Introduction

In, for example, the CatalogueAnnotations story, we want to be able to say something like 'Object N in UKIDSS DR3' is-a 'quasar'. What would that look like?

Let's suppose that we have addressed the problem of how to refer to that object, say with a URI like http://www.ukidss.org/dr3/objs/N.

How about the following?

@base <http://blah/my-sac>.
@prefix skua: <http://myskua.org/claimtypes/1.0/> .
@prefix dc: <http://purl.org/dc/elements/1.1/> .
@prefix iau: <http://www.ivoa.net/rdf/Vocabularies/IAUT93#>

<> a skua:IdentityClaim;
  dc:author "Norman Gray"; # author of the claim -- a URI/email would be better
  skua:subject <http://www.ukidss.org/dr3/objs/N>;
  skua:type iau:Quasars;
  dc:date "yyyy-mm-dd".

Alternatives and criticisms:

• iau:Quasars is a vocabulary concept, not strictly a 'type', therefore the semantics should be that the object 'is an instance of this concept', in the sense that were you searching for information about this concept, this would be a resource you might be interested in. Better would be to use an ontological type here (for example from the CDS Ontology of Astronomical Object types), where the 'this object IS A type' is more direct. This might be being over-fussy, but vocabularies are largely for helping searching, rather than for making deductions from.
• Perhaps this would be a reasonable use for the RDF reification vocabulary (though I understand there are few defensible uses for that)