I think it is great that there are all of these views on the importance of consensus and how that ties to the nurturing of community and harmony toward a common purpose. I wouldn't even hang around ASF, after having grasped what little of that as I have, if it was not the case.
And it is obvious, from this discussion, that there needs to be room for a PMC to establish where they stand on certain kinds of deliberation and resolution. And that is something the PMC determines by whatever the starting process is. A concern I have is related to podlings, ones that do not have many experienced committers, where customization happens without explanation by mentors and newbies will be launched into the world with different ideas about how things are. For me, I am thinking that simple principles, in particular what -1 means when it is not the specific case of a release or commit, need to have simple practices that everyone can use as a starting point before having to understand the nuances. (And I am not in favor of -1 from anyone counting as a hard-stop veto, which is what I have been seeing in at least one place.) The requirement for mentor concurrence on personnel matters is also an opportunity for coaching if things seem to be going pear-shaped. (I don't know that is a principle for podlings, it is how it worked when AOO was in the incubator. We also did not have any mentors go native.) Another thing has to do with confusion of committer invitation and [P]PMC invitation. I understand that a combined invitation is trumped by the [P]PMC policies and there are *specific* procedures in place for how concurrence of higher powers is obtained before an invitation should be presented to the invitee. And, while different PMCs have different ways of doing all of this, I think it is important to understand that they are different things and local principles will determine how they are amalgamated, discussed, and resolved and whether invariably tied together. I'm not seeking some sort of absolutism. I do think there needs to be a reasonable stable ground for the principles, ones that can be easily followed, and that lean toward the fostering of community. I don't think it is a good idea to be all over the map when a PPMC fires up with initial committers that are not ASF vaccinated and whose further arrivals need not be either. - Dennis -----Original Message----- From: Greg Stein [mailto:gst...@gmail.com] Sent: Tuesday, March 24, 2015 06:47 To: Marvin Humphrey Cc: dev@community.apache.org Subject: Re: Veto! Veto? [ ... ] I dislike all voting, yes. Consensus through discussion is definitely a better approach. Concretely: I don't think there is any specific recommendation for how a PMC/community decides upon new committers. I've seen many mechanisms. In fact, within Apache Subversion, a committer can be added by any *singular* PMC member, no vote required (but their resulting commit rights are limited). For PMC Members, Roy has stated [on general@incubator, on 1/31/2012] that: "Well, it boils down to the fact that making someone a PMC member gives them veto power over the changes you make. The only way that works socially is if everyone currently on the PMC agrees that person is a peer." >... Cheers, -g
