W3C home > Mailing lists > Public > www-ws-desc@w3.org > February 2006

Fwd: Async Conversations captured in WSDL Model

From: Ramkumar Menon <ramkumar.menon@gmail.com>
Date: Mon, 20 Feb 2006 11:20:07 +0530
Message-ID: <22bb8a4e0602192150j4babadd0kfa0f571c95eea06d@mail.gmail.com>
To: www-ws-desc@w3.org
Oops! I missed the www-ws-desc@w3.org address in the recipient list.
pls take a look at this....

rgds,
Menon

---------- Forwarded message ----------
From: Ramkumar Menon <ramkumar.menon@gmail.com>
Date: Feb 20, 2006 11:18 AM
Subject: Re: Async Conversations captured in WSDL Model
To: Charlton Barreto <charlton_b@mac.com>


Hi Charlton,

 Thanks. I absolutely agree with your point. My only contention was on
the fact that by defining an asyncinteraction, the primary intent is not to
define what services a particular service interacts with, but rather
the *capability
of a Service*, thru the fact that
a) A Service "A" provides the implementation for a portType within the WSDL.
[or in other words, the list of portTypes that  a service implements]
b) Any other  Service can implement another portType within the WSDL so as
to initiate interactions with "A". or be a follower in interactions with
"A". [or in other words, the list of portTypes that partner Services need to
implement, in order to interact with the Service "A".]

i.e., this enables definition of communication endpoints of a Service, in
asbolutely similar fashion as that of synchronous request/reply paradaigm.

The fact that the request  message is being used to initiate an
orchestration is outside the scope of WSDL - Anything could have possibly
done using the initiated message. Orchestration is just one of the things.
 Do let me know your valuable thoughts/feedback on this.

rgds,
Menon

On 2/19/06, Charlton Barreto <charlton_b@mac.com > wrote:
>
> WSDL 2.0 supports a number of asynchronous MEPs. However, as WSDL is a
> model for describing web services rather than how they are orchestrated, I
> don't see the use of supporting the async interaction patterns as you
> mention. These relate to how services are used in conjunction with one
> another, rather than consumer/service interaction. Your reference to BPEL is
> prescient - orchestration languages are where such patterns apply, rather
> than description languages for the services themselves.
>
> -Charlton.
>
> On Sunday, February 19, 2006, at 02:47AM, Ramkumar Menon <ramkumar.menon@gmail.com
> > wrote:
>
> >
> ><<Original Attached>><div>Hi Folks,</div>
> <div>Is there a plan to inherently support description of async request
> response communication&nbsp;primitives within WSDL.[ i.e. rather
> than&nbsp;adding custom extensions]</div>
> <div>This is on the lines of the&nbsp;&quot;partnerLinkType&quot;
> extensions that are used by BPEL Processes.If not, is there a demanding
> reason for the same ?</div>
> <div>&nbsp;</div>
> <div>-Menon</div>
> <div><br>-- <br>Shift to the left, shift to the right!<br>Pop up, push
> down, byte, byte, byte!<br><br>-Ramkumar Menon<br>A typical Macroprocessor
> </div>
>
>


--
Shift to the left, shift to the right!
Pop up, push down, byte, byte, byte!

-Ramkumar Menon
A typical Macroprocessor


--
Shift to the left, shift to the right!
Pop up, push down, byte, byte, byte!

-Ramkumar Menon
A typical Macroprocessor
Received on Monday, 20 February 2006 05:50:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:38 GMT