07 — Workstream F: Governance Row Bridge & OSPA Packet
07 — Workstream F: Governance Row Bridge & OSPA Packet
Status: OSPA human-gated; row-bridge DEFERRED (deliberately, with reason)
Live governance state (re-verified 2026-06-03)
| Object | Count |
|---|---|
governance_object_ownership (active = OSPA) |
0 |
governance_responsibility_scope (active) |
6 (approval, audit, execution, health, policy, render) |
governance_candidate_object / _state / candidate_scan_run / ruleset / gov_worker_cursor |
0 |
v_governance_object_inventory |
35 (collection-grain, object_type=collection) |
v_object_owner_gap |
210 (= 35 governed+BIRTH_REQUIRED collections × 6 scopes) |
The seam is, by ratified design, collection-grain: inventory = collection_registry where
governance_role='governed' AND coverage_status='BIRTH_REQUIRED'. Gap = inventory × active scopes −
effective accountable owners.
Is the row bridge live? No — and applying it now is NOT clearly safe
A row-grain bridge could expose DOT/pivot/birth rows (not just collections) as governance candidates. But:
- A naive bridge over
birth_registry(1,121,521 rows) × 6 scopes ≈ 6.7M synthetic "gaps" — that is noise, not signal, and could be mistaken for a real, enormous obligation. - The row-grain governance model is unratified. Which rows
require_owner, at what object grain, under which scopes — these are owner/design decisions. Prior macros deliberately deferred P5 for exactly this reason. - The macro's "apply read-only bridge if safe" is conditioned on safe; manufacturing millions of misleading gap rows is not safe. Decision: do not apply unilaterally. Provide paste-ready, scoped DDL instead (critical families only) for the owner to ratify.
Paste-ready (NOT applied) — scoped row-bridge for owner review
A future, ratified bridge should be scoped to critical-family rows (dot_tools ~309, pivot ~37,
iu_cmd 54 → hundreds, not millions) and gated on a ratified requires_owner rule per species/role.
This is a governance design deliverable, not a blind apply.
Does OSPA / human approval still block? Yes (ospa=0 → L2 rollout NO-GO). Cannot be faked.
OSPA approval packet (for L2/L4 owner)
| Field | Content |
|---|---|
| What needs approval | Seed ≥1 governance_object_ownership row (owner_kind=accountable, lifecycle_status=active) to open L2 rollout; and/or ratify + authorize a scoped row-grain bridge |
| Why | Unblocks ownership operationalization; lets born DOT/pivot rows become governable |
| Risk of not approving | Born DOTs/pivots stay ungoverned; owner_gap stays collection-grain (210); dot-pivot-update + the 16 files cannot enter governance |
| Rollback | Delete seeded ownership rows; DROP any bridge views (inert) |
| Verify after | fn_preflight_guard() row governance_gate_ospa flips CLOSED→OPEN; v_object_owner_gap begins to close |
Governance impact of THIS macro's remediation
Birthing 22 pivots does not create new governance obligations at the collection grain (the
collection pivot_definitions was already in inventory/gap). It does make those 22 rows eligible
for a future row-grain bridge. No governance object was created or mutated this run.
Completion: governance state re-verified; OSPA packet ready; bridge decision explicit (deferred,
with reason); no fake OSPA, no governance bypass.