David F GliddenandClaude Opus 5
ef6fa94dcd
The record is updated where information ARRIVES and never where a reader LOOKS.
Three instances found the same day, each with the correct information already
written down elsewhere:
REVIEWED-81 (AUTHORIZED 07-28) -> PENDING-81 still reads open
the ARC open-work register, which -> those 6 trackers still read live
says in its own description that
it supersedes them, with a
per-file verified verdict
register entry A2: "the content -> register entry B3, one page below:
half (B3) is now unblocked + "[OPEN - gated on A2]"
started" (06-17)
The third is the sharpest: both halves sit in the SAME document, so no tooling
gap explains it. ~/CLAUDE.md already rules this under Memory Discipline
("mark the superseded record as superseded"); the rule is being applied to facts
and not to trackers, PENDING items or register entries, which are also facts
about state.
Executed under steward authorization, arrears only:
- 16 trackers stamped `superseded_by:` + `superseded_on:` AND a visible in-body
banner. The banner is the point — metadata alone repeats the defect, since the
finding is about what a reader sees. 9 obsoleted by the 2026-07-07 MemPalace
wind-down; 7 already carrying the register's own verified `DONE-retire`.
- B3's label corrected, with the two-month error left visible rather than
silently amended.
- Nothing moved, renamed or deleted: 381 pointers resolve, 0 dead, 0 mis-authored.
HARVESTED BEFORE STAMPING, and this is why the harvest came first:
`project-mempalace-library-incremental-remine-strategy` carried a steward-verbatim
rule that is instrument-independent — never run a single multi-day index; order by
criticality; gate each tranche on READBACK, not on write-success (933,576 drawers
once stored successfully and could not be read back at all). It lived only inside
a MemPalace tracker, so retiring the instrument would have retired the rule with
it. ~/CLAUDE.md names that exact hazard: "state the obligation first and the
instrument second, or the next retired tool takes a rule down with it." Now
`feedback-bulk-indexing-runs-incrementally-with-readback.md`, and it applies to
studium-engine corpus work.
Held back deliberately: 2 correctly-parked files with named dependencies, 2 named
nowhere (returned to the steward), and 39 trackers unread — recorded as a gap,
not as a finding.
Measurement note: every earlier estimate of tracker staleness was wrong, because
both mtime and git-last-commit were reset across 283 files by the 2026-06-06
normalization sweep. The honest figure excludes it: median 90 days.
Filed as PENDING-142 ADDENDUM 2 per steward direction (fold, do not file anew).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013Y6t6qx7cpaCu5xGdD36u4
8.3 KiB
8.3 KiB
name, description, type, originSessionId, permalink, superseded_by, superseded_on
| name | description | type | originSessionId | permalink | superseded_by | superseded_on |
|---|---|---|---|---|---|---|
| ARC apparatus refactor — todo list (2026-05-06) | Tracks the multi-session refactor flowing from the universal dateline + photo-apparatus + three containment patterns spec consolidation (2026-05-06). Captures what was decided, what's done, what remains, and the location data the steward provided for the content audit. | project | 3bf7ae75-1138-4745-9024-0c40b3755ad3 | claude-memory/project-arc-apparatus-refactor-todo-2026-05-06 | project-arc-open-work-register.md | 2026-08-17 |
SUPERSEDED 2026-08-17. Register E4 →
[DONE—retire]— Tiers 1/2 shipped; the Tier-3 residual lives on as register B1.Current record: project-arc-open-work-register.md. Kept for detail and provenance — do not read the status below as live.
ARC apparatus refactor — todo list
Spec foundation locked 2026-05-06 across §V (containment patterns), §VII (universal dateline + photo caption), and §V applicability table. Implementation is multi-session work; this list tracks state.
Spec decisions locked (2026-05-06)
- Universal dateline — every content type carries a dateline, including glimpses and photo essays
- Format:
City, Country, Day Month Year- Country always included (half-located is a lapse)
- Local form preserved (Wien, Austria; Sevilla, Spain)
- Day included
- Three containment patterns:
- Text-primary (essay, meditation, observation, fragment, chamber, lex, about): vessel applies (3-lane: left margin = register marks; body = body+dateline+sources; right margin = sidenotes)
- Photo essay: vessel + .full-bleed image breakout; caption block at end of piece
- Glimpse: no vessel, image at full register, caption beneath, dateline at piece end
- Caption (photo apparatus) for glimpse + photo essay only:
- Content: technical EXIF (ƒ-stop · shutter · focal length) + austere subject identification when needed for comprehension
- Position: beneath image (glimpse) or end of piece (photo essay)
- Italic, subordinate, baseline-aligned (Müller-Brockmann)
- Test: helps viewer SEE, never tells them what to think
- Boundary rule: if image's comprehension needs more than austere caption identification, the work is not a glimpse — reclassify to meditation or essay
- Container class name:
.bears-apparatusproposed (role-named, replaces.has-margins)
Spec commits: e42ae0c (consolidation), 137ddd1 (dateline alignment fix)
Done 2026-05-06
- ✅ §V applicability table extended (universal dateline; new Caption row; new Vessel row)
- ✅ §V.b Containment patterns subsection added
- ✅ §VII Dateline section revised (universal, format, alignment principle)
- ✅ §VII Captions section revised (photo apparatus, two placements, EXIF + identification)
- ✅ Dateline alignment fix: removed margin-auto + max-width centering hack from
.dateline. On grid pages now aligns body-column-left via existing> *rule. (commit137ddd1)
Remaining work — ordered by readiness
Tier 1 — bounded SCSS + template work (~1-2 hours, ready)
.has-margins→.bears-apparatusrename- Rename selector across
_sidenotes.scss(3 references) - Update spec implementation notes that reference
.has-margins - Templates: apply
bears-apparatusclass to article elements based on content class (post.html, about-page.html, lex.html, sequence-index.html). Excluded for class=glimpse and class=photo-essay. - Investigation needed: Pandoc templating doesn't easily support negation; cleanest is a context variable from site.hs (
bears_apparatus = class != "glimpse" && class != "photo-essay")
- Rename selector across
- Extend §XII grid scope to observation/meditation/fragment
- Currently grid scopes only to
.post-content, article.essay, article.lex, article.about, .has-margins - Per spec, observation/meditation/fragment should also bear the apparatus
- Once
.bears-apparatusrename lands, this becomes automatic (templates apply the class for all bearing types) - Visual change for these types: body-column-width instead of full-page-content-width — verify acceptable
- Currently grid scopes only to
Tier 2 — content audit (~30-60 min once tooling ready)
- Add
location:frontmatter to all text posts that lack it
Locations the steward has provided (canonical, use exactly):
Essays:
| Title | Location |
|---|---|
| The Death of a Hero | Seoul, Korea |
| The Discovery of Matthieu Taliesin | Barcelona, Spain |
| From Mushi-Ken to Morra – The Ritual of Choosing | Barcelona, Spain |
| Vespers for the living | Cremona, Italy |
Observations:
- All Paris, France
- Exception: 2024-01-11 → Chez Bourny, Champagnat, France
Glimpses:
- Locations extrapolated from titles (per steward's note)
After the Reply essays — already have locations (Wien, Barcelona, etc.)
- Reformat existing datelines to new spec form (City, Country, Day Month Year):
- Existing pieces with location-but-no-country (e.g., "Wien" → "Wien, Austria")
- Existing pieces with location-but-no-day (verify post.html template handles day-rendering correctly; some pieces may have only month-year in frontmatter)
- Update post.html template to render dateline in canonical format (currently renders
$dateline_date$— confirm the day component)
Tier 3 — EXIF integration (investigation needed before scoping)
- Investigate EXIF reading at Hakyll build time
- Possible approaches: (a)
exiftoolpreprocess emitting JSON → Hakyll context, (b) Haskell library called from site.hs, (c) manual frontmatter (least automated, most reliable) - Need: f-stop, shutter speed, focal length per image
- Decision needed: does this happen for ALL images on the site, or only for glimpse/photo-essay content?
- Possible approaches: (a)
- Caption template — once EXIF is available, render caption per spec:
- Glimpse: beneath image (template includes after
<img>block) - Photo essay: collected at end of piece in Image-Apparatus block (parallel to §VII.b Sources)
- Glimpse: beneath image (template includes after
- Content audit: identify all glimpses + photo essays needing caption migration; verify EXIF data is intact in source images
Tier 4 — verification (after each tier)
- Visually verify dateline alignment across all six content types after each tier lands
- After the Reply sequence — validated form, must remain unchanged visually
- Death of a Hero — first individual-essay Sources instance (already shipped 2026-05-05); dateline must reformat to "Seoul, Korea, 5 March 2016"
- At least one glimpse with caption (use Felsenreitschule as a candidate after EXIF tooling is ready)
Open spec questions (note in §VII when decided)
- Glimpse dateline alignment — RESOLVED 2026-05-07. Locked as centered per Bringhurst's coherence principle (don't mix symmetric and asymmetric apparatus); follows the symmetric register the glimpse piece already establishes (full-bleed image + centered figcaption). Sourced in §VII Dateline Alignment subsection (AldineXXI-specification.md). First-instance verification on
2015-03-02-5e-arrondissement-paris-france. - ISO in caption EXIF — currently spec says "judgement call, not default." Lock once first photo essay decides.
- Photo essay end-of-piece block name — "Image apparatus" tentative. Lock with first instance.
Notes for future sessions
- Spec changes today were heavy — ~116 insertions in AldineXXI-specification.md (commit
e42ae0c). The Pass 1 SCSS audit + spec consolidation + dateline fix were three substantial pieces of work. The refactor remaining is bounded but multi-session. - The .bears-apparatus rename is the gateway to making observation/meditation/fragment dateline alignment work. Until that rename lands, those types keep current visual behavior.
- The EXIF investigation is the unknown-scope item. Treat it as its own session.
- Cross-reference:
feedback-fittingness-as-refactor-test.md(the elegance test for code) andfeedback-deploy-is-not-commit.md(session-end discipline).
Drift patterns to watch
- Structural-counter-as-health-signal: a green build doesn't mean spec compliance. Verify visually on representative pages of each content type.
- Deploy ≠ commit: every refactor pass ends with
git statuscheck and explicit commit before declaring done. - Cosmetic refactor: don't rename or restructure unless redundancy or unclarity is being removed. The fittingness test applies.