- From: Jan Grant <Jan.Grant@bristol.ac.uk>
- Date: Tue, 11 Jun 2002 16:15:18 +0100 (BST)
- To: "R.V.Guha" <guha@guha.com>
- cc: Brian McBride <bwm@hplb.hpl.hp.com>, patrick hayes <phayes@ai.uwf.edu>, w3c-rdfcore-wg <w3c-rdfcore-wg@w3.org>
On Tue, 11 Jun 2002, R.V.Guha wrote: > As per W3C processes, we could rule the layering problem out of scope, > etc. and quickly move to completion. But if RDF aspires to be a basis > for OWL and other languages, it has to solve the layering problem. If it > does not, OWL and others will be forced to introduce constructs which > make RDF and them non-monotonic, etc. This is the key issue. I'm strongly in favour of something like this proposal. Current approaches (if there are enough to generalise and categorise as "current") for layering seem to require layered MTs to simply add constraints over the underlying ones. We've already seen the trouble trying to do this properly can cause, in the case of only two layers. It also means that the future invention of new features is stymied, because it requires a retroactive "widening" of the underlying layers*. The timescale for this may be short, but it looks like the best stab thus far at actually telling a convincing story about "layering". I would much rather see a basis for the semantic web that is believable, even if it does cause some process upsets in the short term. jan * or just throwing them away and doing one's own thing, a popular option. -- jan grant, ILRT, University of Bristol. http://www.ilrt.bris.ac.uk/ Tel +44(0)117 9287088 Fax +44 (0)117 9287112 RFC822 jan.grant@bris.ac.uk New Freedom of Information Act: theirs, to yours. Happy now?
Received on Tuesday, 11 June 2002 11:16:25 UTC