- From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
- Date: Mon, 03 May 2004 10:17:23 +0100
- To: "Uschold, Michael F" <michael.f.uschold@boeing.com>
- Cc: Alan Rector <rector@cs.man.ac.uk>, benoit.doumas@free.fr, public-swbp-wg@w3.org, Jeff Pan <zpan@cs.man.ac.uk>, Ian Horrocks <horrocks@cs.man.ac.uk>, Ulrike Sattler <sattler@cs.man.ac.uk>
Uschold, Michael F wrote: > Alan is now suggesting that to this we add another task, which naturally > follows: what things CANNOT be done easily or conveniently or AT ALL > using OWL. This can then be fed back to [?WHO?] at the W3C as > requirements for extensions/updates to OWL. > Alan: > I'd like to come in just to say that in a recent set of travels and > discussions with potential users of OWL, the inability to express > numeric constraints involving inequalities for things such as "Big > Wheel" etc. is a big problem. Virtual all my biomedical examples > require normal values, value ranges, etc. which need this kind of > expressiveness . > It is in scope for the XML Schema Datatypes TF to address the specific issue of how to (slightly) increase the expressivity of RDF and OWL to permit user-defined datatypes (and hence at least a minimal solution to the big wheel problem). Quite how to deploy that solution (in terms of the standards) is yet to be determined ... since there was support for issuing a first datatypes note that raised the questions, I think this is a plausible question to ask the community at that point. Personally, if the necessary changes to any recs were small, I would favour errata against the current recs. That might not fly ... another option might be to have a rec track doc ourselves; or simply to record the consensus and await the next revision of RDF, OWL and/or XML Schema to deploy this consensus within the Rec track. Without having thought about it much, I think small errata is the simplest path, but might meet opposition, and having a rec track doc ourselves would be a mistake. This might leave us with simply documenting the current position and suggested changes (which seems to be Mike's position) I would be nervous about generalizing this - e.g. QCC were considered by webont and rejected, I would be very uncomfortable with this group revisiting that - whereas the user-defined datatypes issue was desired by both RDF Core and WebOnt, but we did not have the effort to put into resolving the problems (i.e. to coordinate adequately with XML Schema WG, and to make sure that any solution for these problems in RDF and OWL does not interfere poorly with work they are doing) Jeremy
Received on Monday, 3 May 2004 05:18:51 UTC