RE: [ws-caf] the issue on Context type identification

["Green, Alastair J." <Alastair.Green@choreology.com>]

We do have a proposal, which is my motion from the last meeting, which
has been tabled. It was, to paraphrase: to have an optional
WS-Context-defined context-identifying attribute whose value is a URI. 

I moved to table (U.S. English sense) on the grounds that Greg believed
(and Kevin disbelieved) that this feature was possible to achieve
through existing facilities (attribute extension, if I understood
correctly), and that therefore my proposal was moot, unnecessary,
redundant etc.

On subsequent reflection, I'm not sure that the issue is XML-mechanical.

The proposal to have a value which is opaque is important in this
proposal. According to me, there are two ways you can do this. All the
parties in a particular contractually-governed interaction can agree
that a header element attribute of name N, value type TV means: treat as
a context, or you can write that kind of convention down once in a
single standard specification, and decide what the value of N and TV are
within the standard spec's namespace etc. 

Contract or statute, if you like. 

However, I am not an expert in the mechanics of XML, and I've learnt
that those who are can throw a different light on these kinds of
statements or assumptions by non-experts like me. 

So I am waiting for someone who is a mechanical expert to say: "that's a
wrong-headed way of posing the problem", or "OK, you are right to pose
the problem in that way: implementation choices do not contradict or
alter the overall approach being considered".

If the experts say the latter, then I think the statutory approach is
preferable, which is my motion as it stands. 

This will give Kevin's customers the opportunity to build what he wants,
will give Peter's customers the opportunity to build what he wants, will
give the sceptics the opportunity to do nothing, and will clear the
issue off the table.

Procedurally: if the mechanical experts will speak, then once we have
listened to them we can either vote the tabled motion up or down at the
F2F, or we can have amendments etc to make counterproposals.

Alastair

-----Original Message-----
From: Mark Little [mailto:mark.little@arjuna.com] 
Sent: 23 June 2005 15:09
To: Newcomer, Eric
Cc: ws-caf
Subject: Re: [ws-caf] the issue on Context type identification

I don't think so at the moment, but I'd like to think we could be by the

end of the week if we put our heads together and really drive this as a 
priority.

Mark.


Newcomer, Eric wrote:

>Are we at the point at which we have one or more concrete proposals we
>can vote on?
>
>Eric
>
>-----Original Message-----
>From: Mark Little [mailto:mark.little@arjuna.com] 
>Sent: Thursday, June 23, 2005 6:36 AM
>To: 'ws-caf'
>Subject: [ws-caf] the issue on Context type identification
>
>I'm conscious that there has been very little follow up on the only 
>outstanding issue we have against WS-Context: whether or not (and 
>potentially how) we can identify an element within a SOAP header as a 
>context. I'd like to encourage people to discuss this via email before 
>the face-to-face next week so we can close this issue down quickly and 
>move on to the WS-ACID specification.
>
>Thanks,
>
>Mark.
>
>
>  
>