- From: pat hayes <phayes@ai.uwf.edu>
- Date: Thu, 31 Oct 2002 19:12:47 -0600
- To: connolly@w3.org
- Cc: www-webont-wg@w3.org
Dan, a thought about that issue that came up in the telecon today. How about saying that the distinction encoded in the media type is something like that between *exhibiting* some RDF, on the one hand (the XML type, though I think that's a lousy way to do it: what's it got to do with XML in particular? If someone invents a new syntax for OWL, do they have to label it as XML?? Ah well, never mind. ) and *publishing* it on the other. Exactly what publishing means needn't be made achingly precise, but I have in mind that the latter is intended in some sense to convey the content of the RDF rather than simply exhibit the form. The point of this is that publishing might be (usually will be) asserting it, but it could be querying it, or using it to convey a query, or even in some cases perhaps denying it, if some community starts using RDF to mark up email debates, who knows. But the point is that what is relevant, as it were, to a publication is the *content* expressed by the RDF, however that content is established by whatever criteria are appropriate; so however this RDF is being used in a speech act, as it were, it is more than just a piece of syntax; its not just being 'exhibited'. And as a kind of unmarked default, ie unless there are technical reasons to assume the contrary, a simple publication in an RDF document at the top level on a webpage is taken to be an assertion that the proposition expressed by the RDF is true (not the RDF - too narrow - but the P. expressed by it, note. Which is?? Let the lawyers decide.) This would keep the door open for things like DQL-in-DAML, or even for things like rules-ML-in-OWL, to be 'published' in the exactly the same sense (so using the same media type) but perhaps a slightly morphed sense of 'content' being assumed, appropriate to whatever the form of the stuff being published. And it gets over potential snags like someone complaining that the innards of a log:not are part of the document so must have the same media type, etc. etc. BTW, this needs to be done not just for OWL, but for all SW content languages. Isnt there some way to have a 'publish' media type that would work for all of them? Can a document have more than one media type? So it could be OWL and published, or RDF and published, or whatever. Or is that too much to ask? Pat -- --------------------------------------------------------------------- IHMC (850)434 8903 home 40 South Alcaniz St. (850)202 4416 office Pensacola (850)202 4440 fax FL 32501 (850)291 0667 cell phayes@ai.uwf.edu http://www.coginst.uwf.edu/~phayes s.pam@ai.uwf.edu for spam
Received on Thursday, 31 October 2002 20:13:05 UTC