Those residues are now fixed and the guard rewritten so this failure mode is structurally caught — see
RESPONSE-REREVIEW-2026-07-14.mdfor the current disposition of record. This file is kept as the historical response to the first review.
Point-by-point disposition of every finding. Done = implemented and verifiable at
the pointer; Partial = implemented with a stated residue; Deferred = not done,
with the reason and the owner of the decision recorded. Regression protection:
scripts/check_retracted_claims.py (since upgraded to invariant rules + a fixtures
self-test; see the correction note above).
Major findings
| # | Finding | Disposition |
|---|---|---|
| M1 | Closest prior art (Montgomery 1995; Barilari–Bossio–Franceschi) | Done. Article §6 “Relation to prior work” with explicit not-new list; abnormals upgraded from fine print (Lemma 1.3, §6); BBF verified against arXiv:2504.09274; refs added to article, D1, D3, D4. Residue: the §6 comparison is statement-level and says so — to be refined against the papers’ proofs before any journal submission. |
| M2 | Period theorem vs caustic conjecture conflated | Done. “Exact period-average law” everywhere (incl. Part 6 scoreboard cell); Conjecture A’s structure (steps proved/gap) in §3.4; R6 probe replaced “beam stops refocusing” with the measured band statement; even-multiplicity guard run subcritical (0 suspects at ε=0.45/0.48) and stated; M2.4 physical-operational scope now in Remark 3.0. |
| M3 | §4 non-sequiturs (two-radius, exact inversion, β=2, σ) | Done. Two-radius separation retracted (single combination at leading order — article, D5, Part 3); inversion labelled leading-order estimator; β=2 rescoped to the period average with the derived caustic −½ sign-flip prediction; σ language replaced by sensitivity bands; Prop 4.1 remainder honest at O(ε³). Since superseded in the good direction: caustic c₂ = 1−¾β now DERIVED (Prop 4.2, run_o6_caustic_c2.py). |
| M4 | Singular-point Q / blog volume language / D2 criterion | Done. D3 fine print (earlier pass); article Remark 2.6 + Prop 2.7 (what the estimator converges to); D2 rewritten: estimator confounding named, full-measure screen labelled “working rule, no converse claimed”. O1 (volume at the null) remains open and is leddered as such. |
| M5 | Smaller math issues (notation, codimension, bundle, dimensions) | Done. t_c = r_L F(θ₀,ε); g ≥ 0 convention with reflection argument; codimension 8 (×2 sites); ℝ-fibre vs U(1) hypothesis in Remark 1.2; dimensions/w↔q/m in Remark 3.0. |
| M6 | Detection vocabulary | Done. Three-level taxonomy (blind / independent local confirmation / tangent-cone consistency check) defined in article §0 + Part 4, applied in README rows and Parts 5/8; “none is blind detection” stated. |
| M7 | Solar validation below solar-physics standard | Partial by design. Immediate honesty items done (r3 cap/prose, finder-params in captions, LOS/45s provenance, T3 complexity-index demoted to candidate). The rebuild itself (SHARP/720s radial, trilinear census, ensembles, tracked cadence) is chartered with pre-registered verdict rules (PROGRAM-G1) and not executed; accordingly, no solar discovery claim is made anywhere in the current layer. Owner of execution: next program phase. |
| M8 | Null/fold ≠ reconnection; fold naming; boundary crossing | Done. “Fold in a real-data-anchored boundary-interpolation family” naming; no-boundary-crossing caveat on the fold-wall claim; reconnection language reserved for non-ideal evolution (Part 6 subtitle + body). |
| M9 | T96 retraction not propagated | Done. README row, S3 banner (SUPERSEDED), Part 7 scoreboard, PROGRAM.md; Cluster/MMS gate recorded in the S3 banner; no magnetospheric claim pending. |
| M10 | Jupiter overstatement | Done. l≤18 continuation wording (×5); 0.85 = conventional mapping surface, Lowes ≈ 0.81; census caveat → then the census RAN (J2): three nulls (third found, Part 8 corrected), converged at two resolutions; coefficient ensemble (model-difference proxy, honestly labelled): patch 40/40, null 33/40 + position box; bald-patch scoping fixed (M10.5). Residue: formal covariances remain unpublished — the proxy is the best available and is labelled as such. |
| M11 | Statistical language | Done. No σ theater; bands labelled numerical/fit sensitivity; §0 convention; model-uncertainty-dominates note. |
| M12 | Record vs publication layer | Done. Superseded banners (S1, S3), PROGRAM.md status header, Part 7 sync, stale-claims regression script; current layer = README + article ledger. “Seven parts” claim: not found anywhere in the tree (checked); treated as already-fixed or mislocated. |
Minor findings
| # | Finding | Disposition |
|---|---|---|
| 1 | “SR” in article abstract/theorems | Done (none present; posts define the abbreviation at first use). |
| 2 | Units / w ↔ q/m | Done (Remark 3.0). |
| 3 | Heisenberg claim planar-scoped | Done (Part 2 + article Cor. 2.4 table). |
| 4 | Contact vs quasi-contact | Done (Lemma 1.3; Part 2 planar note). |
| 5 | Null ≠ reconnection site | Done (M8). |
| 6 | Full solar figure provenance | Partial. Series names, dates, downsampling, sample-file provenance present (Parts 4–7); NOAA/HARP + pixel scale + disk position for every figure is mandated by G1a going forward; retrofitting every legacy caption is folded into the G1 phase. |
| 7 | Finder parameters beside null counts | Done (Part 4 estimator box; Part 5 R3 caption; Part 6 hunt caption; Part 8/J2 census parameters). |
| 8 | “Pre-registered” qualified | Done (self-registered sense defined; repeated in G1). |
| 9 | Data-class labels in captions | Done where load-bearing (Parts 5/6/7/8 captions state real / extrapolated / model-derived); full sweep of early-series figures rides with G1’s caption standard. |
| 10 | Split Part 6 into two posts | Deferred to the author. An editorial restructure of an interactive post with cross-references; flagged, not executed unilaterally. All correctness issues inside Part 6 are fixed in place. |
| 11 | Title/abstract conditional | Done (subtitle rewritten; caustic identification named as open in the callout and ledger). |
| 12 | Regression tests for retracted claims | Done (check_retracted_claims.py, 15 patterns, wired for manual/CI use). |
The review’s two “decisive frontiers”
- Prove or refute the exponential-profile conjugate-time identity. Not closed — but materially sharpened since the review: the identity is now known to be exponential-specific (V1 counterexample), its first-order caustic shift is derived (τ₁ = 2π sin θ₀), the β = 0 coincidence of caustic and period average is a theorem at second order (Prop 4.2), and the supercritical regime is mapped (R6). O2 remains the program’s top open problem, now with a much smaller haystack.
- One astrophysical test at external standard. Chartered (G1) with frozen verdict rules; Jupiter’s J2 census + ensemble executed the reviewable half (item 12) and corrected the program’s own claim in the process. No discovery claim is made until G1 runs.
New results produced by implementing the review
- Caustic profile law derived: c₂(β) = 1 − ¾β (arbitrary profiles, leading order).
- Conjecture A delimited: exponential-specific; supercritical band structure measured.
- Jupiter: third envelope null found by the cell-prefiltered census; prediction now carries a survival fraction under the stress ensemble and a sample position range.
- The launch-averaged statistic confirmed absent from a bounded guiding-centre literature search (O4 closed as a search record).