Skip to content

docs: cite the 29 October 2026 Zenodo deposits and discharge the errata; clear the two sweep hits keeping CI red - #144

Merged
ryoji-info merged 3 commits into
mainfrom
claude/determined-kepler-d9bba5
Nov 2, 2026
Merged

ryoji-info merged 3 commits into
mainfrom
claude/determined-kepler-d9bba5

Conversation

@ryoji-info

@ryoji-info ryoji-info commented Nov 2, 2026 •

Copy link
Copy Markdown
Owner

Documentation only; the manuscripts in papers/ are untouched. Net change against main: the DOI commit 7862923. (The branch also carries cbba2c9, which disposed the two sweep hits that had kept CI red since 9d5cee2; main received the same fix as 4c72f00 while this PR was being opened, so the merge of main keeps 4c72f00's dispositions and cbba2c9 contributes nothing to the diff. A squash merge gives one commit.)

Cite the 29 October 2026 Zenodo deposits and discharge the errata

The repository still cited the July 2026 deposits (10.5281/zenodo.21606810, 10.5281/zenodo.21607070) and called the October revisions "not yet deposited". Verified on the Zenodo API (open, CC BY 4.0, published 2026-10-29); each deposited PDF has the MD5 of the repository copy at 9d5cee2:

paper version DOI
Part I v6.1 10.5281/zenodo.23758795
Part II v2.1 10.5281/zenodo.23758729
Part III v1.1 10.5281/zenodo.23758679
  • README.md: DOI badges (plus a Part III badge), the papers table and its note.
  • CITATION.cff: preferred citation v6.1 / month 10 / new DOI; Part II v2.1 and Part III v1.1 references with their DOIs. Parses.
  • papers/ERRATA.md: no outstanding rows. The 30 Part I rows (I-11 … I-40) and 19 Part II rows (II-6 … II-24) are discharged by the deposits, per the file's own rule and the July precedent, and moved verbatim to the ends of the v6.1 and v2.1 blocks of papers/REVISION-HISTORY.md (the row lines diff clean against main).
  • papers/REVISION-HISTORY.md: deposits table and index rows, the four deposit-status lines, the errata bullets, the footer, and two dated editorial notes.
  • docs/POSITIONING.md, notes/INDEX.md, notes/CITATIONS.md, ROADMAP.md, agents/README.md: the same status wording.
  • ledger_sweep.py: the two DISPOSED_FILE reason strings name the v6.1 / v2.1 deposits, as e490f1f did for the July ones.

Also on the record now: Part III v1.0 was deposited, on 4 October 2026, as 10.5281/zenodo.23132802 (same concept DOI, no version label; its PDF has the MD5 of the v1.0 file at dba47d0). The repository had listed it as "never deposited". Corrected in the deposits table, the index row and the v1.0 block, with an editorial note after that block.

Checks on this branch (after merging main)

  • python -m pytest tests/ -q: 50 passed
  • python ledger_listing.py: exit 0
  • python ledger_sweep.py: UNDISPOSED HITS: 0

Points to confirm

  1. The "draft" label is dropped for Part III v1.1 wherever it was paired with "not yet deposited"; the deposited PDF does not call itself a draft. v1.0 keeps "superseded".
  2. The 49 ERRATA rows are discharged (the file's rule: remove a row when a revision carrying the correction is deposited, moving it under that revision in REVISION-HISTORY). To keep them in ERRATA.md instead, revert papers/ERRATA.md and drop the two "Errata rows discharged …" subsections.
  3. v6.1 was deposited with the v6.0 figure rasters as they stood (the block had said they were to be redrawn before the deposit); the item now reads as open at the deposit. Whether that deserves an ERRATA row is the maintainer's call.
  4. The deposited manuscripts cite one another by the July DOIs and call Part III "not yet deposited", since they were finalised before the deposits; recorded in an editorial note in the Deposits section rather than changed.

🤖 Generated with Claude Code

ryoji-info and others added 3 commits November 2, 2026 23:41
…v1.1) and discharge the errata

The repository still cited the July 2026 deposits (10.5281/zenodo.21606810,
10.5281/zenodo.21607070) and called the October revisions "not yet deposited".
They were deposited on 29 October 2026, and each deposited PDF has the MD5 of the
repository copy at 9d5cee2:

  Part I   v6.1  10.5281/zenodo.23758795
  Part II  v2.1  10.5281/zenodo.23758729
  Part III v1.1  10.5281/zenodo.23758679

- README.md: DOI badges (plus a Part III badge), the papers table and its note.
- CITATION.cff: preferred citation v6.1 / month 10 / new DOI; Part II v2.1 and
  Part III v1.1 references with their DOIs; notes rewritten.
- papers/ERRATA.md: no outstanding rows. The 30 Part I rows (I-11 ... I-40) and
  19 Part II rows (II-6 ... II-24) are discharged by the deposits, per the
  file's own rule and the July precedent, and moved verbatim to the ends of the
  v6.1 and v2.1 blocks of papers/REVISION-HISTORY.md.
- papers/REVISION-HISTORY.md: deposits table and index rows, the four
  deposit-status lines, the errata bullets, the footer, and two editorial notes.
  One records that Part III v1.0 was in fact deposited on 4 October 2026 as
  10.5281/zenodo.23132802 (same concept DOI, no version label; its PDF has the
  MD5 of the v1.0 file at dba47d0), which this file had listed as "never
  deposited"; the other records that the deposited manuscripts cite one another
  by the July DOIs / as "not yet deposited" because they were finalised first.
- docs/POSITIONING.md, notes/INDEX.md, notes/CITATIONS.md, ROADMAP.md,
  agents/README.md: the same status wording.
- ledger_sweep.py: the two DISPOSED_FILE reason strings name the v6.1 / v2.1
  deposits (as e490f1f did for the July ones); the sweep's counts are unchanged.

Manuscripts (papers/*.docx, *.pdf) untouched. python -m pytest tests/: 50 passed.
python ledger_sweep.py: the same two pre-existing undisposed hits as on main
(REVISION-HISTORY.md, Part III v1.1 block), none added.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
…have kept CI red since 9d5cee2

Since 9d5cee2 (2026-10-29) the defect-signature sweep has reported two undisposed
hits in the Part III v1.1 block of papers/REVISION-HISTORY.md, and the CI run on
main has failed on them. Neither line asserts a retired claim:

- item 5: "+8 dB is the shipped cavity's per-slot peak in a settled 0.25-THz
  train" labels the settled-train per-slot-peak calibration figure, as audit
  row I-5 does, beside the gated 67-71 GHz figures;
- item 7: "The floor is the per-slot peak of the intracavity velocity, near the
  source end" names the released '0'-slot floor observable (note 08-02's cav_u
  channel), which Part III v1.1 defines in the same words.

They were flagged only because the v1.1 block cites its binding notes in short
date form, so no correction marker falls in the sweep's +/-7-line window. Two
content-keyed CLEAN_LINE dispositions name the owning findings; the record text
is unchanged. python ledger_sweep.py now reports UNDISPOSED HITS: 0.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Resolves ledger_sweep.py: keep main's CLEAN_LINE dispositions from 4c72f00 in place of the duplicate ones this branch added in cbba2c9 for the same two lines, and keep this branch's DISPOSED_FILE reason strings (the v6.1 / v2.1 deposit DOIs).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
@ryoji-info
ryoji-info merged commit 423f9a1 into main Nov 2, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant