research: 2 Amara courier-ferry absorbs (live-lock taxonomy + canonical agent-attribution convention)

docs/backlog/P2/B-0071-rename-otto-275-forever-out-of-live-lock-class-codex-pr-17-2026-04-28.md

@@ -0,0 +1,80 @@
+---
+id: B-0071
+priority: P2
+title: Rename otto_275_forever memory out of "live-lock 9th pattern" subclass + reclassify per Otto-352 taxonomy split
+status: open
+opened: 2026-04-28
+class: substrate-rename
+trigger: codex review on PR #17 thread (chatgpt-codex-connector P2)
+---
+
+# B-0071 — Rename `feedback_otto_275_forever_manufactured_patience_live_lock_9th_pattern_2026_04_26.md`
+
+## Source
+
+Codex review on PR #17 (ID: PRRT_kwDOSIvLus5-CMw2):
+
+> P2: Reclassify manufactured patience out of live-lock taxonomy
+>
+> This memory encodes manufactured patience as a `live-lock` subclass, but
+> the companion taxonomy absorb in the same commit explicitly reserves
+> `live-lock` for multi-agent/multi-worktree no-progress and separates
+> `manufactured patience` as its own class. Keeping the old label in the
+> memory name/frontmatter will continue routing single-agent inaction
+> incidents under the wrong class and undermine detector/playbook mapping
+> that depends on the canonical taxonomy.
+
+## Why deferred (form-2 deferral with tracking, not in-PR fix)
+
+PR #17's scope is the verbatim research-doc absorbs + memory landing of
+those absorbs. Renaming the file mid-PR cascades:
+
+1. File rename (`live_lock_9th_pattern` → manufactured-patience class label)
+2. Frontmatter `name:` and `description:` rewrite to drop "9th pattern in
+   Otto-2026-04-26 LFG branch-protection live-lock taxonomy" framing
+3. `memory/MEMORY.md` index update (paired-edit lint requires same-PR)
+4. Cross-references in other memory files (grep for the old filename
+   + the "live-lock 9th pattern" framing)
+5. Verification that no in-flight branch / PR / docs reference the old
+   filename
+
+Doing this inside PR #17 expands the PR scope substantially. The codex
+reviewer's substantive correction is accepted; the rename is the correct
+long-form fix; PR #17 lands as-is with this backlog row tracking the
+follow-up.
+
+## Acceptance criteria for the rename PR
+
+- [ ] File renamed to drop `live_lock_9th_pattern` substring (proposed:
+      `feedback_otto_275_forever_manufactured_patience_2026_04_26.md`
+      — keeps the substantive class label, drops the misclassified
+      taxonomy reference)
+- [ ] Frontmatter `name:` field updated to remove "9th pattern in
+      Otto-2026-04-26 LFG branch-protection live-lock taxonomy" framing
+- [ ] Frontmatter `description:` updated to refer to manufactured-patience
+      class per Otto-352 taxonomy split
+- [ ] Body references reframed to cite the 3-class split per Otto-352
+      (concurrent-thrash / stuck-loop / honest-wait) and place
+      manufactured-patience as separate-from-live-lock
+- [ ] `memory/MEMORY.md` row updated (paired-edit lint)
+- [ ] Cross-references in other memory files audited + updated
+- [ ] No CI breakage from broken xrefs
+
+## Composes with
+
+- **Otto-352** (`memory/feedback_live_lock_term_split_three_distinct_classes_otto_352_2026_04_26.md`)
+  — the canonical taxonomy split that this rename realigns to
+- **Aaron 2026-04-26** *"we discussed a while back that the live-lock
+  defintion i gave was overly broad but the word itself is fine"* — the
+  framing-correction that drove the Otto-352 split
+- **Otto-279** history-surface attribution carve-out — Otto-NN
+  references stay legible on memory/ surface
+
+## Why P2 not P1
+
+The misclassification is real but the *substantive* content (manufactured-
+patience pattern, Otto-275-FOREVER discipline, counterweights) is correct.
+Detector/playbook mapping per the codex finding would be undermined if a
+detector implementation existed today and routed by class label — but no
+such detector exists yet. The rename is durability-of-substrate work, not
+operational-blocker work.