Re: [ebsoa] New Chair - John Hardin

[john hardin <john@sanghainteractive.com>]

Thanks Monica and thanks to all for the warm welcome.

I have copied a number of folks on this, because I would like to enlist
their assistance for the ebSOA editing activities or just for ideas as 
we move forward. Also Jamie and Dale asked me about addressing the 
ebXML-jc group. I had a good conversation with Jamie the other night, 
and got some direction from both him and from Duane. I would very much 
like to address the ebXML-jc, and get a braindump how the two groups 
need to dovetail. I'll plan on being on the call next Wed if allowed.

My primary vision for the ebSOA group is to move forward with the work 
from the charter (pattern catalog) and act as a facilitator towards
interoperability among the upcoming business process frameworks (WS*, 
ebXML and SemWeb are three that are working to gain dominance). The
request-response model is a very simplistic view that isn't currently
mapping well onto the true complexity that large industries have
developed in their shared business processes. For example, no one will 
argue today that inventory visibility for the automotive industry is a
request-response model between two partners. Healthcare is another area 
that is very complex, and requires multiple interactions between large 
numbers of process participants to complete the information exchange, 
with forks and decisions during the execution.

The other reality is that process participants require constant realtime 
feeds of information from many points in the process, which are owned 
and offered by various companies of various sizes and IT capabilities. 
This spells out, to me at least, the need for a network operating 
framework, where processes run in an "app fabric" that is dependable and 
very open. Then we can layer event causation techniques and other things 
like shared industry process B.A.M. on top. The network REALLY IS the 
computer....

To accomplish this kind of integration, I believe that we need to:

a) work towards interop between the WS and eb camps (ebXML v3 is closing 
that gap rapidly - thanks Dale!)

b) gather and document, then publish and evangelize, real world use 
cases that prove SOA principles using ebXML and WS and shine a light on 
successful implementations

c) include both WS* and ebXML patterns, along with SemWeb and agent
patterns (no rocks thrown, please - the SemWeb/Agent stuff is the 
long-term target)

d) work with WS-I and the specs to ensure that we have consistent
direction towards convergence

My belief is that the entire IT industry is poised on the verge of a
major break in what I think is so far a stagnation. The potential of the
internet is clear - and everyone is aware that it's potential is more
than just human readable web pages. We all feel it, business has been
preparing (somewhat) for it, and the engineers all want to build it.

However, this level of information integration won't happen unless:

a) the path to implementation of standards based messaging between
partner, or producer/consumer, applications, is very clear

b) all players involved in complex processes can participate

c) the integration effort doesn't involve such a long, drawn out mapping
effort everytime (this kills progress on the entire movement)

The industry will continue to stall until we solve some of these
problems. I would like to help by completing the patterns and the
catalog of patterns, and publish the use cases to describe it. There
also needs to be a major outreach effort to all industry groups that are
building shared business processes (which by the way, look much more
complex than simple request-response models).

That's it, I guess....

John
313.930.5323

Monica J. Martin wrote:
> John,
> On a related point (to ebSOA that is), previously Matt MacKenzie was the 
> ebSOA representative to the ebXML-jc.  Perhaps you could provide some 
> insight into your overall vision of the ebSOA TC based on its original 
> charter. I've asked Jamie Clark to give an update on ebSOA and SOA TC 
> efforts. In the past, it was anticipated that ebSOA would assist in 
> defining the next generation architecture for ebXML. Your thoughts would 
> be greatly appreciated. Thanks.
> 
> 
> 
> 

-- 
~~~~~~~~~
john c hardin
sanghainteractive.com
313.930.5323 cell
mailto:john@sanghainteractive.com

"The new electronic interdependence recreates the world in the image of 
a global village."

     Marshall McLuhan, "Gutenberg Galaxy", 1962