Repository navigation
main fix - #523
Merged
Merged
main fix#523
Conversation
briansrls
marked this pull request as ready for review
April 18, 2026 04:42
Contributor
Author
|
ChatGPT review in progress... (view conversation) Check back in ~30 minutes for the full review. |
briansrls
added a commit
that referenced
this pull request
Apr 18, 2026
Lands docs/phase-plan-2026-04-18.md as the director-chat coordination snapshot for the post-merge-wave phase. Derived from the 2026-04-18 draft with director review baked in. What this doc is: - Snapshot of where the four lanes stand after tonight's 10 merges. - Paste-ready dispatch briefs for the next batch (§3). - Pre-clearance tracker for medium-horizon items (§4). - Stage 3c readiness exploration (§6) — the "strategic horizon" work. - Operating model for the director + child-chat fan-out pattern (§7). Director review findings incorporated: - Staleness: #523 was shown "in flight" but already merged as 977224a — doc now reflects merged state with explicit SHA. - DB-numbering audit: draft reserved DB-18/19/20/21/22 speculatively; dropped because DB-7 (symbolic cost), DB-8 (fixed-point ratchet), and DB-15 (test-infra R2) already cover the work. Also noted DB-17 is allocated to reference-resolution-provenance, so draft's use of DB-17 for WorkflowEffect would have collided. - Stage 3c exploration (§6) filled in: compiler.dag is a 310-line cycle meta-model (not compiler internals); Lane 1 Stage 1e is a hard blocker per lane3-self-hosting-cycle.md; 4 genuinely-open questions named. - STOP-AND-ESCALATE rule added to each §3 dispatch brief + §7 — child chats halt when pre-clearance turns out to have been wrong, rather than patching forward and losing the structural signal. - §5 split: items already in ROADMAP (point to authority) vs migration candidates (new from this session); explicit dispatch decision on which items go to which chat. What's next: - Current §2 dispatches close; §3 briefs go out in order (Python pilot gated on PR 2.5). - §5b migration happens via combined XS brief (§3) + dedicated ROADMAP-hygiene chat (NOT folded into the XS brief). - Next director session commits the first 60 min to §6's open questions regardless of tactical surprises. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
5 tasks
briansrls
added a commit
that referenced
this pull request
Apr 18, 2026
* Phase plan 2026-04-18 — post-merge-wave coordination snapshot Lands docs/phase-plan-2026-04-18.md as the director-chat coordination snapshot for the post-merge-wave phase. Derived from the 2026-04-18 draft with director review baked in. What this doc is: - Snapshot of where the four lanes stand after tonight's 10 merges. - Paste-ready dispatch briefs for the next batch (§3). - Pre-clearance tracker for medium-horizon items (§4). - Stage 3c readiness exploration (§6) — the "strategic horizon" work. - Operating model for the director + child-chat fan-out pattern (§7). Director review findings incorporated: - Staleness: #523 was shown "in flight" but already merged as 977224a — doc now reflects merged state with explicit SHA. - DB-numbering audit: draft reserved DB-18/19/20/21/22 speculatively; dropped because DB-7 (symbolic cost), DB-8 (fixed-point ratchet), and DB-15 (test-infra R2) already cover the work. Also noted DB-17 is allocated to reference-resolution-provenance, so draft's use of DB-17 for WorkflowEffect would have collided. - Stage 3c exploration (§6) filled in: compiler.dag is a 310-line cycle meta-model (not compiler internals); Lane 1 Stage 1e is a hard blocker per lane3-self-hosting-cycle.md; 4 genuinely-open questions named. - STOP-AND-ESCALATE rule added to each §3 dispatch brief + §7 — child chats halt when pre-clearance turns out to have been wrong, rather than patching forward and losing the structural signal. - §5 split: items already in ROADMAP (point to authority) vs migration candidates (new from this session); explicit dispatch decision on which items go to which chat. What's next: - Current §2 dispatches close; §3 briefs go out in order (Python pilot gated on PR 2.5). - §5b migration happens via combined XS brief (§3) + dedicated ROADMAP-hygiene chat (NOT folded into the XS brief). - Next director session commits the first 60 min to §6's open questions regardless of tactical surprises. