- From: <jos.deroo.jd@belgium.agfa.com>
- Date: Fri, 6 Apr 2001 22:42:18 +0100
- To: phayes@ai.uwf.edu
- Cc: www-rdf-logic@w3.org
> > > [Drew] > > > In my opinion, this is an extremely messy way to approach what is > > > basically a simple problem. At any point in this downward spiral we > > > can jump ship and switch to a non-RDF language. Indeed, the only > > > reason to stick with an RDF language inside the quotes is to fool > > > ourselves into thinking we haven't left RDF; it's RDF outside the > > > quotes and RDF inside. But the stuff inside the quotes requires all > > > sorts of machinery that we didn't need outside them, so we really are > > > fooling ourselves. > > > >Pop-up an RDF node as/into an RDF graph *in place* (somewhat by-value). > >Its content is not asserted, only quoted in a *non-opaque* way (as RDF). > > That is a contradiction. What do you mean by 'non-opaque' quoting? the *structure* (elements and relationships) inside the ~quoting~ is engine interpretable (non-opaque, transparent) e.g. an N3 context how would you call that (as a verb)? > >We certainly can feed resolution-based logic/proof engines that way. > > Not while retaining consistency you can't. In a sense, of course, you > can input any string of characters you like into any engine you like, > and *something* will happen. Not a very useful sense, though. that *structure* is a *logical form* and not just *something* (otherwise we have of course "garbage in, garbage out") -- Jos De Roo, AGFA ps you seem to have some interesting points about negation, but I have to re-read them (as I was close to the belief that open-world-negation was impossible)
Received on Friday, 6 April 2001 16:42:37 UTC