Gerrit Holl writes:
If voting is limited to a select group (which could be as small as Python core developers, or as large as anyone who has ever had a pull request merged into cpython, or something in-between), then a vote could be a way to measure opinions after a lengthy discussion fails to reach a consensus.
I'm not sure what the benefit of "measuring opinions" is supposed to be, when those opinions don't bring real resources with them, and few of them are informed beyond "sounds cool" and "YAGNI". If a discussion fails to reach consensus, human brains do OK at holding a fairly detailed summary of it, including who held what opinion when discussion ended -- far more informative than the result of a vote. What a vote can do for you is make an up or down decision, or choose among alternatives. But from the project's point of view, these decisions are rarely pressing (except maybe security fixes, and those are not going to be discussed publicly, let alone put to a general vote!) If an issue is still controversial after a long discussion, it's usually because there are competing interests in play, and somebody has to lose something they want. In those cases, it's almost always best to table it, and see if any technical progress is made on reconciling differences about the issue over the next release cycle. Sure, tabling issues frustrates non-committer proponents (who are also usually the proponents of voting schemes, what a coincidence!), but that's normally better than frustrating committers who are against it.