- From: Michael Steidl via GitHub <sysbot+gh@w3.org>
- Date: Thu, 12 Oct 2017 06:40:14 +0000
- To: public-poe-archives@w3.org
* It is an [explicit requirement of RDF](https://www.w3.org/TR/rdf11-concepts/#section-triples) that each triple MUST have a "_subject, which is an IRI or a blank node_". * It was a decision of this WG to build POE ODRL on top of RDF. * Therefore JSON-LD is an implicit option for ODRL syntax and [JSON-LD defines the keyword](https://www.w3.org/TR/json-ld/#syntax-tokens-and-keywords) @id "_used to uniquely identify things that are being described in the document with IRIs or blank node identifiers_"; can be used as identifier of the subject of a triple. * For a better compatibility with earlier ODRL versions the ODRL-internal name uid was adopted for identifiers and the JSON-LD @context at http://www.w3.org/ns/odrl.jsonld defines already that uid is transformed/renamed to @id By that I conclude that it is primarily an editorial error that no uid is defined for Constraint and Logical Constraint as these classes MUST have an identifier for the subject (= an instance of the class) and re-using the definition of uid used for other classes does not change the IM in any way. Therefore I suggest to add the uid property name to Constraint and Logical Constraint as correction of an editorial error as it is a goal of a Recommendation to express the same thing in a consistent way across a document. (Using uid for some classes and @id for Constraint & Logical Constraint generates the same result in terms of RDF triples but is not consistent for the ODRL user.) -- GitHub Notification of comment by nitmws Please view or discuss this issue at https://github.com/w3c/poe/issues/278#issuecomment-336035223 using your GitHub account
Received on Thursday, 12 October 2017 06:40:04 UTC