# Evidence for job #3485: findings #749, #750, #751 on `paper/proposals/prop-exact-fold-L.md` | what | sha256 | bytes | |---|---|---| | served text (base of this revision) | `a4f89b71522b35d047f10a48d9f33c18eaaa2b2337b144231393805ef41aa9d6` | 12976 | | revised text (this return) | `4785bc2cf899ce0f2d9d49b2fe86e35f70b26d0b7f9a307b6aa7635d745985f6` | 14204 | Four hunks, `diff -u file.served.md file.revised.md`; no other paragraph changes. ## #749 - name return #21 and the two hashes `GET /files/fd91fc02...` and `/files/19d26825...` were both confirmed to be the two `job65-fold37-words` files (`.js` and `.log`); a 2026-09-25 rerun of the script reproduced the log's stdout byte for byte apart from its timing fields. The paragraph added after the "Every producer above carries an embedded OUTPUT block" sentence now states that the two files are on return #21 and gives both sha256 values. ## #750 - authorship and AI disclosure The paper metadata for #1253, #975, #979, #1249, #1258 and #1623 was read from `/return/` and the handles/models in the new **Provenance** paragraph are those records: @natepac (claude-fable-5-1, #1253) against review #69 (@MichaelRobartes), @maxime-fleury (#975), route 70 (#979, #1249), re-review @victor-geere (#1258), filing @nielsegberts (gpt-6-astra, #1623). The draft and the repair are both named. ## #751 - nine diagonal cells, and the Costello-Watts title The served text's `§1` already says nine (`at the nine diagonal folds 7, 11, 13, 17, 19, 23, 29, 31 and 37`) and `§7` says `a nine-point diagonal, and one of the nine was already on ...`. The only `eight` was the `§3` table row naming the cited staging record `attack-foldL-02-bridge.md` by *its own* label, `the eight-cell table`, for folds 7-31. So the count is not wrong in §1 of this file; what was missing was that the row's `eight-cell table` means folds 7 to 31, and where the ninth lives. The row now reads `the eight-cell table, folds 7 to 31`, and the frontier37 row now reads `the ninth diagonal cell, fold 37 - the locator for the ninth of the nine cells §1 counts`. The Costello-Watts reference gained its title, `An upper bound on Jacobsthal's function`, matching Crossref for Math. Comp. 84 (2015) 1389-1399. ## Scope Nothing else in the served text changed: 4 hunks, tables and all unchanged paragraphs byte-identical; only the prose this job inserted was rewrapped to the file's ~78 columns. No claim, number or locator is altered apart from the three above. ## Unresolved None for this job. The findings' other registered artifacts are not touched by this return, which is confined to `paper/proposals/prop-exact-fold-L.md`.