after hours discussion of dark triples [Was: Agenda for RDFCore WG Telecon 2002-04-19]

[...]

> 19: Suggest after hours discussion of dark triples

Since quite some time I am/remain convinced about *unasserted* triples.
I think to have given (be it brief) evidence of that in
-> http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0178.html
-> http://lists.w3.org/Archives/Public/www-webont-wg/2002Apr/0087.html
I couldn't better explain it than Pat in
-> http://lists.w3.org/Archives/Public/www-webont-wg/2002Jan/0195.html
or Tim in
-> http://lists.w3.org/Archives/Public/www-rdf-logic/2001Jun/0124.html
There is just more in the world than asserting RDF triples and
RDF is too useful to restrict it to that.
I also think we need UT sooner than later, so therefore my support.
However I have to admit that we have trouble finding motivating
examples in the realm of OWL. I found Peter's paradox quite convincing.
Main examples are RDF query and premis graphs, but this is indeed out
of scope for WebOnt, however not for the Semantic Web activity I think.
It's quite easy to write a paradox on top of OWL e.g.
  this log:forAll :x .
  { :x a :R } log:implies { :x a [ owl:complementOf :x ] } .
  { :x a [ owl:complementOf :x ] } log:implies { :x a :R } .
and asking
  :R a :R .
and to be honest, I'm not afread of such cases, if we have GLUT
(graph level unasserted triples)

--
Jos

Received on Thursday, 18 April 2002 18:50:02 UTC