- From: Bijan Parsia <bparsia@isr.umd.edu>
- Date: Wed, 31 Mar 2004 18:39:11 -0500
- To: David Martin <martin@AI.SRI.COM>
- Cc: public-sws-ig <public-sws-ig@w3.org>
On Mar 31, 2004, at 2:57 PM, David Martin wrote: > Bijan - > > Thanks for commenting... > > Bijan Parsia wrote: > >> Quick response. >> On Mar 31, 2004, at 2:54 AM, David Martin wrote: >> [snip] >>> <process:hasEffect> >>> <???:Formula> >>> <???:inLanguage rdf:datatype= >>> "&xsd;anyURI">...swrl...</???:inLanguage> >> I wouldn't use a literal here. The problems that faced us with >> parameterTypes don't apply. I expect swrl et al to have uris and >> could be made to be of type, say, LogicFormalism. > > So then someone would declare the LogicFormalism class somewhere, and > someone would also declare an instance of that class corresponding to > SWRL (or DRS or SWRL++ or whatever) and then we'd mention that > instance as the value of inLanguage (or whatever better name we come > up with for that property). Is that what > you have in mind here? Exactly. [snip] > On some property wrapping ???:conjuncts. (well, conjuncts would need a > parent node then). Hmm. That makes it a touch less appealing. The > problem is that *our* "formula" really is a wrapper for other people's > "formual" with extra metadata. So a bit of repetition seems very hard > to avoid. > > Yeah, that's what I was thinking also. Cool. Cheers, Bijan Parsia.
Received on Wednesday, 31 March 2004 18:43:37 UTC