Scope and contributions

Status: P1-R2 candidate / 0.1.0-rc.1. The repository and Wiki are publicly readable as of 2026-09-30. Final scholarly release, artifact availability and citation remain pending #55/#56. The ontology files and governed evidence linked below are authoritative.

Paper 1 studies selected semantic distinctions, operational mapping, and bounded Pharma federation. The nine headline claims have separate evidence roles and ceilings; none is a blanket ontology validity claim. The selected 27/27 Jira fields are dispositions of one schema. The 8/8 SQL↔SPARQL result applies to eight predeclared tasks. Independent transferability is unsupported on the current evidence.

How the candidate developed

1. Evidence to boundary. Source mining and cross-source reconciliation exposed risk vocabulary, conflicts and coverage gaps. The G1 decision permitted a bounded Paper-1 conceptual synthesis with conditions; it did not make missing source families complete.
2. Boundary to model. The G2 architecture decision assigned Core and module/profile responsibilities for the selected distinctions. The resulting P1-R2 candidate manifest names six Turtle modules, SHACL and the local import policy. A formal candidate is not independent domain validation.
3. Model to executable check. The deterministic semantic build checks declared candidate inputs. The relational parity work examines selected SQL↔SPARQL tasks and projection losses. Passing those tasks supports the measured scope, not a universal lossless mapping.
4. Check to claim. Evidence and VVEAA separates observed checks from claim interpretation. Independent expert review (#51), shortage-domain holdout (#31), final claim calibration (#53), assurance (#54) and scholarly release (#55) remain open. Negative and partial findings remain visible in those gates.

This is the current candidate's research route, not a complete entity-level version lineage. Release train distinguishes the evidence, conceptual, semantic, executable, evaluated and publication-bound states; a later planning stage is not inherently a stronger scientific result.

Preliminary nine-claim calibration · Claim-to-method/evidence matrix

Final abstract, comparison, limitations and submission wording await #13/#31/#51/#53/#54/#55/#56. The completed #14 comparison gate does not resolve the remaining #31 impact review. Return to Home.