# P1 related-work ledger

**Research cut:** 2026-07-22  
**Status:** provisional author ledger, pending paper review  
**Rule:** each relation below is a feature-level comparison, not a subsumption
claim.

| Source cluster | Established result or capability | P1 use | CatDB delta, if any | Residual or nonclaim |
|---|---|---|---|---|
| B01, ologs | Types as objects, functional aspects as arrows, facts as commuting diagrams, and functors between ologs | Foundation for typed paths and equations | No mathematical novelty claimed | P1 does not establish that ologs are incomplete or non-executable |
| B02–B03, functorial data migration | Schemas as categories, instances as set-valued functors, schema functors inducing migrations, finite graph-and-equation presentations, relational realization | Foundation for mappings, paths, and pullback | Proposed normalized artifact and evidence boundary | P1 does not claim the first categorical or relational execution model |
| B04–B05, algebraic integration/databases | Typesides, equational schemas and instances, mapping-induced migration, and algebraic integration | Closest mathematical baseline | Profile v1 is an intentionally smaller operational profile | General typeside terms, observation equations, general pushforwards, and queries are outside P1 |
| B06, model management | Algebraic operations over schemas, mappings, and data | Model-management positioning | Explicit immutable artifact and obligation ledger are proposed | No claim that model management is new |
| B07–B09, FQL/CQL | Executable categorical integration, scientific case studies, and a maintained language/IDE | Closest executable baseline | Cross-language fixture and evidence-label contract remain proposed | No implementation-parity, performance, compatibility, or novelty claim |
| Codd 1970 | Relations, keys, operations, and logical/physical independence | Relational comparison | Typed functional projection and loss accounting are proposed | Bags, nulls, SQL constraints, views, recursion, and physical design are not subsumed |
| Chen 1976 | Conceptual entities, relationships, and attributes | ER comparison | Selected functional elaboration into paths and attributes | N-ary relationships, participation, weak entities, and cardinalities need explicit treatment |
| Fagin et al. 2005, data exchange | Source-to-target dependencies, solutions and universal solutions, chase constructions, and certain answers | Mapping, migration, and integration baseline | Typed categorical path/equation representation and cross-language validation are candidate deltas | P1 provides no general dependency language, chase-termination result, universal-solution theorem, or certain-answer guarantee |
| B37–B40, RDF/OWL/SPARQL/R2RML | Triple and dataset semantics, ontology profiles, graph-pattern queries, and relational-to-RDF mappings | Semantic-web comparison | A typed closed-world projection is only a proposed bridge | Open-world entailment, blank-node scope, arbitrary predicates, and SPARQL are not path equations |
| B41 plus GQL/PG-Schema papers | Standardized property-graph language; labeled property multigraphs, graph patterns, schema and key work | Property-graph comparison | Functional-arrow projection with residuals | Edge identity, multiplicity, properties, labels, inheritance, path matches, and updates remain native features |
| B42, Ontop | Mapping-based virtual semantic access and SPARQL-to-SQL rewriting | Federation boundary | CatDB would add its own typed obligation records | P1 does not claim unprecedented semantic federation |
| B43, AsterixDB | Typed semistructured data with open/closed records and nested collections | Document/multimodel boundary | Closed total fragment may be represented | Open records, missing/null distinctions, nesting, arrays, and runtime breadth are not represented |
| B44, HNSW | Physical approximate-nearest-neighbor retrieval | Vector boundary | Vector result may be typed as candidate evidence in later work | Similarity does not establish identity, path typing, or a schema law |
| B65, HoTTSQL | Machine-checkable SQL semantics and verified rewrites | Proof-status discipline | Future CatDB-specific Lean coverage may differ | The provisional paper contains no machine-check claim |

## Required final-review checks

1. Replace living CQL documentation with a pinned revision or archived
   snapshot before making release-specific claims.
2. Use ISO/IEC 39075:2024 for normative GQL claims; use pre-standard papers
   only for open technical detail and history.
3. Do not promote any CatDB bridge from “proposed” until its named fixture,
   implementation, and preservation/loss tests pass.