re: RDF Core working drafts

At 04:46 PM 5/15/02 -0500, Dan Connolly wrote:
> > Comments on the Model Theory:
> >
> > The model theory makes it very clear that RDF reification is not related in
> > any way to reificiation.  This is good, but does raise the issue of why
> > reification has been retained.
>
>That's a good question. I advocated dropping it altogether.
>   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0218.html
>
>The decision seems to have been made in the 15Feb telcon, with
>discussion in several previous telcons....
>   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Feb/0476.html
>
>The chair took an action to add a "fix/drop reification" issue 11Jan
>   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0095.html
>but I don't see it in the issues list.
>   http://www.w3.org/2000/03/rdf-tracking/
>
>So I'd agree that the WG owes a rationale for keeping reification.

I think there were the following reasons:

(a) that there was a use case involving "partial reification" that was felt 
to be of practical use, even if the semantics were weak.  (I forget the 
details.)

(b) that the reification vocabulary is specified in RDF V1 and is probably 
used in some applications to construct descriptions of statements.  To 
withdraw the vocabulary would be a language change with uncertain effect on 
deployed software.  For example, Jena has interfaces specifically geared to 
dealing with reifications in a model.

(c) even with weak semantics, it is arguably better to have a common 
vocabulary for the intended purpose of describing statements than having 
folks (re)invent their own.

I'd fully agree with anyone who argues that a more appropriate tool here 
would be N3-style formulae, but I think we've concluded that such a 
development is out of scope for our current effort.

#g


-------------------
Graham Klyne
<GK@NineByNine.org>

Received on Thursday, 16 May 2002 05:10:07 UTC