>Individually, or acting in concern, the small body of ARC members (as >few as 4, I believe, makes a quorum) can choose to derail or reject a >project (though in principle rejections rarely happen, and are usually >accompanied by quite a lot of justification in e-mail and opinions as to >why a project is rejected. From a pure numbers standpoint the project >team may have a "majority" in that it is made of 10-20 people or more.
Please don't tar "derail" and "reject" with the same brush. Derailing can be done by an individual ARC member and only means that the case is non-obvious or non-controversial or even that someone wants to make sure an "opinion" is written to summarize an important part of the ARC discussion. Derailing is just switching the points (or pointing the switches) to the longer variant of the development process. >Perhaps you feel that that there should be an appeals process whereby a >project could appeal a decision to the larger engineering membership at >large. I'm not sure how such an appeals would work. But you could >draft something up as a strawman if that is what you want,. and present >it to OGB or one of the impacted communities for a formal vote. Right, just as there is one inside Sun where "the business" can override decisions. Clearly there can be different such bodies for OpenSolaris and a particular distribution (a distribution can always "fork") Casper
