RE: abstract model

At 09:29 AM 9/15/00 +0100, Jan Grant wrote:
>PS. What are the engineering implications of the Reification mapping
>that Brian proposes?
>- does a statement's reification have a URI? (or more than one, since a
>reification is a resource which may have multiple URIs that symbolise
>it)*

IMO, a resource representing the reified statement can have a URI. There 
may be more than one such resource.

>- in which case is there a way of determining that URI for any
>particular reification?

Sure... look to the corresponding resource.

>- given a URI (or just a resource) is there a way of determining which
>statement, if any, it reifies?

If the reification is complete, there should be only one possible statement 
that it reifies.  The RDF model talks about a _set_ of statement triples.

>PPS. I've no problem with Brian's intuition about "Jan made a statement
>about Dan, but I'm not exactly sure what" - it's a nice idea, but I
>don't think it justifies a "the reification" viewpoint.

I think the (partial) reification is fine.  It's the mapping that I cannot buy.

#g

------------
Graham Klyne
(GK@ACM.ORG)