W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > July 2001

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

From: Sergey Melnik <melnik@db.stanford.edu>
Date: Tue, 10 Jul 2001 14:52:33 -0700
Message-ID: <3B4B7921.3F786BFB@db.stanford.edu>
To: pat hayes <phayes@ai.uwf.edu>
CC: bdehora@interx.com, w3c-rdfcore-wg@w3.org
pat hayes wrote:
> >Pushing aside resource mumbo jumbo (red herrings).
> I disagree; the problem is to get this mumbo jumbo consistent, since
> our remit is to make the M&S spec coherent.
> >To me, it
> >comes down to these 3 questions:
> >
> >1: Feasibility: can we give RDF Literals URIs in the M&S and
> >stay within the charter?
> Yes, but....
> >2: Validity: should we give RDF Literals URIs in the M&S?
> No, because....
> >3: Purpose: if we did give RDF Literals URIs, what are we
> >using the RDF to describe? (it's not at all clear to me)
> If you use the literal, you are describing a value directly; if you
> were to use the URI of the literal, you would be referring to a
> representation of the value. And there really isn't much utility in
> doing that, as far as I can see. Still, you *could* do it, and you
> would be referring to the same thing if you did so.

If a literal constant is just a value from {"urn:data:literal:",
Unicode*} as suggested in
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Jul/0042.html, we
should not need an additional identifier.

> But further to 2.; while we should not give literals URI's , they are
> nevertheless resources. Thats all we need to say.
> Pat Hayes
> PS. Heres how Ive been translating this debate internally, which may
> help get the central issues clear.
> resource  <---->  entity (ie anything at all, anything in the
> universe or out of it, anything that can possibly be referred to or
> quantified over; anything that can be conceived of in the mind of
> Man, God or Tim B-L; the most general category possible.)
> URI  <------>  name, referring expression
> literal  <----->  numeral, a special kind of referring expression
> from which the referent can be computed.
> [...]

I think we need a glossary like that at some point! Mine is:

entity (constant) = anything identifiable
resource (constant) = URI, name, referring expression
literal (constant) = like yours

Received on Tuesday, 10 July 2001 17:26:09 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:53:50 UTC