Is the object identifier semantically the same as the artefact identifier?

[phochste]

There was a discussion. during he 23 september 2021 meeting in Boekentoren, if the object @id should be semantically equivalent to the artefact identifier.

E.g. the as:object states object: http://somewhere.org/pub/1234

• When following the URL and retrieving the RDF version, should the subject of the RDF be also http://somewhere.org/pub/1234 or at least be semantically equivalent?

[phochste]

In the COAR group there is a related discussion about this fact. The current strategy is to add a global persisent identifier to the artefact (e.g. a DOI) as use the ietf:cite-as property in an object (and the RDF representation of the artefact).

But there are issues with ietf and JSON-LD. See also the discussion here: https://github.com/ietf-wg-httpapi/linkset/issues/45#event-5429025819