Frequently asked questions
Common questions about OpenRiC - is it a product, what it costs, RiC-O alignment, SPARQL, writing, AI, and how to get involved.
Is OpenRiC a product? No. It’s an open contract (specification) plus a reference implementation. Any system can implement it; the clients work against any conformant server.
What does it cost? Nothing. Spec is CC-BY 4.0; reference implementation, viewer and capture are AGPL-3.0.
Which RiC version does it target?
RiC-CM 1.0 and RiC-O 1.1. The reference serializer emits zero non-canonical rico: terms; anything RiC-O lacks lives in the openricx: extension namespace.
Do I have to implement everything? No - pick a profile. Core Discovery (read-only) is the minimum useful claim.
Is there a SPARQL endpoint?
Yes, under the SPARQL Access profile - read-only /sparql. See SPARQL queries.
Can I write data, not just read it? Yes - the Round-Trip Editing profile covers POST/PATCH/DELETE with an audit trail. See Creating entities. On the reference server, creation is currently open (no key); edits/deletes need a key.
How is AI-suggested data handled?
The modelling wizard can propose models, validated against RiC-CM before display. The draft Inferred-Provenance profile requires any machine-asserted data to be visibly distinguishable from documented fact (prov:wasGeneratedBy; absence ⇒ asserted fact).
How do I know a server is conformant?
Run the conformance probe and check the profiles it declares at GET /api/ric/v1/.
Is my data locked in? No - that’s the point. Standard serialisations (JSON-LD/Turtle/RDF-XML, round-trip-guaranteed), a full export path, OAI-PMH, and SPARQL. See Exporting & harvesting and the Portability profile.
How do I get involved or report something? Open a thread in GitHub Discussions or email johan@theahg.co.za.
Next
Something missing or unclear? Tell us in Discussions or email johan@theahg.co.za.