# CHANGELOG entries, wave 3 partition I (the flagship split) *(Staging file. The parent merges these into `research/history/CHANGELOG.md` in one pass; appliers never edit that file directly. House form: newest first, grouped by the document changed, each entry naming the claim as it stood, what replaced it, and the artifact that forced the change.)* --- ## 2026-08-17: the flagship's wall chapter becomes a companion note ### paper/wall-note.md (new) **Created as the child of `paper/moire-primes.md` §7 and §7A**, on the split approved in the consistency campaign (`qc-CAMPAIGN.md`:337-345: Chris permits breaking papers into sub-papers, with bidirectional crosslinks as the condition). It holds the five door paragraphs and the four face paragraphs at full grain, 362 lines, and the flagship falls from 1,097 to 778. The condition of the split was that no inbound pointer breaks. Both headings stayed in the flagship at their own numbers, with their text unchanged, so every document that cites `paper/moire-primes.md` §7 or §7A still resolves; `node research/qc.js` reports `refs` and `crosslinks` at zero before and after. Every moved block was word-diffed against its original. The only differences the diffs found are four bare section references made explicit, since `(§8, …)`, `(§4)` and `(§9)` twice would have read as references to the note's own sections once moved. No constant, calibration marker, level stamp or script citation changed. Door 2's *Retracted, and left visible* block, Door 5's expert-read note and Face 3's double self-reversal on K* all travelled byte-identical. ### paper/moire-primes.md §7 **"What this project adds is a *surveyed perimeter*: five independent routes, each rigorous until its last step"** had already been retired in wave 2. The paragraph that replaced it, the door-failure map, is now compressed into the stub and gains the five door names, which it did not carry: Door 1 is Legendre's budget (3ⁿ), Door 2 the Fourier budget (2ⁿ), Door 3 the moment ceiling, Door 4 the removal ledger, Door 5 coverings and constructions. All eleven content items of the retired paragraph survive, audited one by one in `research/history/staging/applied-I.md`. A flagship reader still learns that there are five doors, what each is called, that Doors 1 and 4 are one classical object at two truncation depths, that Doors 1, 3 and 5 run to a final inequality, that Door 2 fails structurally at x = 11 and that Door 4 reverses on the numbers before any limit is taken. **The section's two-moiré conclusion and the blockquote, "The Scour never achieves perfect local alignment with the Grain", stayed.** It is the paper's native-form restatement of the conjecture, promised by name in the abstract and echoed in the coda, and a flagship whose body never states its own headline reformulation would be a shorter paper and a worse one. ### paper/moire-primes.md §7A **Stubbed to the section's own opening parenthetical and its own closing summary**, both verbatim but for four inserted face names and three words forced by the move ("Each face below" lost "below"; calibration is marked "throughout the note"; the reversals "stay visible there"). The summary already stated in one clause each what every face establishes, so no paraphrase was written to replace it. A flagship reader still learns that there are four faces and what each is called: Face 1 the anchored bias, Face 2 the overlap channel, Face 3 certificate depth, Face 4 the exponent. ### paper/moire-primes.md §8 **"the real tile is consistently *quieter* or luckier than its ensemble in four places"**, which stood at `:910`. Replaced by the surviving count and the measured range: the tile is quieter than its ensemble on the per-prime strike variance, at 0.34 to 0.62× ensemble, and the campaign's other three sightings dissolved under exact control. `research/NATAL-CAP-CAMPAIGN.md`:77-83 is headed **"THE ANCHORED CALM — one sighting, not four"** and lists what each of the three turned out to be: the sub-random house splits belong to every rotation rather than to the anchor, the sub-CRT pair and triple overlaps are the dead origin plus short-window arithmetic, and the drift toward 0.793·E is the anchored bias β, a separate object. The flagship was claiming four surviving sightings where one survives. This is the unlanded half of `applied-F.md`'s handoff C-5. Wave 2 applied the pointer half, adding `research/anchored-calm.md` beside the campaign scoreboard, and left the count. **`(§7A, Face 2)`**, the pointer for corr(VR, S) ≈ 0 at all three exactly enumerated levels, is now ``(`paper/wall-note.md` §2, Face 2)``, because that measurement moved. --- ## Checked and found already correct, recorded so the work is not repeated Three defects the campaign carried as open inside this file were verified at their home documents and needed no edit, all three having landed in wave 2. **Door 5 and the distinct-moduli hypothesis (`qc-papers2` P-11).** The door states that Hough (2015) and Balister–Bollobás–Morris–Sahasrabudhe–Tiba (2022) require distinct moduli, that our classes `{0, −2 mod q}` use each modulus twice so neither theorem applies to our object, and cites Klein–Koukoulopoulos–Lemieux (2024) Theorem 3 as the theorem that does cover multiplicity 2, **with no numeric constant available at s = 2**. Verified against `research/covering-dive.md`:91-92, the home, which carries Theorem 3 verified from PDF and marks the translation to our setting `[INFERRED]`; the paper reproduces both the bound and the marker, and names the gap that matters, that KKL says a multiplicity-2 system cannot cover ℤ and says nothing about covering a finite interval, which is all a zone is. **Face 4 and Lemma V (`qc-papers2` P-3, `qc-status` A-1).** "Missing exactly one ingredient" is gone. The text says that at the working point the ladder measures, s/u ≈ 1.2 sits outside the range s ≤ u that Lemma V is stated in, so the operative assumption is a Gaussian maximal law for the sawtooth and that maximal inequality is the whole price of the route. Verified against `research/sift-limit-attack.md`:47-53, which declares itself the home and the authority for that distinction. **§8 against Face 2 on the anchored calm (`qc-papers2` S-3).** The two sections describe the same object and now agree: the mechanism is proven, the anchor's two strike classes being mirror-adjacent and gluing into a single cyclic window, and the calm is uncorrelated with survival. Verified against `research/anchored-calm.md`, which grades the mechanism PROVEN at all x and all q and says in terms that the calm is not a route to the conjecture and must not be read as one.