On 27/11/2009, at 2:27 AM, Todd Thiessen wrote: >> The logic here is flawed because it is from a single >> perspective of an >> individual who finds it burdensome to validate each and every release. > > It isn't just me. Both Paul and Brett expressed similar concerns ;-).
I don't think this was my point. I'm fine with either more or less frequent releases (just not as infrequent as 2.0 -> 3.0-alpha-1 or 3.0-alpha-2 -> 3.0-alpha-3 :) I'm not fine with circumventing review or pushing releases outside the ASF. I'm also not pushing for duration for "testing" purposes. That's part of it, but as Jason said automation can reduce the need over time (though it's never going to be 100% so there's some value in testing). However, it is mostly for an opportunity to review changes. If we do "8 releases a day", then the chances of something undesirable slipping in that others might have noticed increases greatly. We've had things go into actual releases that can do bad things to the remote repo that we may need to live with forever, so it's not unthinkable that such things would be more common under more frequent releases. Sadly, I think the review aspect is frequently lost as people +1 releases they haven't looked at the code changes for, and sometimes haven't even tested. I actually think frequent releases during alphas is less important than frequent releases after final (3.0.1, 3.0.2, etc). Those consuming the alphas are going to be much more bleeding edge and happy with nightlies / source code. We just need regular milestones for those that have come to depend on it like Archiva, or Tycho as Jason mentioned. So, in summary - ain't broke, don't fix it :) - Brett --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org For additional commands, e-mail: dev-h...@maven.apache.org
