Re: I have a trouble with The RDF Model

Pat,

 You (and others) have raised the question of what exactly the "graph"
(the directed labeled graph which is often drawn in rdf discussions) is.

 I have a simple (albeit potentially controversial) answer --- it is the
range of the interpretation of RDF expressions.

 Guha

pat hayes wrote:

> >At 09:44 AM 11/27/00 -0600, pat hayes wrote:
> >>The domain of a model theory, more or less by definition of the
> >>term "model theory", is the expressions of the language (or perhaps
> >>more exactly, the parsings of those expressions according to the
> >>syntactic rules of the language.)
> >
> >I'm not sure I'd know a model theory if it leapt up and bit me, but
> >that's a useful start for me.  What would you say is the "range" of
> >a model theory?
>
> I'm glad you asked. The range is the interpretations. Thats what a
> model theory does: it defines (a mathematical model of) what counts
> as an interpretation of the language, and gives rules for how to
> interpret the expressions of the language in such an interpretation.
> (Other names for interpretations are 'possible world', 'set of
> circumstances' 'state of affairs' and 'model of the world'.) Exactly
> what counts as an interpretation, speaking now mathematically,
> depends on the language; more intricate langauges require more
> complicated notions of 'interpretation'. Propositional logic requires
> simply an assignment of truth-values to the basic proposition
> letters. First-order relational logic requires a set over which the
> quantifiers range, and denotations defined over this set for all the
> names and relation names in the language. Programming languages have
> traditionally required functional domains obeying certain fixpoint
> properties. Modal logics require more complicated interpretations
> with multiple domains linked by accessibility relations, and so on.
> 'Ontology' lanugages like DAML, which essentially describe a
> heirarchy but don't say anything much about it, seem to require only
> some sets with some relations between them as an interpretation, as
> in Peter Patel-Schneider's OIL semantics. Right now Im not yet sure
> what RDF really needs, but Im working on it.
>
> >>So I repeat: are you saying that the 'at' assertions are part of RDF, or not?
> >
> >I'd say not, but that it is possible to _model_ the 'at' assertions in RDF.
>
> The trouble with that answer is, I really do not know what it means.
> What sense of 'model' are you using? Do you mean it is possible to
> *describe* them in RDF? Or that it is possible to *simulate* them in
> RDF? Or that they are some kind of assertional *extension* to RDF? Or
> *axioms written in RDF syntax*? Any help would be appreciated.
>
> Pat Hayes
>
> ---------------------------------------------------------------------
> IHMC                                    (850)434 8903   home
> 40 South Alcaniz St.                    (850)202 4416   office
> Pensacola,  FL 32501                    (850)202 4440   fax
> phayes@ai.uwf.edu
> http://www.coginst.uwf.edu/~phayes

Received on Thursday, 30 November 2000 14:05:13 UTC