- From: Graham Klyne <Graham.Klyne@MIMEsweeper.com>
- Date: Fri, 25 Jan 2002 16:40:01 +0000
- To: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>
- Cc: <w3c-rdfcore-wg@w3.org>
At 02:04 PM 1/25/02 +0000, Jeremy Carroll wrote:
>Further clarification ...
> > So, in effect, the denotation of *any* node is a pair (in any model of an
> > RDF graph)?
>
>Nodes labelled with URIs map to resources
>Nodes labelled with strings map to pairs
>Blank nodes can map to either resources or pairs.
Er, what happens with:
ex:someURI rdf:value "12345" .
?
[That's a repeat of the question I asked in IRC - I think it needs to be
clarified.]
> > Ah, I think I get it. In any model (which in your MT is defined relative
> > to some set of datatypes as well as the URI vocabulary), the
> > interpretation
> > can be _any_ pair of <literal,value> that is a member of some datatype
> > mapping. rdf:type and schema statements may have the effect of
> > eliminating
> > some potential candidates from the set of models?
>
>Almost correct.
>The doc allows _any_ pair (whatsoever).
>There is no requirement that there are any datatypes (an earlier draft,
>which I think you looked at, used xsd:string as a default, that's been
>dropped, I had made a mistake), so there is no requirement that the pair
>belongs to a datatype.
Ah, I see. That was an unwarranted assumption on my part (based on the
interpretation being with respect to some set of datatype mappings). Yes,
I now see why that doesn't work.
#g
------------------------------------------------------------
Graham Klyne MIMEsweeper Group
Strategic Research <http://www.mimesweeper.com>
<Graham.Klyne@MIMEsweeper.com>
__
/\ \
/ \ \
/ /\ \ \
/ / /\ \ \
/ / /__\_\ \
/ / /________\
\/___________/
Received on Friday, 25 January 2002 11:44:37 UTC