- From: Monica J. Martin <monica.martin@sun.com>
- Date: Thu, 17 Jul 2003 12:45:33 -0600
- To: Martin Chapman <martin.chapman@oracle.com>
- CC: Francis McCabe <fgm@fla.fujitsu.com>, Steve Ross-Talbot <steve@enigmatec.net>, "Champion, Mike" <Mike.Champion@SoftwareAG-USA.com>, public-ws-chor@w3.org
> > >Chapman: even though we don't have agree teminology yet perhaps we sould start using >the term agent instead of party or participant. > mm1: Would this create any confusion if we also understand there could be an agent that can be queried to acquire state? This was mentioned when we discussed state synchronization in the F2F. > > >>Frank McCabe: Following the form of the WSA, we have agents that provide and request >>Web services. This is based on the intuition that a service is >>fundamentally about the potential for action, and that actors >>(computational and otherwise) are the entities that do the acting. >> >>Chapman: I think there is a fundamental terminology issue here that needs to be >>cleared up. An entity (avoiding any overloaded word) that sends a message to a web >>service (and may expect a response depending on the wsdl) doesn not >>iteslf have to be a web service. This is the most fundamental building block. >>Furthermore this interaction supports an MEP (in soap teminology) and >>pattern (in wsd teminology). Perosnllay if we can not describe these meps in a choreography >>language we have failed, and hence I do not think that mep choreogaprhy is any >>different from web service choreography. >> >> >>
Received on Thursday, 17 July 2003 14:33:33 UTC