Re: [stp-dev] Re: [Open letter to STP regarding BPEL and other verticals

[Kevin McGuire <Kevin_McGuire@xxxxxxxxxx>]

Hi Alex,

Great to hear from you and thanks for
the explanation, good discussion.

The potential for shared technology
between BPEL and BPMN is very attractive.  As they are close, but
as you mention below not identical, this will require some time to understand.


Cheers,
Kevin McGuire

"Alex Boisvert" <boisvert@xxxxxxxxxxx>
Sent by: stp-dev-bounces@xxxxxxxxxxx

03/15/2006 05:40 PM

Please respond to
STP Dev list

To	"STP Dev list" <stp-dev@xxxxxxxxxxx>
cc	mike.norman@xxxxxxxxxxxxx, Mike.Milinkovich@xxxxxxxxxxx, michal.chmielewski@xxxxxxxxxx, Greg Adams/Ottawa/IBM@IBMCA, Ward Cunningham <ward.cunningham@xxxxxxxxxxx>
Subject	Re: [stp-dev] Re: [Open letter to STP regarding BPEL and other        verticals, from the Eclipse BPEL Designer team

Kevin McGuire wrote: 
3) >>I
will leave it to the appropriate parties to comment on BPMN.
This is actually the one that has me the most concerned wrt. overlap and
direction. 
Hi Kevin,

I do share your concern about overlap and would certain like to discuss
direction of the BPEL/BPMN Designers.  When Intalio was approached
late in 2004 related to the STP project, it seemed a good fit to host the
BPMN Designer effort.  If it appears to you that there's a better
place for the BPMN project, I would be happy to revisit this organization
with you and the STP project members.

The goals of BPMN are to provide organizations with better communication
means through standard business process diagrams, and to ensure that these
diagrams embody sufficient information to define and ultimately execute
as BPEL processes.

As such, I do think both BPEL and BPMN Designer will end up with similar
underlying models, although some differences in semantics can lead to complications
and divergent models for levels above BPEL.  Whereas the BPEL designer
attempts to give a high-fidelity representation of BPEL execution concepts,
BPMN starts with similar but different visual concepts that do not map
1-to-1 with BPEL elements.  There's been a lot of recent (blog) discussions
about BPEL and BPMN metamodels and the possibility of having a common MOF-based
metamodel.   The OMG is also trying to realign BPMN within the UML
2.0 metamodel so my guess is BPMN may all of a sudden inherent a significant
formal metamodel from this effort.

As for our current effort, we started with decoupled approach consisting
of a GEF-based UI layer and a corresponding EMF-based model drawn from
the most recent BPMN specification.  There are real challenges in
reconciling the current BPMN + BPEL specs.  When in doubt, we currently
defer to BPEL since this is our first and most important execution target
today, and provides a more formal and stable foundation.   This work
helps provide feedback to the BPMN spec team in terms of the mapping to
BPEL.

I hope this helps your understanding of our direction and current work.
 I would be happy to discuss with you ways to work together so we
can better leverage each project's work and align directions.

regards,
alex
_______________________________________________
stp-dev mailing list
stp-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/stp-dev