Education

Semantic interoperability

Different systems need shared meaning, not simply matching names or transport formats.

How it works

A dictionary term, schema property and policy obligation can refer to connected but distinct concepts. A governed ontology preserves identity and provenance; exports map core fields into available representations. Graph and tabular formats differ in expressiveness, so parity is scoped to supported fields.

Illustrative explanations describe the boundary, not a customer result. For specific evidence and integration scope see the linked product page.

Formats preserve different expressive scope

A row export can expose definition, domain, category and standing; a graph format can carry typed relationships and source references. Compare canonical IDs and supported fields across representations, not byte equality between unrelated serializations. Document qualified input profiles and limitations before selecting integration adapters. Deployment and interoperability.

How a catalog and an ontology work together

A catalog inventories assets and supports discovery. A semantic runtime can add evidence-qualified meaning, concept identities, relationships and controlled exports; the catalog remains the discovery or operational surface. Matching a column name is not enough to conclude that two systems use the same concept.

In the v1.89.332 controlled rich example, two Payment Settlement senses remain distinct and review-required. Its 22 constituted concepts and 20 graph edges are not a customer-wide integration result. Inspect the actual rich dataset and the native profiles and OEM export details.

Evaluation: select a representative dictionary and schema, compare identifiers, definitions and source locators with expert review, then test downstream projection parity. Keep unsupported dimensions UNKNOWN rather than merging by label.

Technical note by OntoGuard · product evidence and review scope are identified in the linked materials.

Evidence and buyer questions

How does SI integrate with a data catalog?

A catalog inventories data and documentation. Semantic Intelligence constructs evidence-qualified meanings and reusable contracts and graphs that can feed existing catalogs. It complements catalog and graph platforms, not a universal replacement.

Inspect source or product evidence