Applied Ontology Readiness - Status

Unit ID: applied-ontology-readiness Status: AOR-1 to AOR-9, AOR-3b, AOR-12 and AOR-13 committed (6e36586, 6e9acb1, 54caeb6). AOR-10, AOR-11 and AOR-14 to AOR-17 implemented and self-validated, awaiting the human’s commands (uncommitted). Every slice is now implemented. Last updated: 2026-09-25 Trigger: human request, 2026-09-25 Plan: applied-ontology-readiness.md Sketch: applied-ontology-readiness.md Review request: applied-ontology-readiness-review.md ADRs: A-83, A-86 (with its addendum) and A-87 to A-96 Accepted (2026-09-25).

Current position

On 2026-09-25 the human reported every AOR-2 to AOR-9 command passing, accepted ADRs A-87 to A-92, chose content-hash version IRIs for generated documents, asked for Executable to align directly to PROV-O if that imposes no functional restriction, and asked for the Phase C ADRs and ADR-A83 to be drafted. The LOG.md sign-offs are the human’s to record.

Later on 2026-09-25 the human ratified ADR-A83, A-86 and its addendum, and A-93 to A-96, chose path encoding B for ADR-A90, and asked for AOR-14 to AOR-16 as one Quantification change with AOR-16 including the compiler work. The ADR-A83 harness was delivered as the eligibility-compiler unit’s Part B (VP). The agent ran the checks listed under Validation below. The human has not yet run them.

Slices

AOR-4’s Protégé step has a walkthrough: Loading LATTICE in Protégé.

Versions in 54caeb6 (against 6e9acb1)

All MINOR. Foundation and its vocab, Vocabulary, MORK: 0.2.0 → 0.3.0. Quantification, Party and its vocab, Surface and its vocab: 0.3.0 → 0.4.0. Eligibility, Instrument, Behaviour (spec and vocab each) and the applied capacity execution spec: 0.4.0 → 0.5.0. Executable: 0.3.0 → 0.4.0. applied/insurance is a sketch and untouched, by instruction.

Versions in the uncommitted change (against HEAD)

All MINOR. Quantification, Party and its vocab, Surface and its vocab: 0.4.0 → 0.5.0. Eligibility, Instrument, Behaviour (spec and vocab each) and the applied capacity execution spec: 0.5.0 → 0.6.0. Executable: 0.4.0 → 0.5.0. Mork: 0.3.0 → 0.4.0 (ADR-A97). The catalog is regenerated.

Validation run by the agent (2026-09-25)

check:mork-compilers (92 passed, reasoner tests included), check:python-root, check:ontology-versioning (29 documents), check:ontology-catalog (3 known defects), check:reasoning-testkit, check:reasoning-isolation, check:vocabulary.

Decisions awaiting the human

1. MTP pins: whether build should stop rewriting pins.lock.json (an MTP decision that mork-teaching-pack.md reserves for an ADR).

Discovered, outside this unit

• The persistence-compiler-iri-sync commits (772ea66, d19b7cf) bumped persistence/spec without regenerating the import catalog, so check:ontology-catalog failed at 3e7d911. Regenerated on 2026-09-25 (uncommitted). .github/copilot-instructions.md now reminds agents of every versioning step.
• ontology/mork/spec/Mork.ttl did not load in the OWL API and was not OWL 2 DL. Repaired under ADR-A97 (Mork 0.4.0, MTP re-pinned), which unblocks the eligibility-compiler unit’s A2 reasoner half (status).

• tools/literate_extract.py still resolves layers at <root>/<layer>, so its documented --check reports drift everywhere since the move to ontology/. The aggregate mise run check catches this, and CI runs only when dispatched.

• Link repair is planned as documentation-link-repair, pending. Its first slice fixes the checker scanning ignored docs/_site/.
• ontology/applied/insurance and the two MORK examples stay in KNOWN_DEFECTS, by instruction.
• Both GitHub workflows are manual-only (workflow_dispatch).
• rdflib: the MIN aggregate fails on two unbound values, and a grouped subquery is evaluated under the outer binding (AOR-8 VP).

Token record

Blockers

None.