- From: David Martin <martin@ai.sri.com>
- Date: Wed, 17 Dec 2003 22:17:25 -0800
- To: Austin Tate <a.tate@ed.ac.uk>
- Cc: public-sws-ig@w3.org
Hello Austin - Austin Tate wrote: > At 21:46 16/12/2003 -0800, David Martin wrote: > >> (1) Make "who does what" more explicit in OWL-S process models, by >> relying on "participant", and adding additional, related, constructs >> as needed to spell it out. > > > > Using a general and extendible constraint formulation we can have > "perform activity actor" constraints (see NIST PSL and PIF) rather than > having to add lots of specialised attributes. > > I am still arguing for a simple underlying structure that just express a > process model as a set of activities and a set of suitable constraints, > and a set of annotations of form key=value. Yes, I understand and am sympathetic to this general approach. But I don't think it's really any different than what I had in mind when I wrote (1) above. After all, an OWL property instance with a value can be thought of as a key=value pair, and OWL certainly provides a general, extendable framework. Further, I am certainly not advocating "lots" of attributes; just the minimum number needed to get the job done. Finally, I don't see that "perform activity actor" is any less "specialized" than the OWL-S "participant" property, or any other property we may find that we need to clarify "who does what". Regards, David
Received on Thursday, 18 December 2003 01:23:03 UTC