Re: [RTF] Glossary entry ""metric of Architectural component"

On Tue, Jul 02, 2002 at 05:09:10PM -0500, Damodaran, Suresh wrote:
> I don't think I understand your comment, Mark.
> Hardware is used today to implement security techniques. See  [1] for a
> sample.
> I will not speak for some of the companies involved who are well represented
> in this WG, but I definitely think some components of Web Services 
> will be implemented in hardware.

Oh, most definitely.  I never meant to suggest otherwise.  But
"component" is not necessarily - using your term - an architectural
component.  For example, a random number generator (hardware or software
based) would not be an architectural component in our architecture, but
might be called a "component" by some people.

> It is not appropriate to say that WSA is
> a "software" architecture. Just "architecture" would suffice, IMO.

Well, no, I'd disagree quite strongly with that.  What we're designing
is a software architecture.  But some of its software components can
be deployed within hardware.  For example, dedicated hardware built to
be a high performance proxy would be considered an architectural
component by REST.

> Having said this, I will let the h/w company representatives to pick up this
> debate
> if they chose - this is not something I want to spend time debating.

I agree that there's not a lot of value in debating hardware versus
software.  My main objective here though, is to ensure two things;

- that we reuse as much existing terminology in the field of software
architecture as we reasonably can
- that the terminology we do use, whether new or existing, is self-
consistent

MB
-- 
Mark Baker, CTO, Idokorro Mobile (formerly Planetfred)
Ottawa, Ontario, CANADA.               distobj@acm.org
http://www.markbaker.ca        http://www.idokorro.com

Received on Tuesday, 2 July 2002 20:21:42 UTC