Re: #rdfms-literals-as-resources in scope?

On Thu, 12 Jul 2001, pat hayes wrote:

> >Graham Klyne wrote:
> > >
> > > At 02:52 PM 7/10/01 -0700, Sergey Melnik wrote:
> > > >resource (constant) = URI, name, referring expression
> > >
> > > I think that's clearly at odds with RFC2396 (which seems the nearest thing
> > > we have to a universally accepted starting point for defining
> >these things):
> > >
> > > [RFC 2396, section 1.1]:
> > >
> > >        Resource
> > >           A resource can be anything that has identity.  Familiar
> > >           examples include an electronic document, an image, a service
> > >           (e.g., "today's weather report for Los Angeles"), and a
> > >           collection of other resources.  Not all resources are network
> > >           "retrievable"; e.g., human beings, corporations, and bound
> > >           books in a library can also be considered resources.
> >
> >Hm. From reading M&S it feels though that Resources are
> >URI-identifiable/-ied things.
>
> Sure, but
> (1) they are the things, not the names that, er, identify them
> (2) it isnt absolutely clear what 'identify' means, but many people
> seem to think that it means 'denote'
> (3) It isnt clear whether being a resource means actually having a
> URI, or only being the kind of thing that could possibly have a URI,
> ie anything at all; your -iable/ied contrast.
>
> >I think this is *the* top-priority issue
> >that we have to clarify and hold on (e-)paper.
>
> It sure would be good to get this clarified.  Mutual
> misunderstandings about these issues seem to get in the way of just
> about every 'semantic webbish' discussion.

2nded! (or 3rd-ed...).

This business seems to be at the heart of much of the murkiness around RDF.

I'm not sure we'll ever have a crisp answer to the question "by virtue of
what social/legal event does a URI string come to denote some entity",
since generations have failed before us with similar questions. But
there's still plenty of room for greater clarity.

Regarding identified vs identifiable, I much prefer the latter (since
anything is identifiable) as it removes us from having to seek answers to
inpossible questions such as the one above.

Dan

Received on Thursday, 12 July 2001 20:58:15 UTC