Re: Proposal to increase the required majority for all RFCs

On Sat, May 17, 2014 at 8:57 PM, Andrea Faulds <ajf@ajf.me> wrote: > Good evening, > > Given the recent controversy on the 64-bit patch vote, and after some discussions on IRC[0] > with some others, I think we should amend the current voting rules[1]. > > Currently only 50%+1, a simple majority, is required to pass ‘non-language changes’. > However, I feel that this is too small of a hurdle to pass. Under this rule, contentious changes can > pass despite a very thin majority. The rationale seems to be that non-language changes have less > broad effect and hence don’t need wide approval. However, changes which don’t affect the > ‘language’ can still have wide-ranging effect and be very problematic. It is also a quite vague > requirement, as it is not always clear what change affects the ‘language’ and what doesn’t. > Hence, I propose that we require a supermajority of 2/3 to pass RFC votes.. This system is > currently used for ‘language’ votes, but I feel it ought to extend to everything. This a much > bigger hurdle to climb over than only a simple majority, but it means that only changes with broad > consensus are likely to pass. It also means that the results of a vote will be less contentious, as > there need to be at least twice as many votes in favour than against for it to pass. Finally, this > change would mean there would be no interpretation issues as to what constitutes a language change, > as all changes must meet the same bar. Language changes was meant for new features in the language exposed to the users, things we will have to maintain for ages (think safe_mode and other new language constructs). Internals on the hand keep changing without impact on userland, hence we won't have to keep using one implementation for years. Cheers, -- Pierre @pierrejoye | http://www.libgd.org