Re: TDL Model Theory

• From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
• Date: Thu, 24 Jan 2002 14:40:32 -0000
• To: <w3c-rdfcore-wg@w3.org>
• Message-ID: <JAEBJCLMIFLKLOJGMELDAENHCCAA.jjc@hplb.hpl.hp.com>

>               Idiom P in this MT requires the range of a
> property to be a literal-value pair. But that is wrong, for P. If I
> want to say that the range of a property is xsd:integer, I mean to
> say that the range is whatever xsd says integers are, ie the value,
> not a literal-value pair. One hundred and twenty-three is an integer;
> the pair <'123',123> is not an integer (according to XSD).

Short version:

P and S-B have rdfs:range as referring to the value space.
S-A has rdfs:range as referring to the lexical space.
My proposal explores the third possibility of having rdfs:range as referring
to the map.

I don't think there are any a priori reasons that rule that out. It is our
model theory and we are free to choose what parts of the world get modelled
where. Since rdf:type is found in the model I find it more natural to do the
datatyping work in the model.

Longer version:
omitted.

Jeremy

Received on Thursday, 24 January 2002 09:40:34 UTC