Lance Waterman wrote:
As one of the Sybase BPEL developers, this certainly sounds reasonable to
me. Perhaps we can start a discussion around what the "bake off" criteria
will look like ( i.e. what BPEL constructs are fully/half/not supported,
what set of unit tests should be supported, etc ... ). As an example; in
looking through the Agila code ( which could have rendered faulty
assumptions ) it appears that the BPEL "Scope" construct is not currently
supported. I rate "Scope" as quite important to have in a BPEL
implementation and its not a simple thing to implement. Likewise, I am sure
there are criteria that the Sybase donated engine does not fully measure
against.


While I'm not sure how far along the BPEL engine in Agila is (Twister), I wouldn't be surprised if it wasn't at the same level of completeness as your implementation.

geir

Lance

On 2/14/06, Geir Magnusson Jr <[EMAIL PROTECTED]> wrote:


Bill Flood wrote:
Geir, approaching Agila was our first avenue.  We looked at what they
had
and I initiated several conversations about donating to that incubator
project.

We offered a base line upon which to build but there did not seem to be
any
uptake although both committers said they were happy to have us come in
and
provide coding help on what they already had.
I don't really know.  I wasn't part of that conversation, although I did
have a few discussions w/ people after this all started.

I expect that what you just said might have been the problem.  They
already had a BPEL engine, and it sounds like you were suggesting they
stop and reboot on your codebase.  After all, they have some users and
wouldn't want to just drop them.

I was a little mystified.
Jumping in and bringing their stuff up to where we already were seemed
counter productive given the large gap in code maturity and capability
so we
passed.

Based on the support we have seen for our contribution from others in
Apache
thus far, I have to believe that our impression wasn't just the result
of
our inherent subjectivity

I believe we did approach them in good faith and I'm not sure why there
was
disinterest in our offer via Agila but here we are.  We left the
previous
conversation on good terms.  At this point, my preference would be that
the
Agila folks look at the contribution and see if they want to become part
of
that larger community for this new baseline.  To me, it's not about
ownership, it's about critical mass in the community to carry something
forward.
How about making a fresh start then...  If the Agila people are
interested, put out a call for any and all other implementations of BPEL
that might be donated and build a larger community, mixing the best of
anything that is donated to get the best BPEL engine and community we can?

In the same way that we built Geronimo from "best of breed" J2EE-ish OSS
projects that are out there, I'm sure we could do a similar thing with
BPEL.

Maybe do a "bake off" to help find the best codebase, and have the
community collaborate around that?  (I'm not sure what that would
entail, actually...)

Would you be interested in that?

geir



---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to