The FAIR principles for research data (Findability, Accessibility, Interoperability, Reusability) are a set of guidelines to facilitate sustainable availability of (research) data.
Inspired by Fluree's "Making Data FAIR", I decided to look at ArcadeDB in terms of FAIRness.
So, to check whether a tool X helps making (research or enterprise) data (more) FAIR, I ask for each principle Y: How does X make data Y?
How does ArcadeDB make data findable:
- A unique identifier is assigned to each record and never reassigned.
- Custom metadata for types and properties beyond SQL
COMMENT ON. - The identifier is a mandatory property of every record.
How does ArcadeDB make data accessible?
- Query Languages: SQL dialect, Cypher, Gremlin, GraphQL, MQL (MongoDB), SPARQL (via Gremlin plugin)
- Interfaces: HTTP, Java, JDBC, MongoDB, Redis, Postgres
- Drivers: CHICKEN Scheme, .NET, Ruby, Rust
How does ArcadeDB make data interoperable?
- The native data format is JSON.
- Data types map to standard Java data types.
- Links between records are based on unique identifiers.
How does ArcadeDB make data reusable?
- Property fields (i.e. for licenses) can be made mandatory.
- Provenance can be enforced by immutable records.
- Export formats: JSONL, GRAPHSON, GRAPHML
So overall, ArcadeDB makes data fair by its feature set. While this feature configuration is likely not unique, a polyglot multi-model NoSQL/NewSQL database is rare.
Top comments (0)