Re: [w3ctag/design-reviews] WG Revision: SHACL 1.2 Core (Issue #1166)

HolgerKnublauch left a comment (w3ctag/design-reviews#1166)

This is not an official response from the WG (I don't know what the process for that would be), but we have recently made sure that the wording around node expressions is clearly about *derived value nodes* only, and is not really an inference or reasoning capability that produces triples.

However, other documents including the SPARQL-RL and SHACL Inference Rules (currently part of SHACL-SPARQL) do define new inference capabilities, and SHACL rules can rely on triples that are derived using sh:values and sh:defaultValue if they want to. So SHACL is confidently adding new alternatives to RDFS, OWL and similar entailments, although not as part of SHACL Core.

On the second question of reporting failures on unsupported entailments, this has not changed since SHACL 1.0, see https://www.w3.org/TR/shacl/#shacl-rdfs which clearly states that support for entailments is optional. This means that interoperability can be limited, in the same way that RDFS or OWL-based systems are not required to support SHACL. If anyone has specific proposals on what we can/should change, we welcome more input.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/1166#issuecomment-5363952076
You are receiving this because you are subscribed to this thread.

Message ID: <w3ctag/design-reviews/issues/1166/5363952076@github.com>