RE: Web Service Definition [Was "Some Thoughts ..."]

Roger,

I think one has to be flexible about orchestration.  I can think of web
services that encapsulate complete orchestrations and infact provide
discrete services.  I don't think that anybody would argue about whether
they are web services or not.  

The confusing part is a service that fulfills all other requirements of a
web service (uri, standard web protocols, non-human interactions e.t.c. )
with the only caveat that its methods have a state associated with them and
there is a legal sequence for invoking its methods.  In my opinion, it is
still a web service.  Take the example of WSFL.  The flow models defined in
WSFL have infact a portType associated with them, that describes the
interface.  However, the exposed operations in that TYPE can only be invoked
in a state dependent legal sequence.  It is still a web service however.

Thanks,

 
_______________________________________________
Waqar Sadiq
 
EDS EIT EASI - Enterprise Consultant
MS: H3-4C-22
5400 Legacy Drive
Plano, Texas 75024
 
phone: +01-972-797-8408 (8-837)
e-mail: waqar.sadiq@eds.com
fax: +01-972-605-4071
_______________________________________________
 
 

-----Original Message-----
From: Cutler, Roger (RogerCutler) [mailto:RogerCutler@ChevronTexaco.com] 
Sent: Friday, March 01, 2002 10:50 AM
To: 'Vinoski, Stephen'
Cc: www-ws-arch@w3.org
Subject: RE: Web Service Definition [Was "Some Thoughts ..."]

I am reluctantly coming to agree with your last point.  The reluctance comes
from my perception that it involves distinctions that are difficult to make,
but I'm beginning to think that it really is necessary.  I came up with a
fairly lengthy analysis of why my suggestions you were replying to (and
thanx for your helpful responses) are in fact not a good idea -- which I
won't bore you with -- but the bottom line was more or less that you can't
avoid the non-human issue.

About orchestrations, I don't think anybody disagrees that they are
important and that we should be considering them.  The point I'm making is
that it seems that most people want to define a "web service" in such a way
that orchestrations are not themselves web services but are composed of
them.  I could go either way on this -- I'm just trying to go with the
consensus here.

-----Original Message-----
From: Vinoski, Stephen [mailto:steve.vinoski@iona.com] 
Sent: Thursday, February 28, 2002 6:50 PM
To: Cutler, Roger (RogerCutler)
Cc: www-ws-arch@w3.org
Subject: RE: Web Service Definition [Was "Some Thoughts ..."]

[ deleted portion ]

> Add that the web service is addressed by a
> URI and you
> eliminate a bunch of other things, like orchestrations, that 
> are not web
> services per se.

I disagree. I am a big fan of orchestrations as related to Web Services, and
thus would not promote a definition that disallows them. URIs do not
disallow orchestrations.

> It seems to me that then you would have a
> definition that
> it would be easy to use to answer the question: "Is XYZ a web 
> service?"
> without getting involved in difficult discussions like 
> whether the thing is
> intended for machines or people, or people interacting 
> through machines, or
> machines using people, or ... 

In web terms, a Web Service *is* defined by the fact that it evolves the Web
beyond browser/web server interactions to true application-to-application
interactions that do not directly involve humans. You can't define Web
Services without addressing this key element.

--steve

[ deleted portion ]

Received on Friday, 1 March 2002 12:17:45 UTC