Nineteen research papers

Which categorical structures have useful, checkable operational meaning for executable schemas?

Can a categorical semantic layer compile to efficient backend plans while remaining outside row processing?

What semantic information must be stated before executing a schema migration?

Under what assumptions can source-to-domain and domain-to-consumer mappings replace pairwise integration logic?

How can canonical queries be rewritten through typed mappings into source-native plans?

How should mapped views respond to data, schema, and mapping changes?

How can identity evidence remain explicit without collapsing uncertainty?

When can writes against a canonical schema propagate safely?

What is a semantic branch, diff, and merge when schemas, mappings, data, provenance, and reconciliation evolve together?

Which CatDB operations commute, converge, conflict, or require coordination?

Which consistency guarantee belongs to each CatDB operation?

Can semantic metadata improve routing, pruning, and distributed plan selection?

How should provenance compose across mappings, migrations, reconciliation, materialization, and queries?

How should CatDB select federation, local execution, caching, partial materialization, full materialization, or incremental refresh?

What production architecture preserves the chosen semantics while supporting safe and efficient execution?

Can a compact functional model act as a semantic oracle for mappings and migrations?

Which invariants justify Lean kernel checking in the first formalization?

Can one Markdown corpus support multiple typed projections with visible mappings and provenance?

Does the combined evidence support CatDB as a semantic data layer for evolving, integrated, and distributed information systems?