- From: Shi, Xuan <xshi@GEO.WVU.edu>
- Date: Wed, 25 Jan 2006 11:33:43 -0500
- To: "'public-sws-ig@w3.org'" <public-sws-ig@w3.org>
- Cc: "Shi, Xuan" <xshi@GEO.WVU.edu>
The fundamental difference between OSRR and others, as I think, may be that OSRR is requester-oriented, while others are provider-oriented. OSRR focuses on how to tell requesters to understand and compose a request to consume the service in an easy and simple way. OWL-S, WSMO, etc. focuses on how service providers handle multi-services together to accomplish the multi-purpose task. For service requesters, if they want to consume Web servcies by OWL-S, WSMO approaches, they need to first understand the complex frameworks and processes described in OWL-S, WSMO, etc. before they can consume the servcies by integrating varied services themselves. The result may be only 5,000/600,000,000 people in this world can consume the so-called semantic Web services. Dr. Battle just called for attention to WSDL 2.0. That's a good footnote - if few people can use it, why do we care about it when HTTP can do what WSDL 2.0 would provide? I think we need to consider more about how requesters can understand and consume the services, other than how providers process the request, and go back to HTTP if it can handle everything we can do in a simple way.
Received on Wednesday, 25 January 2006 16:34:06 UTC