Skip to content

docs(threads-o8q): record Nova's refused Phase 5 sign-off in the status ledger - #26

Merged
BunsDev merged 1 commit into
mainfrom
docs/phase5-status-ledger-refresh
Aug 9, 2026
Merged

docs(threads-o8q): record Nova's refused Phase 5 sign-off in the status ledger#26
BunsDev merged 1 commit into
mainfrom
docs/phase5-status-ledger-refresh

Conversation

@BunsDev

@BunsDev BunsDev commented Aug 9, 2026

Copy link
Copy Markdown
Member

What

docs/phases.md calls itself "the honest status ledger" and asserts "If a doc or deck implies otherwise, this page wins." Its Phase 5 section was dated as of 2026-07-21, was 19 days stale, and omitted the single most important fact about the phase: Nova formally refused sign-off on 2026-07-29.

A stale authority doc is worse than a stale ordinary doc, because its whole function is to be the thing you check when other sources disagree. "Phase 5 is active" and "Phase 5's sign-off was refused and nobody has picked up the work" are very different states, and only the second is true.

Bead: threads-o8q. Docs-only — no code paths touched.

What was wrong

  1. .7 and .12 listed as in progress; .2, .13, .8 listed as open. All five are closed, as is .11. Only .9 and .10 remain.
  2. Nova's 2026-07-29 refusal appeared nowhere in the doc.
  3. The five named remediation blockers (threads-okc, 980, dgg, zav, 3jx) did not appear at all.
  4. The one-sentence summary said only "Phase 5 is active", which reads as healthy in-flight engineering. No remediation bead has been claimed in the 11 days since the refusal.
  5. The summary-table row carried the same stale breakdown.

What changed

  • Top-line summary and summary-table row now state the refusal and that no remediation work is claimed
  • Ledger re-dated 2026-08-09 with the correct closed/open split
  • New subsection recording what Nova affirmed as coherent (Channel/ApprovalPath separation, delayed apply, thin fail-closed Cave) versus what blocks, as a five-row per-bead status table
  • Records the 2026-08-09 re-verification: all four open blockers still true at coven 59c5be4, 129 commits after the commit Nova reviewed — with the static-verification caveat stated plainly
  • Names the related open beads explicitly not in the gate chain, so they aren't mistaken for blockers
  • Two new Known housekeeping discrepancies: PHASE-0-DESIGN.md §9's stale "before v0.2 freeze" heading, and threads-xpo (Channel::Deliberate specified but unreachable from the daemon)

Deliberate non-changes

  • specs/PHASE-0-DESIGN.md is untouched. It is FROZEN and change-controlled, and this page's own words are "These docs describe it; they do not amend it." The §9 heading problem is therefore recorded, not fixed at source. Amending it would need a post-freeze amendment bead, as threads-986.12 was.
  • Phase 5 keeps its [ACTIVE] label. The doc defines [BLOCKED] as "waiting on a named decision", and Phase 5 waits on engineering, not a decision. Making the refusal explicit in prose is accurate; relabelling loosely would introduce exactly the drift this PR is about.

On the §9 finding

Three of the four "open questions (need resolution before v0.2 freeze)" were settled by events after the 2026-07-14 freeze — §9.2 by Phase 2 merging as coven PR #382; §9.3 on 2026-07-15 as Shape B per PHASE-3-PORTABILITY.md §6; §9.4 by PHASE-4-CAVE-SURFACES.md and the 2026-07-17 Phase 4 freeze. Only §9.1 (does federation force a fourth fabric level) is genuinely open, and it is correctly deferred. The heading is the stale part, not the content.

Verification

Docs-only; nothing to build or test. Every status claim was checked against the live bd board and the repo on 2026-08-09 rather than carried over from the previous text.

Note for reviewers: cargo clippy -D warnings currently fails locally on rustc 1.95.0 at staging.rs:127 while CI is pinned to 1.88.0 and green. That is pre-existing and unrelated to this PR — tracked as threads-bnu, whose stated root cause I have corrected on the bead (CI is pinned; only local toolchains are not).

…us ledger

docs/phases.md claims precedence over every other doc in the family
("If a doc or deck implies otherwise, this page wins") but its Phase 5
section was 19 days stale and omitted the most important fact about the
phase: Nova formally refused sign-off on 2026-07-29.

