RE: Draft language on MEPs, synchronous, and asynchronous.

> I would suggest that being "up at the time" is the ultimate in 
> implementation details.

The point I was making is about "asynchronous" being identified with lack of precise sequencing, and I gave you examples of common use of the term where that identification does not hold.

> -----Original Message-----
> From: Geoff Arnold [mailto:Geoff.Arnold@Sun.COM]
> Sent: Thursday, May 01, 2003 12:09 PM
> To: Ugo Corda
> Cc: www-ws-arch@w3.org
> Subject: Re: Draft language on MEPs, synchronous, and asynchronous.
> 
> 
> 
> On Thursday, May 1, 2003, at 02:31  PM, Ugo Corda wrote:
> > The only part that concerns me is the statement "Other MEPs allow 
> > messages to be sent without precise sequencing, and these are 
> > described as "asynchronous"".
> > The statement seems to exclude important cases where messages are 
> > exchanged using a precise sequencing (e.g. request-response) but in 
> > such a way that the receiver does not have to be up at the time the 
> > submitter sends the message, or the submitter does not have to wait 
> > until the receiver has a response ready and can just collect the 
> > response later (cases that are usually classified as asynchronous).
> 
> I would suggest that being "up at the time" is the ultimate in 
> implementation details. As long as the
> binding described in the WSDL can be satisfied and the message 
> transmitted across the Internet,
> I don't care if it triggers a bell next to the bed of the guy who has 
> to boot up the mainframe
> and enter JCL by hand to start the application..... ;-)
> 
> > In any case, I would rather go ahead with your proposed solution 
> > rather than spending other weeks debating the issue. Someway I have 
> > the feeling that Web services users will happily do their 
> synchronous 
> > and asynchronous exchanges without waiting for our Glossary 
> definition 
> > first ...
> >
> 
> I certainly hope so!
> 
>

Received on Thursday, 1 May 2003 15:17:47 UTC