David F GliddenandClaude Opus 4.8
3f9a89b00c
Basic Memory v0.21.6 first sync over the live memory dir (steward-authorized live-dir trial, Option A 2026-06-06): adds permalink: to frontmatter, refolds long YAML description lines, strips final newlines. Bodies untouched — verified via full diff classification. From this commit forward, any diff in claude/memory shows only what Basic Memory or the session writes. Trial design: MemPalace untouched as incumbent; git status check on this dir at every wrap; end-of-day evaluation (recall quality, sync robustness, rebuild-from-files, malformed-file behavior). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
3.2 KiB
name, description, type, originSessionId, permalink
| name | description | type | originSessionId | permalink |
|---|---|---|---|---|
| Catalogues in chamber-library are stale — walk the trees, not the catalogues | Steward correction 2026-04-26. Any catalogue or index file in chamber-library is presumed stale. For inventory, recall, or substrate-survey, walk the actual file tree. Do not rely on CHAMBER_CATALOGUE.md, README.md, or similar summary documents as authoritative. | feedback | 87dfe384-b862-4deb-9676-24e43046528f | claude-memory/feedback-catalogues-stale-walk-trees |
Catalogues are stale — walk the trees
Steward verbatim 2026-04-26: "that catalogue is stale (any catalogues you may come across will be stale)"
The rule
Any catalogue, index, or summary file inside chamber-library/ (or by extension any of the steward's working repositories) is presumed stale unless its mtime / content has been verified against the actual tree it describes. This includes but is not limited to:
chamber-library/converted_texts/CHAMBER_CATALOGUE.md(last updated 2026-03-28; the tree has grown substantially since)chamber-library/README.md(claims "305+ texts, 147+ philosophers" — likely undercount)- Any subfolder
README.mdfiles - Auto-generated indices in
indices/(currently empty but the principle holds)
For inventory, substrate-survey, or recall of "what is available," walk the actual directories with ls, find, or recursive read. The tree is the canonical record; the catalogue is a frozen snapshot.
Why
The chamber-library grows through deliberate steward-additions (Mauss 2026-04-07, Derrida 2026-04-05, observer-problem-sources 2026-04-09, Aldus Younger + Teall + Jaccottet:Handke:Lichtenberg/ 2026-04-21–25, etc.). Catalogue updates lag and may never catch up. Trusting the catalogue means missing the most recently filed substrate — exactly the material most likely to be load-bearing for current threads.
This is also a particular shape of the canonical-library-not-checked-first family: catalogue-as-shortcut over actual tree-walk. The catalogue is a kind of canonical-library check, but a degraded one.
How to apply
When the steward asks "what's available" or "have a look at chamber-library" or when surveying for cluster-substrate or humic-layer voices:
- Walk the relevant subtree directly (
ls,find -maxdepth N -type f, orGlob). - If a catalogue is open in context, treat it as a navigation hint at best — verify each pointer against the actual file before claiming presence.
- For "what's new since" questions, prefer
git log --stat --since=YYYY-MM-DDor directory mtime sort over reading any catalogue's "last updated" line. - If a catalogue is needed for some downstream use (e.g., reader-orientation), surface that the existing catalogue is stale and propose either a fresh walk-derived inventory or a generation script — don't quietly re-publish stale data.
Cousin patterns
canonical-library-not-checked-first(2026-04-24, recurred 2026-04-25)working-from-mental-summary-not-rereading-the-source(2026-04-25)attribution-by-pattern-match-instead-of-verification(2026-04-24)
These all share the same underlying drift: substituting a compressed representation for the actual material. The lex-method's "read before claiming" applies to catalogues as much as to source texts.