On Jan 12, 2008 3:37 PM, Rahul Akolkar <[EMAIL PROTECTED]> wrote: > On 1/12/08, Jochen Wiedmann <[EMAIL PROTECTED]> wrote: > > On Jan 12, 2008 3:25 PM, simon <[EMAIL PROTECTED]> wrote: > > > > > I don't see the point of that at all. Commons-xxx projects do not exist > > > for the purposes of testing maven plugins. If commons-xxx can build > > > successfully with version N of a plugin, then there is no reason to ever > > > use any other version of the plugin. > > > > > > Well, one minor case is with report plugins, so that the reports > > > generated by related modules look similar. But maven can't help with > > > that anyway, as it doesn't provide <pluginManagement>. > > > > Well that boils down to the question, whether we should have a > > commons-parent or at least whether we should have only one. For > > example, we might as well have a pom with reports only. > > > > I am personally quite happy about it because I have the feeling that > > it helps me. That's the reason why this discussion engages me. Others > > (including you?) don't seem to keen on using its features. > > > > I have no particular feelings if a subproject decides to use a > > particular version of the commons-parent pom or not to use it at all. > > OTOH, I'd ask that those who want to use it should be left more alone. > > > <snip/> > > Also note: > * I believe RMs are left alone to choose ant, m1 or m2, and details > that make sense therein
+1 - that has always been our way and as long as your next point is upheld, it just comes down to what is easiest for the individual RM. > * If you have an expectation to be left alone in things such as the > generation and content of L&N files, that expectation is unrealistic > (in other words, there are some things that interest / should interest > everyone and for better or worse, we're consensus driven) +1 there too. We have some minimum requirements that we argue about from time to time, but it is important for us to maintain consensus on what they are and when we play the RM role, we adhere to them. Call me an optimist, but I think that thanks to the m2 experts among us, we are getting close to having a simple, easy-to-use build environment for m2. We had that for some time with m1 and it helped make cutting commons releases less of a "black art." The /releasing pages document what is still a fairly arcane process using m1 that will become easier with m2. We will get that documented, these long threads will die, and we will get back to talking about (our own ;) code. That's my theory and I'm sticking to it :) Phil > > -Rahul > > > --------------------------------------------------------------------- > 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]
