- From: Seth Russell <seth@robustai.net>
- Date: Fri, 18 Oct 2002 07:23:24 -0700
- To: pat hayes <phayes@ai.uwf.edu>
- CC: fmanola@mitre.org, "www-rdf-comments@w3.org" <www-rdf-comments@w3.org>
pat hayes wrote: >> In fact, I believe the case you are purporting as the most common >> will actually be so rare, that there should be a prize given to the >> designer of such an agent. > > > Software agent technology is almost routine these days. Amusingly > enough, I am typing this while sitting in a DAML PI meeting listening > to a talk about a large-scale deployed military information system > called SONAT which uses software agents to connect DAML ontologies and > draw conclusions automatically. The field has got to the point where > people are worrying about things like strategies for reconciling > conflicting access policies between agents. > > (BTW, the Cmap tools (version 3) uses what is essentially agent > technology under the hood in order to support real-time collaboration.) Wooh! I didnt mean to imply that heuristic search techniques like those embedded in Prolog are not extremely useful for the semantic web. Rather I actually believe the opposite. Unfortunately my scolorship is deficient here, and I cannot tease apart those useful techniques from the assumptions embedded in your semantic theories given the writings that I can find on the web today. So, shame-faced, I'll just retract my alligation. Seth Russell http://radio.weblogs.com/0113759/
Received on Friday, 18 October 2002 10:24:05 UTC