xml-dev - Re: RDF-Schema (was Re: Why validate?)

[Dan Brickley <Daniel.Brickley@bristol.ac.uk>]

On Thu, 18 Nov 1999, David Megginson wrote:

> Michael Erdmann writes:
>  > In Robins mail about the semantic level of XML as well as what
>  > David and Tim write about the inadequacy of DTD to represent
>  > meaning, I missed any references to RDF/RDF-Schema (BTW a REC and a
>  > PR of W3C), that ---as I understand--- exactly are aimed at
>  > providing a formal, ontological, shareable semantics in the world
>  > of XML.
> 
> Or, you could argue that RDF-Schema simply defines a different kind of
> structure -- instead of saying what elements may contain, it says what
> classes properties may appear in.  It's the same thing, then, but in a
> more specific problem domain (the object world).

I've been thinking a lot lately about how to characterise what RDF
Schema tries to do. I think it falls into two parts: firstly, we say
'use RDF to describe itself', and secondly, we provide some simple RDF
vocabulary constructs to begin this task. So, the second aspect of RDF
Schema is (for a different domain) broadly analagous to the
elements/attributes level of XML DTDs. But the first aspect of RDF
Schema, ie. building something more intrinsically self-descriptive,
feels different to what we have in DTDs.

That's longwinded; I'll try again. The current facilities actually
provided in RDF Schema are roughly as expressive as DTDs, but the
information model behind what RDF Schema 1.0 gives us is
more easily extended. For example, much richer constraints for RDF
schemata should be pretty easy to define once we have an RDF query
language useable for expressing them.

> That said, I'm a big RDF-Schema fan -- it's quite readable, and I'm
> finding some serious interest in both RDF and RDF-Schema out there in
> the business world.  I'd very much like to see RDF-Schema go to REC,
> even though it's missing some things I think I need.

Do you think these can be layered on afterwards in subsequent
specifications? If so, that's the best place for them.

>  > Or do I got something completely wrong. 
>  > 
>  > Or is XML-DEV not the right place to argue in favor of RDF/RDFS?
> 
> This is the best place.  There's also an rdf-dev list, but nobody
> posts to it.

Yeah but...

RDF-DEV / RDF Interest Group mailing list update time...

As RDF-DEV listowner I decided to merge that list in with the new
RDF Interest Group list we've just set up. The W3C RDF Interest Group
pages (including archives and list signup details) are at:

	http://www.w3.org/RDF/Interest/

Now the RDF-DEV / RDF IG list convergence is mostly behind us it seems a
good time to circulate a call for partipation in the RDF Interest Group.
(For a while I dithered about shutting down RDF-DEV but was definitely
the right thing to do; this was why it took until now to get the RDF IG
details more broadly circulated).

The RDF Interest Group charter is at http://www.w3.org/RDF/IGcharter

We've had 100 posts in 14 days on RDF Interest Group list[1]; that's
roughly what we had in 3+ months on RDF-DEV.

More list-members always welcome. To subscribe to the RDF
Interest Group mailing list, send a message: 

	To: www-rdf-interest-request@w3.org
	Subject: subscribe


Recent discussion has included simplification of RDF syntax, storing
RDF models in relational databases, cambridge communique, proposals for
RDF APIs, and various other thinkings-out-loud...

Dan



[1] http://lists.w3.org/Archives/Public/www-rdf-interest/


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo@ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)