David/dotfiles
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

session 2026-06-13: Studium Steps 3-4 + chamber-library canonical_texts tier (176-source graduation); L1 baton back (Seb replied)

Browse Source
This commit is contained in:
David F Glidden
2026-06-13 10:23:39 +02:00
parent 605fb410b6
commit d31831b6db
4 changed files with 107 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
claude/memory/skill-harvest-register.md
+9
View File
@@ -143,3 +143,12 @@ The single place proposed skills live so they don't evaporate between sessions.
| **clean cruft at the SOURCE layer, not as a downstream transform** | feedback memory | When a source carries conversion cruft (EPUB footnote-links, image-scan embeds), clean it at the SOURCE — producing a new canonical file (new hash, re-anchor) — not as a chunk-time/read-time transform. A downstream strip makes `text_original` a *transform* of the file rather than a byte-faithful *slice*, silently violating the verbatim guarantee. Spec §5 names this the primary path; the steward's instinct ("attend to the source files") corrected my chunk-time-rule fallback. Technique note: when footnote *display* numbers restart per-section (so `[45]` recurs), pair/label by the globally-unique anchor id (`#nf K`/`#anf K`), not the visible number — Pandoc renumbers on render anyway. Governed cleaner built: `chamber-library/scripts/clean_epub_residue.py`. | `feedback-clean-at-source-not-downstream-transform.md` | **PROPOSED** |
*(Both earned in use, both recurring-shaped. The first generalizes today's all-four-indices-defective finding into a standing gate; the second captures the steward's layer/placement correction — his instincts about WHERE a thing belongs were load-bearing twice today, also worth holding.)*
## New proposals (2026-06-13 — Studium Steps 3-4 / chamber canonical-tier day, awaiting steward)
| Element | Kind | One-line | Where it lands | Status |
|---|---|---|---|---|
| **dry-run-first for bulk file operations** | verification-ladder entry | When a mutation touches many files at once (mass `git mv`, graduation, rename), build it **dry-run by default** and emit the full move-map + collision-guard + reference-scan (`git grep` each old basename) BEFORE `--apply`. Proven 3× today (`canonicalize_filenames.py`, `graduate_to_canonical.py`, and the graduation preview) — the steward approved each from the preview, and the preview caught the 7 stale hand-doc references + 0 collisions before anything moved. Kin to the byte-identical gate, generalized to file-tree moves. | `reference-verification-ladder.md` | **PROPOSED** |
| **Symmetria §3 flag: census-through-a-pattern** | Symmetria §3 flag (refines census-through-truncation) | A filter/regex used to *count* or *partition* a set can silently mis-match and the count reads as authoritative. Today `grep -iE 'LOG'` for meta-files matched "episteme**LOG**y"/"pheno­meno­**log**y"/"eco**log**y" → 49 false "meta" files (real = 4). Antidote: verify the pattern against known positives AND a known negative before trusting the partition; anchor patterns (`_LOG\.md$` not `LOG`). One step more specific than census-through-truncation (which is about *coverage* of the scan; this is about *correctness* of the matcher). | Symmetria §3 | **PROPOSED** |
*(Both genuine, both recurring-shaped. The first will pay off the next bulk corpus move — the 237 dirty files graduate incrementally, each a potential bulk op. The second caught a real false-count today and is a clean refinement of an existing flag.)*