- From: Shi, Xuan <xshi@GEO.WVU.edu>
- Date: Mon, 30 Jan 2006 10:32:05 -0500
- To: "'Luke Steller '" <Luke.Steller@infotech.monash.edu.au>, "'public-sws-ig@w3.org> '" <public-sws-ig@w3.org>, "Shi, Xuan" <xshi@GEO.WVU.edu>
- Cc: "Shi, Xuan" <xshi@GEO.WVU.edu>
Hello, You are right! It's the activity of the client, not of the service provider! So the semantics of Web services should describe the meaning of the service - WHAT the service can provide, other than HOW the service will process the request, let alone how or what client could do something - the client-side activities should not be a part of the meaning of a Web service. Otherwise, the definition on the semantics of Web service should be changed to how client/requester can integrate varied services together, other than the meaning of service itself. However, service provider could only deal with the server side activities and could hardly cover all possible use cases for the varied clients. That's why I said that the Service Profile or Goals description may be good enough if such documents can explicitly tell the requester what the service can provide and how requester can invoke it. Service aggregation or integration is the activity of the clients thus it has nothing to do with service provider whose service semantics can NOT cover such content. Best wishes, Xuan -----Original Message----- From: Luke Steller To: public-sws-ig@w3.org> ; Shi, Xuan Sent: 1/30/06 6:28 AM Subject: Re: Internet/Distributed Computing using HTTP/POST: Bridge semant ic W eb and Web services under the same Internet protocol Hello, but its not the provider who is interacting with the services, its the client - hence the client needs to know. Luke.
Received on Monday, 30 January 2006 15:32:24 UTC