- From: Ian Horrocks <horrocks@cs.man.ac.uk>
- Date: Thu, 27 Feb 2003 17:04:11 +0000
- To: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>
- Cc: "Sean Bechhofer" <seanb@cs.man.ac.uk>, "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>, <www-webont-wg@w3.org>
On February 27, Jeremy Carroll writes: > > > The alternative presentation, particularly the approach to equivalence > > and disjointness is, to me, less clear. > > The earlier complexities have gone from the version: > > http://sealpc09.cnuce.cnr.it/jeremy/owl-syntax/2003-21-Feb/dl-syntax.html > > Peter had indicated that he saw semantic difficulties with my earlier > proposal, and I saw that he was right. > > The difference on equivalent classes and disjointness is now simply how to > treat n>2 in such statements, which I don't think is unsuromountable. This kind of thing *is* important when it comes to building tools. E.g., OilEd allows users to assert that a set of classes are disjoint - something that is a pretty common requirement. If saving the file as RDF means decomposing such statements into pairwise disjointness axioms, then when the RDF is read back in it is impossible to know which if any of these should be re-grouped. For users, the result is that every time they go through a save-restore cycle they find all their disjointness axioms have fragmented. This can be both irritating and confusing. I don't think that we give/have given anything like enough consideration to these sorts of practical/implementation considerations in our design. Ian
Received on Thursday, 27 February 2003 12:05:21 UTC