Re: [ws-caf] On the synchronous request/response model

[Mark Little <Mark.Little@arjuna.com>]

Ricardo, RPC bindings have existed since prior to the original release of
the specifications. The current text is clear that the one-way calls are used 
in the document but other bindings are possible. As to whether we want to 
provide these other bindings, I am unsure (whether RPC is normative or not is 
another issues too). Feel free to register this as a different issue (along 
the lines of what bindings do we want to support in the 1.0 release), but it 
shouldn't be within Context, since this affects all of the specifications.

Mark.

----- Original Message -----
From: "Ricardo Jimenez-Peris" <rjimenez@fi.upm.es>
To: <ws-caf@lists.oasis-open.org>
Sent: Thursday, April 01, 2004 1:00 PM
Subject: [ws-caf] On the synchronous request/response model


> bugzilla-daemon@arjuna.com wrote:
>
> >"Because message exchanges may not be synchronous request/response, but
> >asynchronous one-way messages, some means of associating responses with
> >requests is necessary. Therefore AssertionType has the correlation-id to
> >assist in this association task."
> >
> >
> The specification document states as above that both synchronous
> request/response and
> asynchronous one-way messages are possible. However, the WSDL currently
> only provides the interfaces for the asynchronous model.
> Will we add the WSDL corresponding to the synchronous request/response
> model?
> If so, we could help since we have been already done this transformation
> for WS-CTX and are doing the same for the other layers of WS-CAF.
>
> Otherwise, the spec should only refer to the asynchronous style of
messages,
> the only one currently supported in the WSDL.
> Shall I register this as an issue?
>
> Cheers,
>
> Ricardo
>
> --
> Ricardo Jimenez-Peris
> Associate Professor
> Universidad Politecnica de Madrid
> http://lsd.ls.fi.upm.es/%7Erjimenez/
>
>
>