Skip to content

docs: reconcile RELEASING.md + .gpd/STATE.md with live v0.1.0 release#5

Open
Zer0pa-Architect-Prime wants to merge 1 commit into
mainfrom
docs/r5-reconcile-v0.1.0-public-release
Open

docs: reconcile RELEASING.md + .gpd/STATE.md with live v0.1.0 release#5
Zer0pa-Architect-Prime wants to merge 1 commit into
mainfrom
docs/r5-reconcile-v0.1.0-public-release

Conversation

@Zer0pa-Architect-Prime
Copy link
Copy Markdown
Contributor

Summary

Closes the governance drift flagged by the Wave-2 FPO/Repo review (R5): both RELEASING.md and .gpd/STATE.md still claimed v0.1.0 was a local-only internal tag with OWNER_DEFERRED licence, while the public custody-only release v0.1.0 has been live on GitHub + PyPI since 2026-05-04T01:44:58Z (cuneiform-control 0.1.0 Apache-2.0 via Trusted Publishing).

The reconciliation distinguishes the two release types that had collapsed in the prior wording:

  • Custody-only public release (what v0.1.0 actually is): manifest-validation control pack; preserves NO_GO_GOVERNING_GATE_UNMET verbatim; no decipherment claimed; no image-bearing corpora redistributed; Apache-2.0 (code) + CC-BY-4.0 (docs).
  • Scientifically-repaired public release (still not allowed): requires repaired governing gate on new scientific evidence and an extracted rerun surface beyond manifest validation.

Scope discipline

  • Doc-only edits — RELEASING.md + .gpd/STATE.md. No code, no proof artefact, no smoke-runner changes.
  • The failed-gate boundary discipline is unchanged. NO_GO_GOVERNING_GATE_UNMET remains authoritative; smoke PASS recorded; HF custody pinned.
  • The historical Phase-03-close v0.1.0-internal staging tag record is preserved for traceability; the public custody-only v0.1.0 release record is added alongside it.

Test plan

  • RELEASING.md and .gpd/STATE.md no longer assert OWNER_DEFERRED licence or "tag not pushed" while v0.1.0 is publicly live
  • No softening of NO_GO_GOVERNING_GATE_UNMET anywhere in the diff
  • No claim of decipherment, scientific gate repair, or image-corpus redistribution introduced
  • Custody-only / scientifically-repaired distinction is unambiguous

Truth basis

  • Live PyPI: https://pypi.org/project/cuneiform-control/0.1.0/ (Apache-2.0; summary "Does not repair the failed governing gate")
  • GitHub release: https://github.com/Zer0pa/Cuneiform/releases/tag/v0.1.0 (published 2026-05-04T01:44:58Z)
  • HEAD commit on main: 3d29caa (2026-05-04T01:34:04Z)
  • README ## Live Status, pyproject.toml, LICENSE, CITATION.cff, .zenodo.json — all consistent on Apache-2.0 + custody-only scope
  • FPO/Repo Orchestrator citations: LANDING_CARD_WAVE2_CENTRAL_DISPATCH_BRIEF_2026-05-08.md §B row 459 (R5); FPO_DISPATCH_BRIEF_PRODUCT_PAGE_2026-05-09.md §A line 335 ("Repo fixes required: reconcile RELEASING.md / .gpd/STATE.md governance drift")

Closes the governance drift flagged by the FPO Wave-2 R5 review: both
docs still claimed v0.1.0 was a local-only internal tag with
OWNER_DEFERRED licence, while the public custody-only release v0.1.0
(cuneiform-control 0.1.0 Apache-2.0 on PyPI, GitHub release 2026-05-04)
has been live since 2026-05-04T01:44:58Z.

The reconciliation distinguishes the two release types that have
collapsed in the prior wording:

  - **Custody-only public release** (what v0.1.0 actually is):
    manifest-validation control pack; preserves NO_GO_GOVERNING_GATE_UNMET
    verbatim; no decipherment claimed; no image-bearing corpora
    redistributed; Apache-2.0 (code) + CC-BY-4.0 (docs).

  - **Scientifically-repaired public release** (still not allowed):
    requires repaired governing gate on new scientific evidence and an
    extracted rerun surface beyond manifest validation.

The failed-gate boundary discipline is unchanged. Smoke PASS, HF custody
pinned, NO_GO_GOVERNING_GATE_UNMET still authoritative. No code or proof
artefact changes — pure documentation reconciliation.
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.

1 participant