W3C home > Mailing lists > Public > public-ws-addressing@w3.org > November 2004

RE: i0001: EPRs as identifiers (why XML?)

From: Jim Webber <Jim.Webber@newcastle.ac.uk>
Date: Mon, 22 Nov 2004 23:22:44 -0000
Message-ID: <37E80E80B681A24B8F768D607373CA8001854341@largo.campus.ncl.ac.uk>
To: <public-ws-addressing@w3.org>

Paco:

I'm having difficutly in understanding how WS-Addressing is useful at
all to anything other than SOAP. 

[ SOAP Payload ] == message, transfer mechanism TBD

[ SOAP Payload + WS-Address ] == message ready to go on some underlying
network protocol

Versus say:

[ CORBA payload + IIOP info ] == message ready to go via TCP

Or

[ XML payload + HTTP headers] == message ready to go via HTTP

So I see that other transfer protocols have addressing information as
part of their unit of transfer by default. SOAP doesn't have that, hence
the need for a SOAP-friendly addressing mechanism. So I am a little
puzzled as to how WS-Addressing is useful outside the SOAP domain. Can
you exemplify something from your experience to help?

As for WSDL being so "flexible," well that sure goes some way to
explaining how complex it is. Had WSDL gone down the route of describing
SOAP message exchanges, then I don't think we would have all be
thoroughly entertained by Rich's recent article. Perhaps Rich will
commit to a similar article on WS-Addressing if it becomes all-embracing
rather than focussed and useful :-)

Jim
Received on Monday, 22 November 2004 23:26:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:34:59 GMT