Re: ANN: Mozillation project update

At 04:20 PM 4/8/00 -0400, Dan Brickley wrote:
>So (putting on AI skeptic hat ;-) what I'd love to have now is a set of
>practical use case scenarios which show all this logic stuff making itself
>useful for real applications (maybe I can annoy the inference/KR folks on
>this list into building some more cool demos to shut me up!).

I don't know if this qualifies as "real application", since it isn't widely 
deployed yet...

I have an interest in the CC/PP activity to define client capabilities in 
RDF.  I come to this from my work in IETF CONNEG, which takes a far more 
deterministic approach than generalized RDF to matching client capabilities 
and document content.  RFC 2533 describes a feature description and 
matching framework that was modelled on (a subset of) Prolog.

Following the ideas in <http://www.w3.org/DesignIssues/Toolbox.html> I 
would anticipate one could build a similar model, and apply similar 
matching capabilities, to a client capability framework built in RDF.  It's 
not AI, but I think this is exactly the kind of thing an RDF-linked Prolog 
engine could be useful for.

#g


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