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * Address chatgpt-review on #525: fix Track 9 drift + ROADMAP-duplication Reviewer (commit 697c3a5) flagged two BLOCKING single-authority slips where this snapshot drifted from ROADMAP.md (the root doc, not the src/v3 sub-tracker I initially referenced): 1. #516 row listed Track 9 primitives as NonEmptyList/NonSingletonList/ ArityIndex/TransformRef. Authoritative list at ROADMAP.md:765-768 is NonEmptyList<T>/NonSingletonList<T>/ParamRef/TransformRef. ArityIndex was explicitly REJECTED as a standalone primitive at ROADMAP.md:772. Row now restates from ROADMAP with line reference, not from memory. 2. §3 XS brief and §5b both instructed a child chat to "add ROADMAP entry" for ParamRef/TransformRef constructor-validation asymmetry. ROADMAP.md:777-786 already carries that exact debt with the Lane 3c graduation trigger. §3 brief now scopes to a substrate.dag cross-reference comment only; §5b ParamRef row removed; §5a gains the row pointing at the existing ROADMAP entry. Non-blocking finding also addressed: 3. §6 Q3 flagged compiler.dag's v2-path carryover as "no ROADMAP row yet" — a real planning fact stranded mid-doc. Moved to §5b as a migration candidate (pending director pre-clearance on whether it's a real 3c blocker or a v2→v3 bridging detail). Root cause (reviewer's deepest point): doc was behaving partly like a second tracker, partly like a thin read-model. Added explicit Snapshot-discipline rule: "This doc is a thin read-model over ROADMAP. When ROADMAP and this doc conflict, ROADMAP is right and this doc must be fixed — delete the restated fact, replace with a line reference." Plus fixed the parent link from ../src/v3/ROADMAP.md (the sub-tracker) to ../ROADMAP.md (the authoritative root). Also updated §5a to include Track 9 second consumer (IndexedElement.index → ElementRef) as already-in-ROADMAP (ROADMAP.md:770 — "deliberately not pre-declared"), removing it from §5b where I'd incorrectly framed it as needing migration. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * Address chatgpt-review on b7b4dba: authority attribution + framing Non-blocking finding from the re-review: DB-17 attribution said "per ROADMAP §Scheduled deletions" without distinguishing which ROADMAP or citing the allocating design doc. Reviewer couldn't verify the attribution in the uploaded ROADMAP file. Fix: name the allocating authority explicitly (docs/design-reference-resolution-provenance.md) with the ROADMAP reference as a secondary pointer, not the primary one. Design-question finding: doc claimed to be "thin read-model over ROADMAP" but in practice reads from ROADMAP + DB design docs + post-l15-phase-plan. Reviewer rightly flagged the framing as understating what the doc actually depends on. Fix: broaden the snapshot-discipline framing to "thin read-model over ROADMAP and the design docs ROADMAP references." Added explicit authority order on conflict: (1) specific DB design doc for what it locks, (2) ROADMAP for active-deferrals / structural debt, (3) lane/phase docs for sequencing. Future readers who hit conflicting information now know where to resolve from. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * WIP: gunbc design advancement * Address chatgpt-review on 07f3c95: strip citation-count bookkeeping Non-blocking finding: §1 "structural rule banked" had a Q4/Q3 drift (§1 said Q4 cited in ComposedEffect; §2 correctly said Q3), and more fundamentally the citation-count bookkeeping was this snapshot acting as a second authority for a rule not yet in INVARIANTS.md. Fix (reviewer's option b — strip bookkeeping until the memo graduates to a canonical authority): - §1 banked rule: removed the specific "Q5 in PR 2.5, Q4 in ComposedEffect" citation list and the "graduate after Python pilot dispatch" counter. Replaced with a brief statement that the memo is currently cited from dispatch briefs (§2, §3) and graduation is a separate follow-up not tracked from this snapshot. - §5b: removed the "graduation" migration row for the same reason. - §5b dispatch decision: recalibrated row count (4 not 5) and clarified that 2 rows go to §3 briefs, 1 to a dedicated hygiene chat, 1 to director pre-clearance first. Q-number citations in §2 acceptance and §3 briefs remain — those are legitimate dispatch acceptance criteria ("cite Q5" / "cite Q3"), not aggregate bookkeeping. Only the aggregate graduation-tracking from the snapshot is stripped. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Opened from session-dashboard for session
lively-tern-606.