Sanjiva Weerawarana wrote:
I totally agree these codebases are large and complicated (remember I was the one who was surprised how fast people found that the Sybase codebase was nice and cool). If you really think the best solution is not to force them to merge then let's not go down the single project incubation path. That's a path that'll guarantee that the project will never graduate IMO. I have no problem saying we're going to incubate three different BPEL impls at the same time, but I am sure everyone realizes the reality that that's likely to mean none of them will really capture a dominant place in the world simply because 3's just too many. I guess that's fine too.
If we reversed course and went with multiple, I think that Darwin would "lend a hand" there, so to speak. Code mergers and community consolidation between the like-minded and the willing will take place, and obviously there will be competition for users, contributors and mindshare.
From what I understand to the goal of tight embedding of SybaseBPEL to the ServiceMix JBI container, it plays to a different audience - one that wants to use JBI, and have BPEL included, versus people who wish to use BPEL independently from any given service container or management system. Both are legitimate technical objectives IMO.
I'm interested in working on the latter. geir --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
