- From: Katia Sycara <katia@cs.cmu.edu>
- Date: Thu, 8 Jan 2004 15:03:04 -0500
- To: "'Champion, Mike'" <Mike.Champion@SoftwareAG-USA.com>, www-ws-arch@w3.org
- Cc: katia@cs.cmu.edu
Section 2.3.2.5.2 of the current document states that "a service has an identifier", which leads to the unresolved issue in section 2.3.2.5.3 of "what is identified with a service identifier", "service end points, service description, the agent?" A service can have many associated identifiers. The service has a service description that has an identifier, the service can have a UDDI entry that has an identifier, it can have a description of its interaction protocol (expressed in BEPL, for example) that has an identifier, it can have a document expressing additional semantic information (e.g. the service's capability) that has an identifier. Each one of these identifiers could uniquely identify the service (presumably if a requester gets hold of one of these somehow, it knows how to go get the rest if needed). If we agree with this view, I propose to say this in the explanation (section 2.3.2.5.3). This leads to the issue of whether or not keep the text "a service has an identifier" in 2.3.2.5.2. This may also be related to what the relationship of a service and a resource is (which, I think is another unresolved issue). --Katia
Received on Thursday, 8 January 2004 15:03:41 UTC