I understand the desire to not having projects use snapshots, but the
reality is you just sometimes need to build against head. The geronimo
team tries to limit snapshots to projects that do instep releases with
geronimo but that is still about 30 modules spread across 4 projects.
Why use snapshots? If you want to track *current* status, you really want
to turn that over to GUMP, which will pull source for every project you
depend upon, and build Geronimo and all dependents from source. GUMP is the
ASF's Continous Integration project, and will be adding (AIUI) testing, as
well as the ability to freeze a given dependent to a released JAR.
Again I understand the desire, but I think it is unreasonable to assume everyone will be using GUMP or more generally be using a continuous integration system. In the case of Geronimo, I personally build Geronimo, OpenEJB and TranQL on my machine, so I can do cross project refactoring.
-dain
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
