Protocol specification demonstrates controlled semantic state evolution via graph reachability, indicating that canonical status can be derived dynamically without static verification tags.
We introduce OAOL, a protocol for the controlled evolution of canonical semantic state. In contrast to document-centric or repository-centric notions of canonicity, OAOL defines canonicity as reachability over a graph induced exclusively by admitted transitions. Dynamic observations and agent-generated proposals may motivate epistemic change, yet only those transitions that satisfy an explicit admission predicate enter the canonical state graph. Rejected transitions remain auditable but create no canonical edge. Verification status is derived, never stored. This note presents the formal core of the protocol, including the Canonical Semantic State, Labelled Transition System, Admission Predicate, Canonicality as Admissible Reachability, Rejection and HALT semantics, and the distinction between canonicality, auditability, and truth. The formal protocol is intentionally separated from implementation-specific storage technologies and generative models. The complete OAOL protocol specification is composed of the following independently maintained components: • OAOL Labelled Transition System (LTS) v1.0• OAOL Canonical State Schema v1.0 (JSON Schema Draft 2020-12)• OAOL Admission Engine v1.0• OAOL Formal Transition Examples v1.0• OAOL Conformance Test Suite v1.0 Core principles: 1. Canonicality is admissible reachability.2. Rejection preserves auditability but prevents canonical reachability.3. Canonicality does not imply truth.
No takes yet. Share an insight, caveat, or question.
Yuji Marutani (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: