- From: Mark van Assem <mark@cs.vu.nl>
- Date: Thu, 01 Jun 2006 23:21:38 +0300
- To: Dan Brickley <danbri@danbri.org>
- CC: Antoine Isaac <Antoine.Isaac@KB.nl>, Alistair Miles <a.j.miles@rl.ac.uk>, public-esw-thes@w3.org
Hi Dan, > It's entirely natural and healthy for the machine-readable bits not > to capture everything. Otherwise, we'd have no reason to go on inventing > ever richer formalisms :) (OWL, RIF Rules, ...) Exactly. And that's why the argument Antoine brings to the table seems to be not one that should prevent us from introducing a class skos:Term. I would be interested to hear your opinion on the Term-as-class issue as co-editor of SKOS. But let me re-iterate that I think the most important thing is to *flag* the issue now. And, in a future draft, we should at the very least make clear explicitly which choice we make. I can respect the fact that an as simple model as possible will help uptake which can be more important than being complete. I do not (currently) agree with such a decision but I can live with it. However, as we are knowledge modellers, we should make such a choice explicit and explain that we do not support some features of a class of thesauri, among which are prominent ones such as Agrovoc and MeSH. Mark.
Received on Thursday, 1 June 2006 20:21:46 UTC