Jeremy with a week or so more maven experience under my belt, and having revisited the parent and buildtools pom files I feel I am now sufficiently well informed on the maven technical side to +1 a vote if it were to be reinitiated.
Best Regards, Kelvin. On 08/10/06, Jeremy Boynes <[EMAIL PROTECTED]> wrote:
I was asked off-list why I would abandon the release vote when I had the necessary +1's. The answer is tied to the decision making process at the ASF. One of the benefits the ASF provides to its projects is that it is a corporation, an actual legal entity that can perform legally binding actions. This makes the Foundation legally responsible for those actions rather than the individuals themselves. If someone sues, for example for patent or copyright violation, then they would being suing the Foundation rather than the individual. IANAL so there may be exceptions but in general, this is a good thing for the contributors. For this to work, the Foundation must provide adequate oversight over its actions. It does this through delegation from the Board to the Vice-Presidents that Chair the various Project Management Committees. Executive actions are taken by the Chair based on decisions made by the PMC membership. These decisions are made by Vote of the PMC membership, subject to overrule by the PMC Chair and the Board. Releasing software is one of those legal actions. The Foundation is entering into a contract with the user in the form of the Apache License agreement. It is also entering into agreements with suppliers to that software - for example, it must comply with the redistribution provisions of any included software. The PMC Chair acts to release the software based on a vote of the PMC members with a +1 indicating that they support the release. For the PMC Chair to be able to respect that vote they must have confidence that the member had performed their duty of care - for example, that they had reviewed the software and did not know of any issue that affected the Foundation (such as containing third-party software that was not in compliance with Board policy). An Incubator Podling like us is not a recognized committee of the Foundation and to that extent the votes that we cast are not binding on the Foundation. To release the software that we produce we need action by the Chair of the Incubator PMC (IPMC) who will act based on the votes of IPMC members. The IPMC delegates responsibility for this to our community through an informal Podling PMC (PPMC) that in our case comprises all active committers. It is our responsibility to produce a release and vote as if we were on an actual PMC and were responsible for the decision. The outcome of that vote is evaluated by the IPMC members as part of their duty of care over the release; they also do their own review of the content to ensure that we did not miss anything. The more confidence they have in us having acted appropriately, the easier the IPMC vote. In this case, we had enough votes on the content to indicate that there had been review of the code and that it could be presented to the IPMC for a vote. However, during the course of the vote it became apparent that there was considerable confusion in the community about what we were voting on and why; that there were several people who were unclear on the technical aspects of the content or on the process that we would be following. I do not believe that someone can exercise due care when they are unclear on the matter under review and as such that would be sufficient to cause me to change my vote to a -1; as the vote initiator and Release Manager, it was enough uncertainty for me to consider the result inappropriate to present to the IPMC. Voting based on trust of someone else's review is something I consider OK if clearly stated (assuming enough people actually do a review to provide quorum), voting +1 just to get it out I view as questionable. I think the next step here is to clear up the confusion over the content and process so that all interested community members can vote clearly. If we think we are in that position now we can just vote again. However, based on comments this weekend and remaining abstentions I think there are still some issues to be cleared up first. -- Jeremy --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
