On 10/3/06, Roy T. Fielding wrote: > On Oct 3, 2006, at 1:55 PM, robert burrell donkin wrote: > >> That's why we created the PPMC == the entire set of committers of the > >> podling and the Mentors. > > > > this is not policy ATM > > Yes it is -- it was formally voted on during the Geronimo incubation. it isn't in the process documentation. this is the only description we have of current policy. a patch (the process document is RTC) or a message ID would be useful. > >> They do have binding votes on everything > >> *except* releases because we delegated that to them, right? > > > > i don't understand how this can work with the current structure. AFAIK > > PPMCs have no organisational standing (they are no official > > committees) and are not recognized by the board. AIUI they cannot take > > decisions binding on apache. their main role ATM seems to be as a > > training exercise. > > How can they avoid making decisions that are binding on Apache? > Importing code into subversion is a decision that is binding on Apache. importing code is an action. the meaning i intended for 'binding' is an action which is taken on behalf of apache. AIUI anyone with commit privilages could upload tar ball containing source to an apache server. if they upload the tar ball after due process then they may be acting on behalf of apache. if they just upload the tar ball then they are acting as an individual and must take personal responsility for their action. > It just isn't a very important decision. The PPMCs are making binding > decisions every day -- the only condition we placed on them was that > at least three +1 votes had to come from the PMC (whether or not those > people happened to be Mentors) for releases and adding new committers. and for adding new PPMCers, i presume > That does not mean the other votes are ignored. AIUI those actions you mention are the ones which in any project require a majority of binding votes. these are the votes which the committee needs to decide on so that there is organisational (rather than personal) responsibility. there are lots of other votes but IMHO they are primarily social rather than official (though at least of equal importance). > The only reason PMCs have organizational standing is because they > are a group of named individuals on a committee that has been > assigned a task by the board in accordance with the bylaws. A PPMC > becomes an official committee as soon as the Incubator creates it, > since its creation is well within the scope given by the board to > the incubator project and the individuals are named in the status file. but i don't see any official creation process. AIUI the board creates new committees through a top level project resolution voted upon by the board. ATM there is no explicit vote from the incubator PMC, there is explicit policy in the process document and we do not insist that the proposal contains a section about the PPMC. there is no papertrail. > The only question is what authority is granted to the PPMC by the > Incubator, and every podling since Geronimo has acted according to > the policy that all decisions are made by the PPMC with a minimal > quorum of three PMC +1 votes. i don't think that this is in the policy document. a patch or a message ID would be very useful. > > bootstrapping is simply a description of the only process available > > ATM. the mentors (as incubator pmc members) are the only ones on the > > project who have the binding votes required to take decisions (such as > > appointed PPMC members). > > That just isn't true. Somebody took a thread out of context and > applied it where it doesn't make any sense. The restriction that > only a PMC can release software has very little to do with all > of the other decisions a project might make. if the PPMCs were official committees, i would agree. maybe they are but i just don't see the paperwork or the process documentation to support this claim. it would be cleaner if they were official committees but IMO we need a little more ceremony before it is clear that they are. - robert --------------------------------------------------------------------- To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org For additional commands, e-mail: general-help@incubator.apache.org