- From: Dan Brickley <danbri@w3.org>
- Date: Fri, 9 May 2003 09:19:40 -0400
- To: Jeremy Carroll <jjc@hplb.hpl.hp.com>
- Cc: Brian McBride <bwm@hplb.hpl.hp.com>, w3c-rdfcore-wg@w3.org
* Jeremy Carroll <jjc@hplb.hpl.hp.com> [2003-05-09 14:18+0200] > > > > > If they were just constraining the current RDF/XML syntax, it'd > > be bearable. But > > for this to be their legacy for all future syntaxes seems pretty > > heavy, given > > that the triples are implied. > > > > No - strong disagreement. > > In OWL there are many triples which are redundant because they are implied. > However, it is hard to tell which triples are redundant and which are not. > End users need clear and simply guide lines. One such guideline is > "everything must have a type". Everything does have a type: Resource > For syntaxes which omit redundant triples, then we can easily hypothesis an > intermediate stage that inserts them. And anyway future syntaxes for both If the hypothesising is so easy, why can't OWL do that for our current syntax? > OWL and RDF are likely to either: > - list every triple explicitly depends what you mean by 'every'... > or > - be transformation based > > either way I think your argument fails. I've registered my discomfort with the OWL design, but I can live with compromise. It's time to finish these specs, and a few uglinesses are inevitable. This is one of the them, I guess. Dan
Received on Friday, 9 May 2003 09:32:53 UTC