RE: WS, SOA, and the Web

Yes, I agree, I'm just saying I don't think we're at the constraints stage yet with regard to WSA, we are still defining relationships. 

To answer Walden's query, what I'm saying is that we need to define the relationships among the architectural elements first, and that I would agree that components, connectors, and data are indeed examples of architectural elements.  I am not sure the degree to which these elements are applicable to WSA, and think we need to consider the possibility of including other elements.

Really the comment is more about recognizing the stage in the development of the architecture that we're in, I'm not trying to say constraints aren't important.  But they are more important during what I've called the "application" of an architecture to solve specific problems.  

We are progressing the WSA spec more or less along these lines by defining the core concepts (or elements) and relationships and then describing various views or perspectives on those concepts and relationships.  These views relate to various applications of the core concepts and relationships and constraints and properties are significant here because they are brought into the picture by the views (or applications).

Hopefully this is in line with generally accepted thinking about architectures and their application to specific technical problems.

I would only add as a kind of footnote that WSA seems much broader than the Web architecture, since it extends beyond the Web.  This seems more likely to be the bone of contention to me than the core concepts, relationships, and views.

Eric

-----Original Message-----
From: Baker, Mark 
Sent: Tuesday, May 06, 2003 1:08 AM
To: Newcomer, Eric
Cc: www-ws-arch@w3.org
Subject: Re: WS, SOA, and the Web


Hi Eric,

On Mon, May 05, 2003 at 06:18:59PM -0400, Newcomer, Eric wrote:
> Actually I think we are in danger of losing the distinction between architecture as a practice and its application.  REST is an application of architecture, therefore constraints and properties are central.  However, architecture is more generally a study of relationships, and since we are not prescribing the application of Web services architecture, it is a mistake to begin with constraints when we need to begin with the definition of the relationships.

But "constraints" are constraints on relationships;

"A software architecture is defined by a configuration of architectural
elements--components, connectors, and data--constrained in their
relationships in order to achieve a desired set of architectural
properties."
 -- http://www.ics.uci.edu/~fielding/pubs/dissertation/software_arch.htm#sec_1_2

It seems we're in agreement.

MB
-- 
Mark Baker.   Ottawa, Ontario, CANADA.        http://www.markbaker.ca
Web architecture consulting, technical reports, evaluation & analysis

Received on Tuesday, 6 May 2003 11:38:30 UTC