RE: Mapping Specs to the Architecture

Hi,
Some security work will still MAP easily. Others, that are work in progress
(or wanna be work) may not need to be mapped. 
The excersise will be usefull though.
 
Abbie
 

-----Original Message-----
From: Cutler, Roger (RogerCutler) [mailto:RogerCutler@chevrontexaco.com] 
Sent: Wednesday, March 12, 2003 5:53 PM
To: Ugo Corda; www-ws-arch@w3.org
Subject: RE: Mapping Specs to the Architecture


Right.
 
To be more specific about expectations, my personal take is that SOAP, WSDL
and UDDI should be relatively easy to map, Choreography might be a real
problem, but I think some very interesting insights might result from trying
to map the specs involved with the "ilities".  For example, because of the
work of the MTF folk I think that WSDM is going to map nicely.  But what
about security and RM specs?  SAML?  WS-Security?  WS-RM?  etc.  It seems to
me that they may be a bit tougher to place, and thinking about this may
actually help us to see where we need to expand the framework of the
architecture.
 
Or might not.  As I keep saying, it's just an idea.  Might be fruitful,
might not.
 
 
-----Original Message-----
From: Ugo Corda [mailto:UCorda@SeeBeyond.com] 
Sent: Wednesday, March 12, 2003 4:38 PM
To: Cutler, Roger (RogerCutler); www-ws-arch@w3.org
Subject: RE: Mapping Specs to the Architecture


However, although I do not remember the exact words TimBL used (and anyway
he generally talks in such a way that I might not even be able to define the
concept of the exact words he used), it may be that his question was not,
"Have we provided a mapping of specs onto an architecture diagram", but
"Have we provided a diagram on which specs can be mapped".  That's kind of a
different question and probably a better one.

 
Yes, that's the way I originally understood TimBL remarks. In other words,
it would be nice to have an architecture spec that could identify areas
currently covered by standard efforts and company proprietary specs (e.g.
Choreography - we don't need to identify the individual specs if we don't
want to), areas only currently addressed by company proprietary specs (e.g.
Transactions - again, no need to mention specific specs), and areas not
currently addressed by any specification effort, be it from a standards
organization or from specific companies (e.g. Dave Orchard's proposed
caching of SOA responses for higher network perf).
 
Ugo