I'm sorry Nick, I did not read your email closely enough. Local resources and Inline representations are already in the spec and why you were questioning is the value-type for the creator property-value.
I believe the change to dcterms:creator and dcterms:contributor was accidental and they should change back to value-type Local Resource. Thanks, - Dave On Mon, Aug 2, 2010 at 1:08 PM, Dave <[email protected]> wrote: > On Mon, Aug 2, 2010 at 10:03 AM, Nick Crossley <[email protected]> wrote: >> >> I have a related question on the same property. In earlier versions of >> the core spec, the creator and similar resources were defined to be of type >> 'Resource or Local Inline Resource'. In the tidying up of that terminology, >> the definition is now type=Resource, with representation=either. However, >> this is not quite the same: the new spec does not seem to permit a local >> resource, and requires the foaf resource to be gettable. Was this a >> deliberate change, or accidental? What are the implications of requiring >> the creator and related person attributes to be gettable resources? For >> example, is it required that the 'same' person used in two different service >> providers must have the same URI? What does the 'same' person mean? >> Personally, I would strongly prefer that providers be allowed to return >> local resources for foaf:Person or similar resources, avoiding any >> requirement that such things are persistent gettable resources. > > I don't believe we ever had consensus to remove the notion of local > resources from the Core. > I believe what we wanted are two value types for resources: > 1) Resource > 2) Local Resource > 3) Either > And two representation types: > 1) Reference > 2) Inline > 3) Either > I consider this to be a correction in the spec and would like to get it in > ASAP. > Thoughts? > Thanks, > - Dave > > _______________________________________________ > Oslc-Core mailing list > [email protected] > http://open-services.net/mailman/listinfo/oslc-core_open-services.net > >
