RE: BPSS strategic issue: transaction integrity

[Bob Haugen <linkage@interaccess.com>]

Larissa and Dave,

I'm happy to hear that Vitria does not hard code PIPs.
I'm really sorry if I implied differently. 

I asked this question on the lists on Monday, and got no response
about Vitria at that time.  (Also called a bunch of people.) I do not
pretend that constitutes a scientific research program, and am 
always ready to be proved wrong about anything.  At this point,
I am happy to say I was wrong, at least in respect to Vitria.

However, if I was totally wrong about the state of PIP implementations -
if most people don't hard-code them - then that is good news for ebXML.

What I was trying to find out was if ebXML is doing something so
different with interpreted XML for business processes that there
was not enough prior practice to make it realistic.  If there is
significant practice interpreting XML PIP processes then that makes 
the ebXML transaction level very realistic, since it is so close to the PIP
transaction model.

Followup questions:

Larissa Leybovitch:
>I would like to comment that not all the RN PIP software is "hard-coded".
>Vitria's RN solution does NOT have any "PIP hard-coded" components.  Our
>software is based on the generic code (which does not have to be changed to
>each new PIP) that is parameter/table driven and is capable of interpreting
>the XML representations of business processes.  

Excellent! 

>Please adjust your research notes to reflect more accurately the state of
>the solution provider's progress.

I will.

Dave Welsh:
>I'm using Netfish and you're right, we have a super PIP editor. 
>Maybe Bob's refering to something else.

My understand of Netfish's PIP editor from a phone conversation
is that you still need to define each PIP in their editor. Is code
generated, or is the process definition saved in XML and interpreted 
at runtime?  Could you take PIP process definitions in XML from
some other source and run it in Netfish's PIP software?

Thank you all,
Bob Haugen