Re: WS, SOA, and the Web

Eric,

I think the distinction between practice and application is indeed lost
on me, but we've miscommunicated before, so I'm not sure I get you
in the first place.

I do think you're mistaken in your characterization of REST as "an
application of an architecture", which makes it sound like REST was
laid down first, then some people implement it to make the WWW,
where it's plain that the actual process was the reverse.  Is that
what you mean?

Could you discuss your "relations" vs. what the "architecture gurus"
call "connectors"?  I'm wondering if these are synonyms.  If so, that
would actually be cool.

Finally, what's the harm in discussing constraints prior to enumerating
all relations in the system?  Is it even possible to do the latter?  I
doubt it.  Aren't relations between things?  Aren't the definitions
of those things laden with constraints?  If not, what distinguishes
one thing from another?  As you can see, your post motivates a
lot of questions in my mind.

-- Walden

----- Original Message -----
From: "Newcomer, Eric" <Eric.Newcomer@iona.com>
To: "Baker, Mark" <distobj@acm.org>; <www-ws-arch@w3.org>
Sent: Monday, May 05, 2003 6:18 PM
Subject: RE: WS, SOA, and the Web


>
> 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.
>
> -----Original Message-----
> From: Baker, Mark
> Sent: Monday, May 05, 2003 9:38 AM
> To: www-ws-arch@w3.org
> Subject: Re: WS, SOA, and the Web
>
>
>
> On Thu, May 01, 2003 at 04:05:15PM -0600, Champion, Mike wrote:
> > > Agreed with concerns about the .. abruptness .. of the SOA vs
> > > REST material.  I think part of the reason it is abrupt is
> > > because doing a rigorous definition of an architecture style
> > > should take a fair amount of text, but we/I haven't put much
> > > more time into it.
> >
> > I sympathize ... I took an action item at an editors call a couple of
weeks
> > ago to work on the "WSA and the Web" section, and kept digging myself
deeper
> > and deeper into the hole.  The problem is that the Webarch, and SOA
> > definitions that I could find are so general that they not only can
> > encompass almost anything (as the Web does, of course) but can also
> > encompass each other, even though some consider them antonyms.
> >
> > I think we need to try to get some clarified text into the "heartbeat"
> > publication.  I have come to  kindof like the "Direct SOA" and
>
> You seem to have sent that out prematurely, Mike.
>
> FWIW, I don't think it's necessary to compare the breadth/generality of
> the competing architectural styles in order to compare them.  We already
> have a framework for comparison provided by the gurus of software
> architecture; examine the properties induced by the constraints of each.
> This should be fairly straightforward.
>
> I'd be happy to help with that text.  In fact, I think we've got some
> of it already, regarding visibility.  I'll have to hunt that down; it's
> well buried. 8-)
>
> MB
> --
> Mark Baker.   Ottawa, Ontario, CANADA.        http://www.markbaker.ca
> Web architecture consulting, technical reports, evaluation & analysis
>
>
>

Received on Monday, 5 May 2003 19:59:39 UTC