W3C home > Mailing lists > Public > www-ws-arch@w3.org > August 2002

RE: Web service definition

From: Brian Connell <brian@westglobal.com>
Date: Mon, 12 Aug 2002 14:48:12 +0100
To: "Geoff Arnold" <Geoff.Arnold@Sun.COM>, <www-ws-arch@w3.org>
Message-ID: <JMEAIMHKKEFLHHHCKOPBMEFFCGAA.brian@westglobal.com>

Hi,

> Definition: A Web service is a software application identified by a

Don't mean to be totally pedantic, but I believe we should replace the term
'software application' with something different.  A Web Service can be a
sub-part of a 'software application', not necessarily the entire thing.

Interesting question ... does a Web Service only become a Web Service
because it can be defined and described using Web Services standards?  If
this suffices as a definition, then the definition should only revolve
around the standard protocols.

So we should either define Web Services as a *means* of:
defining interfaces and bindings
discovering existing definitions
interacting and communicating with said interfaces.

Or should we also be trying to define characteristics of the service itself
(loosely-coupled, business-oriented)?  Is this not a very important (and
different) part of a definition for Web Services?

Or is it both?  In which case, where are the definitions for
'loosely-coupled' and 'business-oriented'?  Can a software service be a Web
Service if it isn't business-oriented or loosely-coupled?  Obviously, a
software service that doesn't adhere to the Web Services standards is *not*
a Web Service, but is may still be loosely-coupled, etc.

So, should we go the whole hog, and define Web Services in terms of the
characteristics of a service too (loosely-coupled, etc)

Maybe one definition is too ambitious for all of this .. maybe there should
be more than one definition.

Brian Connell
CTO, WestGlobal
Received on Monday, 12 August 2002 09:51:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:25:04 GMT