Steve Brewin wrote:
What I said about fetchmail holds for Maven 2 too.
I'll try this another way. For 2.2.x code, if it ain't broke, don't fix it.
Oh, I agree: the difference is in the definition of broken.
Imho if it is not good enough for your next step then it is broken.
If you have a bike and it is not broken, would you even consider buying
a car? ;-)
We will never achieve fast turn around of incremental releases if we don't
take this approach.
Since I work on James (almost an year) the trunk has never been non
working for more than a week. In fact at step of at most 2-3 weeks I
deployed the trunk somewhere.
If there was much more developers or maybe only PMC votes to make a
release we would already have released something 6 months ago.
E.g: I don't understand why 2.3.0a4 is not published, yet. The current
showstopper has been there since James 1.0, why should this stop an
alpha? We don't pay to release versions, so it should be much more agile.
More revolutionary changes should be saved for major versions. I count both
the Maven and Fetchmail changes in this category.
Revolutionary?
I just can't imagine an user that download James 3.0 and look at the
changelog:
- Upgrade the build system to Maven2
- Refactored the fetchmail to be like before but well formatted and we
now feel better improving james.
Every user would like similar releases.
Imho we should do every step we need to add features, then, when we are
satisfied by new features we plan the testing and the release cycle.
Refactorings and the build tool are only media to the goal: new features.
If we only need bugfixes we can safely apply them to the v2.3 branch and
release 2.3.1, 2.3.2, 2.3.1950...
Regarding the Maven changes, as I previously said, having given Alan commit
priveleges to demonstrate how we might Mavenise, why not give him sometime
to comeback with some results and evaluate them?
So when I say that we should code and then vote you (all) say that we
should discuss and then code. Now that I started a discussion about
Maven2 to try helping Alan understand the PMC position and maybe better
demonstrate his efforts (or maybe to help me do this if he won't do that
for whatever reason) you all say "let's wait for his work"...
No one replied point by point to my 6 reasons. I expected a lot of
+1...-1 results to the single points...
That not a critic on you .. its a critic on anyone.. also me. We must
try to work better together. And not only make critic on each other.
James is a great project but sometimes we take to long ..
And sometimes we forget that deciding to do nothing is sometimes for the
best.
Deciding to do nothing have produced 0 releases in 2 years now.
Maybe we should try something different for 2 years and then evaluate?
Stefano
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
