- From: Frank van Harmelen <Frank.van.Harmelen@cs.vu.nl>
- Date: Thu, 17 Jan 2002 21:52:47 +0100
- To: Nicola Guarino <Nicola.Guarino@ladseb.pd.cnr.it>
- CC: GUIDO VETERE <gvetere@it.ibm.com>, Mike Dean <mdean@daml.org>, www-archive@w3.org, Guus Schreiber <guus@swi.psy.uva.nl>
Nicola Guarino wrote: > I agree with the last point: there is no need to introduce a relation > as primitive if it can be axiomatized. However, you need a suitable > expressivity for axiomatizing it (for instance, to express the > supplementation axiom). Just to keep you posted: one of the outcomes of the Web Ontology Working Group meeting at Bell Labs on Mon/Tue was the decision not to include part-whole explicitly in the language, but to make sure that it could be axiomatised. (I believe it can in fact already be axiomatised in DAML+OIL, including the supplementation axiom). > >Finally, I insist on the need of introducing 'type' vs. 'role' as modifiers > >for class declarations. This distinction (that I learned from your works) > >woud be crucial for for many practical reasons. Have you any comment on > >that ? > > Of course we need these things. Type vs. role is certainly a very > important distinction for an ontology language. This distinction did not make it into the requirements list that was determined at the meeting. In fact, it never even came up. It also did not show up in any of the few dozen use cases that the working group collected. If you want to push for this, bring this to the attention of Guus Schreiber <guus@swi.psy.uva.nl> (who I've cc'ed for this msg). Thanks for your continued input on this issue! Frank. ----
Received on Wednesday, 16 January 2002 15:56:17 UTC