RE: 16 Mar. Conference Call notice (UML+ -> XML)

[Keith.Finkelde@BTFinancialgroup.com]

Bob & BusProcessTeam

Agree that the Business Processes need to be modelled in another language / modelling convention.

I believe that we all recognise the need for a full MDLC (Message Development LifeCycle ) and a method to guide us through
business analytic representations [problem space], requirements , design [logical solution] then transformation to a technical language [physical solution].

We have been using a range of techniques for current business /industry analysis - covering both UML Use Cases (including DFDs and role interactions)
We have then actively using UML in the logical solution space -  (Structural: Message Class Models + Integrity Constraints + Formating Rules;  Behavioural:  Scenarios, Use Cases ; Temporal : State Transitions) expressed in a combination of Rose models + Excel + Word docs.  XML Authority has been used to transform this to a dictionary ( currently - not able to express all cross field constraints) then export/generate to XML protocols (of whatever dialect) [physical solution].

I believe the TMWG / UMM requires some extensions to better model whole-of-industry problems & also to introduce some genuine analytic techniques (of the As-Is business world) in
earlier stages of the lifecycle.  The current version seems to jump straight into specification of To-Be models, without the ability to trace back.   I attach a review of the UMM and recommendations for extension made previously.

regards

Keith Finkelde 
BT Portfolio Services
email: keith.finkelde@btfinancialgroup.com
phone: +61 2 9259 9765



-----Original Message-----
From: Bob Haugen [mailto:linkage@interaccess.com]
Sent: Thursday,16 March 2000 4:13
To: ebXML-BP@lists.oasis-open.org
Subject: RE: 16 Mar. Conference Call notice


Addendum to BP metamodel suggestions:

I know that some people are not aware that the first
cut of the ebXML BP metamodel will not actually be
in XML (or so I understand, anyway).

Here is my current understanding, stated explicitly:
The first version of the ebXML BP metamodel will be
represented in a combination of UML and text.
XML will come later, derived in some way from the UML.

Questions:
1. Is that correct?
2. Will the UML follow the UN/CEFACT N090 Methodology,
as the Registry group is doing?

Respectfully,
Bob Haugen

UNCEFACTMethod.doc