RE: URI for the concept of SOAP MEP?

> -----Original Message-----
> From: Jacek Kopecky [mailto:jacek.kopecky@deri.org]
> David,
> 
> I think I understand your concern about where this could lead. In
fact,
> I'd be perfectly happy if all WGs that create something machine
> processible had an RDF mapping deliverable in their charters. For
> example one big hole in the WSDL RDF mapping will be XML Schema - WSDL
> only refers to schema things by QNames, which would not be necessary
if
> schema had a suitable RDF mapping.

Heh.  Maybe we should do the converse too.  Every semantic Web spec has
a WSDL/SOAP mapping and has to have deployed endpoints.  Oh wait, I
don't see many soap 1.1 and wsdl 1.1 endpoints in w3c.org land.. :-)  

> 
> On the other hand, there's the question of resources, of course. In
> particular, the WSDL WG requests one URI for one class. The
appropriate
> RDF document describing this particular class should be very simple,
if
> indeed the WG decides to have one. And the WG can decide not to
provide
> a URI for the class, which might become a bit of a coordination issue.
> 
> After we (WSDL) got this one URI, we'll be happy for a while, as we
have
> other URIs from you that are perfectly useful even though you don't
> provide RDF documents for them, and then the W3C can consider how it
> could go about creating RDF ontologies at least for its own things.
> 

"wafer-thin".  Zee slippery slope mah friend.  Maybe this should be done
as a rechartering for XMLP?  Wait, we just finished the XMLP
rechartering last week and this wasn't mentioned AFAICT.  I don't recall
seeing a single comment on this on AC forum..  I wonder if I could troll
through the xmlp charter comments to see if anybody did.   Oh wait, I
can't see that either.  Maybe it will be in the charter.

> Maybe a team member could make such an ontology for any new CR W3C
spec,
> and publish it as a note? This would help with CR and we'd have more
> ontologies. 8-)

I agree with the notion that the folks that care about things should do
the work.

> 
> Hope this helps,
> 
> Jacek
> 
> On Mon, 2006-01-23 at 14:46 -0800, David Orchard wrote:
> > I don't see anything about providing RDF descriptions in our
charter.
> > There's only one mention of RDF and that's wrt data model transports
and
> > serialization.
> >
> > I'm not sure how far this will go and I'm a little concerned about
the
> > precedent.  Do all the concepts that will be in any RDF mapping need
to
> > be in a WG's RDF description?  If the generic problem is that a
> > description format references a concept from another spec and the
rdf
> > mapping requires that the concept have a URI, does that mean that
every
> > time a description format finds it needs a URI that originating
> > committee has to rev it's rdf mapping?
> >
> > Further, does that mean that any WG that produces a spec that might
be
> > described - and probably have an RDF mapping - will need to do RDF
> > descriptions for the RDF mapping of the description format?
> >
> > Cheers,
> > Dave
>

Received on Tuesday, 24 January 2006 03:36:47 UTC