On Tue, Jun 5, 2012 at 11:01 AM, Marvin Humphrey <mar...@rectangular.com> wrote: > On Tue, Jun 5, 2012 at 9:53 AM, Patrick Hunt <ph...@apache.org> wrote: >> Isn't this why we vote. To come to a decision when consensus can't be >> reached and allow people to move on. > > When diversity concerns were raised in the ManifoldCF podling by Jukka, > graduation was delayed to address them. > > When diversity concerns were raised in the Lucy podling by Torsten Curdt, > graduation was delayed to address them. > > When diversity concerns were raised with regards to Flume, a VOTE was called. >
I think that's an oversimplification that ignores the work and good faith the Flume community has done to address the concerns raised. Extensive discussion ensued and afaict Ralph's concerns were addressed. He raised issues and the community worked to resolve them. (ie promoting committers to pmc as part of graduation). > Contentious VOTEs force people to take sides and create winners and losers > where none existed before. It is often worth going the extra mile to avoid > imposing a "tyranny of the majority" and fostering alienation. > > In my opinion, there would be a benefit to the Flume community for staying in > the Incubator a while longer and wrestling with how to increase diversity. > Have you folks ever had a non-Cloudera Release Manager? Was there ever a > possibility of a non-Cloudera Flume VP? How about waiting until someone other > than a Cloudera person emerges who is willing to lead a graduation push? I > think the community would develop healthy habits by prioritizing such concerns > for a little while. > > On the other hand, there are benefits to graduation in terms of enlarging the > pool of potential contributors by those who are willing to get involved with > graduated projects but not podlings. Ralph has shown an open mind and has > been weighing the plusses and minuses. We all want what is best for Flume. > All great ideas. But this highlights my concern and the problem I have with many of these discussions. I've been on the other side of the fence on this stuff. You're trying to make forward progress. How do you separate the "must do's" from the "good ideas". From what I could see the Flume community worked with the IPMC to review the status and address any concerns. At which point the vote was moved forward. If you don't agree then -1. That's why we have voting. That's why it's majority vs veto'able. If the vote doesn't pass is will list explicit grounds for a -1, the community can go back, address those, and start a new vote when the issues are addressed. There should be no shame in getting a -1, voting (after healthy discussion) allows the community to speak clearly and definitively and it also respects the community requesting the vote. Patrick --------------------------------------------------------------------- To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org For additional commands, e-mail: general-h...@incubator.apache.org
