On Sun, Nov 16, 2014 at 11:44 PM, Pierre-Luc Dion <pdion...@apache.org> wrote: > I agree that more tests are welcome, we have to try then :) > >> Afaik non-pmc's aren't binding and any votes would merely be an > indication, >> or do i misunderstand? > > All votes are important [1] and count as indicator, who ever vote mean > something, It also show that the community members did some tests or > review of some kind. And whoever doing a -1 with valid justification will > be listen. It would be much more interesting to see an RC with 15 non > binding votes than the 3 minimum binding. > > [1] http://www.apache.org/foundation/voting.html >
Just my 2 cents: Some Apache communities include users@ to get user feedback, and other's do not. IMO, what would be better is for us to engage the user community to help test the release prior to adding the voting overhead. The act of voting should (hopefully) not be that big of a deal. I'd much rather find ways to recruit users to become testers earlier on in the cycle. If you look at our last several releases, QA / quality has been lacking during the release process. It's often been active users that start testing only when they see the vote thread. The question I'd pose is: how do we figure out how to pull these individuals into the cycle earlier. -chip
