W3C home > Mailing lists > Public > xml-dist-app@w3.org > June 2002

RE: [getf] Node MUST process? (was: [GETF] okay, here's an updateddraftwithHenrik's option B)

From: Henrik Frystyk Nielsen <henrikn@microsoft.com>
Date: Wed, 5 Jun 2002 09:55:50 -0700
Message-ID: <79107D208BA38C45A4E45F62673A434D07D01B0F@red-msg-07.redmond.corp.microsoft.com>
To: "Jean-Jacques Moreau" <moreau@crf.canon.fr>, "Christopher Ferris" <chris.ferris@sun.com>, <noah_mendelsohn@us.ibm.com>, <xml-dist-app@w3.org>


Chris and I noodled on IRC on the bullets and I think came up with the
following points:

A) It should say that it applies to a SOAP node receiving a SOAP message

B) Regarding bullet three, correlation/orchestration may be done in any
number of ways, using MEPs is not the only way. For example, I can have
a set of RR MEP messages being orchestrated in a grander scheme.

C) We should probably refer to features as MEP are also features.

D) Whether they are written as bullets or as text doesn't really matter.


So, with that, here is a slight tweak:

	The SOAP processing model applies to a SOAP node
	receiving a single message only, in isolation
	from any other SOAP message. There is no state,
	correlation or coordination at the SOAP processing
	model level, even, for example if using a feature
	which involves sending multiple messages in
	sequence, each subsequent message depending on
	the response to the previous message.

>So may be we should use my earlier bullet points, I you think
>they're easier to understand? I've loosely based <chris tweak=1/>
>on them, but didn't seem to work for you.
Received on Wednesday, 5 June 2002 12:56:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:10 GMT