On Aug 22, 2007, at 11:43 AM, David Jencks wrote:
While it's more fun to watch from the sidelines, I'm afraid I'll have to get involved here to point out that you are wrong about this. trunk contains an entirely new and incompatible with 6.0.x annotation processing framework which is in use in the Geronimo integration. Remy suggested and I agreed that it would not be appropriate for an api change of this magnitude to go in a 6.0.x revision. While Remy was certainly the committer most involved in applying these changes, IIRC there was some support from at least one other committer. In Geronimo we are anxiously awaiting an official tomcat release with these agreed upon api changes, so we tend to view the proposal eliminate trunk with some concern.
All this really depends and hinges on what the current development community sees as the "usage" of trunk... For example, with httpd, trunk also contains incompatible changes from the current latest shipping release (2.2.x). But that's OK because trunk is actually what will be used as the basis for 2.4, where there are no API guarantees. BUT, and this is an important but, changes to the 2.2 branch are done via backports *from* trunk. 2.2 doesn't have it's own trunk, so to speak, but uses main trunk. This is all well and good if that process is understood and used by everyone... Seems to me, this is a main issue with this whole debate. What "is" trunk... --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
