W3C home > Mailing lists > Public > www-ws-arch@w3.org > June 2003

RE: Stateful services (was Web Service Description and stateful services)

From: Savas Parastatidis <Savas.Parastatidis@newcastle.ac.uk>
Date: Tue, 24 Jun 2003 16:34:49 +0100
Message-ID: <BC28A9E979C56C44BCBC2DED313A447001D75880@bond.ncl.ac.uk>
To: "Sergey Beryozkin" <sberyozkin@zandar.com>, <www-ws-arch@w3.org>
Cc: "Paul Watson" <Paul.Watson@newcastle.ac.uk>, "Jim Webber" <jim.webber@arjuna.com>, "Thomas Rischbeck" <thomas.rischbeck@arjuna.com>, <arnaud.simon@arjuna.com>

Sergey,

> 
> It seems to me that the state which you characterized as "3. State as
in
> "data resource"" is a more general case of either 1. Service state or
2.
> Application state
> 
> For example if we have a huge file which needs to be exposed, then if
it
> doesn't matter which customer gets some data from this file
(identified by
> some token), then it looks like that this file is really part of the
> 1.Service state
> However, if a service wants to present to different clients the
different
> views of the same data extracted from the file, then it becomes
> 2.Application/activity state

I disagree.

In the first case, the "service state" is private to the service. Nobody
knows and nobody should care what data the service maintains behind its
interface.

In the second case, the "application state" captures interaction,
application specific information.

In the third case, a resource is exposed through a URI so that it can be
identified and refer to. True, a service may decide to expose part or
the entire resource through its interface, in which case we don't really
care what's behind that interface.

Regards,
.savas.
Received on Tuesday, 24 June 2003 11:35:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:25:21 GMT