- From: Dave Beckett <dave.beckett@bristol.ac.uk>
- Date: Tue, 25 Mar 2003 20:34:13 +0000
- To: Jeremy Carroll <jjc@hplb.hpl.hp.com>
- cc: w3c-rdfcore-wg@w3.org
>>>Jeremy Carroll said:
>
>
> Dave Beckett wrote:
>
> > The first question from
> > http://lists.w3.org/Archives/Public/www-rdf-comments/2003JanMar/0226.html
> > is basically, does anyone use bagID?
> >
> > I know the CARMEN project / MathNet uses bagID with RDF/XML and
> > Dublin Core. See the PDF report from Feb 2002
> > http://www.mathematik.uni-osnabrueck.de/projects/carmen/AP6/zwischenberic
ht_ap6.pdf
>
>
>
> I think this document enhances Tim's case.
>
> The only occurrence of bagID in the doc is to create a bag that has a
> matching aboutEach - which we axed.
It isn't the only one, it is just one I knew of, there were hundreds
(I didn't bother to count) more references on google.
>
> Moreover, judging from the surrounding RDF/XML the document author appears
> to want every triple generated within the XML element to be reified into
> the bag, rather than only the toplevel triples (as we clarified).
>
> i.e. this document shows a use of a construct that is not bagID which is
> useful with a construct which is not RDF.
>
> bagID was an ugly sibling to aboutEach and should have been chopped at the
> same time.
and I agreed with you - at the time.
I also put off implementing rdf:bagID in my parser, expecting we
would kill it. But we didn't.
This point is, the WG made the decision so what new information is
there to remove it? That we removed aboutEach? No, that was before.
If we were to remove it, we would owe people a good reason and would
have to make sure it really wasn't used or useful.
If we were having another last call syntax WD, it would mean new LC
WD test cases, LC WD primer to follow the changes, probably not the
others in this case.
Dave
Received on Tuesday, 25 March 2003 15:37:28 UTC