RE: [xml-dev] Re: Major Historical SOA Milestone Today

["Chiusano Joseph" <chiusano_joseph@bah.com>]

<Quote>
>In addition to those listed, there's also data architecture, which is 
>essentially the representation of the subject areas that define an 
>enterprise's mission, the relationships between them, and the 
>rules/constraints governing those relationships. A data architecture 
>also serves as a "blueprint" of the data that supports an enterprise's 
>mission.

Hmm.  Isn't that the ontology (the concepts plus the relationships among
the concepts that can then be instanced as data)?  I'm not sure the
distinction is there but ... 
</Quote>

I believe we make the distinction that an ontology is suitable for
reasoning (to be reasoned upon) using electronic means such as reasoning
agents. This is at least how I communicate the distinction. So one could
add this distinction to the definition of data architecture above, or
add "unlike a data architecture..." to the definition of an ontology, or
both.

Joe

Joseph Chiusano
Associate
Booz Allen Hamilton
 
700 13th St. NW, Suite 1100
Washington, DC 20005
O: 202-508-6514  
C: 202-251-0731
Visit us online@ http://www.boozallen.com

-----Original Message-----
From: Bullard, Claude L (Len) [mailto:len.bullard@intergraph.com] 
Sent: Thursday, May 11, 2006 2:56 PM
To: Chiusano Joseph; xml-dev@lists.xml.org
Subject: RE: [xml-dev] Re: Major Historical SOA Milestone Today


I am trimming the TO list for politeness.  The others are xml-dev
subscribers.

<Quote>
2.  An architecture is an/a (any takers?).  Assuming this is limited to
architecture as a meaningful term in computer systems, pick one: 
</Quote>

>In addition to those listed, there's also data architecture, which is 
>essentially the representation of the subject areas that define an 
>enterprise's mission, the relationships between them, and the 
>rules/constraints governing those relationships. A data architecture 
>also serves as a "blueprint" of the data that supports an enterprise's 
>mission.

Hmm.  Isn't that the ontology (the concepts plus the relationships among
the concepts that can then be instanced as data)?  I'm not sure the
distinction is there but ...

>Also, what is an example of an architecture that is not
service-oriented? >Some say client-server (but is that really an
architecture?)

I'd say that is a computer architecture given the definitions provided.

Again, not meaningless but meaningful in 'the way that you use it'.

On the other hand, while one can show that all architectures are not
service-oriented architectures, that does not show what a service
oriented architecture is.  So one can (if accepted) look at 1) an act
that returns a value to a user, then ask how many kinds of acts meet
those broad criteria and if other limiting attributes are needed to
distinguish 'service acts'.   It may be that that is the overly
inclusive term until further qualified.

While it is true from one point of view, SOA is an IT concept, from
another, it is a sales concept that can be used to produce a
specification for an IT system.  That is why I say 'domain' because the
intension of use (what is expected in return of using it) is accounted
for.   Given the abstractions for SOA we discussed last time this topic
came up, one can get a lot of good specifications before one ever gets
to the IT means (eg, the abstraction of a business by attributes of the
service types (typically, what kinds of messages are exchanged, their
patterns, etc.)).

len