- top-line summary and summary-table row now state the refusal and that
  no remediation work has been claimed
- ledger re-dated 2026-08-09; .2/.7/.8/.11/.12/.13 moved to closed, so
  only the two human gates .9 and .10 remain open
- new subsection records what Nova affirmed as coherent and the five
  named blockers, with per-bead status
- records the 2026-08-09 re-verification: all four open blockers still
  true at coven 59c5be4, 129 commits after the reviewed commit, with
  the static-verification caveat stated
- names related open beads explicitly NOT in the gate chain
- adds two housekeeping entries: PHASE-0-DESIGN section 9's stale
  "before v0.2 freeze" heading, and threads-xpo (Channel::Deliberate
  specified but unreachable from the daemon)

specs/PHASE-0-DESIGN.md is deliberately untouched -- it is frozen and
change-controlled, and this page describes it rather than amending it.
Phase 5 also keeps its [ACTIVE] label, because the doc defines [BLOCKED]
as waiting on a named decision and Phase 5 waits on engineering.

Docs-only; no code paths touched.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Copilot AI lite review requested due to automatic review settings August 9, 2026 11:07

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Updates docs/phases.md to bring the Phase 5 status ledger current and explicitly record Nova’s 2026-07-29 refusal to sign off, along with the remaining remediation blockers and their current (re-verified) status. This aligns the “honest status ledger” with the stated goal that it should be the authoritative source when other docs/decks diverge.

Changes:

  • Refreshes the Phase 5 ledger date and corrects the closed/open bead breakdown.
  • Adds a dedicated Phase 5 subsection summarizing Nova’s refusal, what was affirmed as coherent, and the per-bead remediation blockers.
  • Updates the Phase 5 summary-table row and extends “Known housekeeping discrepancies” with two newly recorded discrepancies.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

Comment thread docs/phases.md
> This page is the honest status ledger. Labels used throughout: `[FROZEN]` (design complete and change-controlled), `[MERGED]` (landed on the `main` branch of a downstream system that will call this crate at runtime once cut into a release; buildable and integration-tested but **not in a released binary** yet), `[RELEASED]` (in a tagged downstream release users can install), `[ENGINEERING FROZEN]` (code complete, tests green, awaiting one named decision to reach a deployed system), `[ACTIVE]` (open engineering phase with in-flight beads), `[BLOCKED]` (waiting on a named decision), `[NOT STARTED]`.
>
> The one-sentence truth: **phases 0–4 are frozen, Phase 5 (approval semantics) is active, everything downstream is merged but not in a released binary — no enforcement exists anywhere in production.** No released daemon in the wild calls this code. If a doc or deck implies otherwise, this page wins.
> The one-sentence truth: **phases 0–4 are frozen, Phase 5 (approval semantics) is active but its sign-off gate was refused on 2026-07-29 and the four named remediation beads are still open and unclaimed, everything downstream is merged but not in a released binary — no enforcement exists anywhere in production.** No released daemon in the wild calls this code. If a doc or deck implies otherwise, this page wins.
Comment thread docs/phases.md
- **Closed:** `.3` core approval types — `ApprovalPath`, `ApprovalPathKind`, `VetoWindow`, `ProposalClassification` (`approval.rs`); `.4` identity invariant predicates + advisory probes (`identity_invariants.rs`); `.5` `SurfaceRegionPredicate` + Gate-4 replay (`surface_regions.rs`); `.6` delayed-apply scheduler + audit — implemented **daemon-side in coven PR #430** (daemon-owned classification and scheduler, deadline/minimum-visible revalidation, fail-closed committed-evidence replay, cross-platform conditional atomic writes, startup recovery); `.11` authority review findings resolved; `.2` RFC closure/provenance amendments; `.7` Cave veto-window contract; `.8` implementation and migration fidelity; `.12` RFC-0001 approval-tier alignment; `.13` authorized retired-Ward migration fixture. The proposal/decision-record PR #6 merged 2026-07-27 as `091607f`.
- **Open — and this is the whole of Phase 5's remaining state:** `.9` Nova coherence sign-off gate and `.10` Val freeze gate. Both are human gates that agents must never simulate. Related: `threads-3xd` — RFC-0001 amendments (§5.5 closure precondition + §4.2 predicate (iv) provenance).

**Nova's sign-off is BLOCKED, not pending (2026-07-29).** This is the single most important fact about Phase 5 and it is easy to miss from the bead counts alone. Nova ran independent core and integration reviews against coven `f3cd322`, coven-threads `091607f`, merged coven PRs #430/#464, merged Cave PRs #3581/#3628, and familiar-contract PRs #3/#4, and **refused sign-off**. The design choices were explicitly affirmed as coherent — Channel and `ApprovalPath` remain separate axes, delayed apply is correct, Cave stays thin and fail-closed. What blocks is implementation, in five named beads:
@BunsDev
BunsDev merged commit 74569e5 into main Aug 9, 2026
2 checks passed
@BunsDev
BunsDev deleted the docs/phase5-status-ledger-refresh branch August 9, 2026 21:59
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants