This sounds like the votes of more than half of those wo participate are needed for a simple majority in a vote (as opposed to more than half of the entire CTC). Is that correct?

@addaleax Yes, that is correct. Or at least, that is my intention. I guess we'll see if other folks agree with that or not...

So:

The vote-to-schedule-a-vote in the previous paragraph requires more than half of the CTC. (I actually think that should be changed to half of the attendees although there should need to be a quorum, but I don't feel strongly enough about it to try to bundle it into this PR.)

Once there's an actual vote on a decision on a contentious issue, that vote is decided by simple majority (or even tiny plurality—see below!) of the participants as long as there is a quorum participating.

So, for example, there are currently 18 people on the CTC. Let's say there was a contentious issue around whether we should deprecate a particular API or not. And a vote is scheduled to decide. Let's say 8 people don't show up to the meeting and do not register a vote in the GitHub issue for the meeting agenda. Let's say the remaining 10 people all vote either in the issue tracker or else show up at the meeting to register a vote. Hooray, that's a qurorum, the vote counts! Let's say 7 of those 10 people say, "I looked at the issue and I just don't have a strong opinion either way. I abstain. I want a decision to be made, but I don't care what that decision is." That leaves only 3 people casting non-abstention votes. Let's say 2 of them vote for deprecation and 1 votes against. Then it's time to deprecate that API.

Of course, that scenario would be hopefully exceedingly rare. We don't actually have too many votes in my experience (and when we do, they almost always end up being very lopsided). But the point of the quorum rule isn't to make certain that a large majority get their way. It's to simply guarantee sufficient support that the decision has legitimacy without the actual decision-making process being a source of dramatic delays or inaction. It's an absolute floor for what a decision might look like in the absolute worst of times.

In reality, I think we tend to work pretty hard to get at least 10 votes for anything, so the scenario I describe above is highly unlikely to ever actually occur.

This sounds like the votes of more than half of those wo participate are needed for a simple majority in a vote (as opposed to more than half of the entire CTC). Is that correct?

This is incorrect.

It needs to be half of the entire CTC.

See the above:

vote must be approved by a simple majority of the CTC

It does not say "of the CTC present", it says "of the CTC".

Specifically, it says that it's a simple majority of the ctc members participating so long as more than half of the ctc members participate (even if only as abstentions)

@Fishrock wrote:

This is incorrect.

It needs to be half of the entire CTC.

The "vote must be approved by a simple majority of the CTC" text in the current document refers specifically to calls to vote (that is, a motion to close) on a contentious subject, not the votes on the contentious subjects themselves. (Or at least that's how it appears to me.)

And there's a good reason for that. If those were the rules, then:

• It only takes a small number of inactive members or abstentions to make it impossible to take action on a divisive issue.
• It puts pressure on people who maybe should abstain to actually cast an ill-informed vote instead. If we're making a decision that only half the body is adequately informed and has adequate expertise to address, then just that half should have a say. People who know what they don't know should be able to abstain without making gridlock more likely. I don't think it's wise or realistic to expect people on the CTC (or anywhere in the project) to understand everything about the project. Abstaining is OK as long as you're not making it a way of life on every topic.
• This makes proposals easy to block (basically, anyone who abstains or doesn't participate is essentially a "no" vote). In itself, that may not be a bad thing, but this encourages gaming the system. If you want your way, propose the opposite of what you want as the text to be voted on and you've got a head start because anyone not voting is effectively voting for your side. This is not a problem on a motion to close discussion because you can't have a motion to not close. There's a motion to close and a motion to table and that's it. But it is conceivably a problem when it comes to actual voting on the contentious issue at hand.

To my reading, at least, the current doc does not address how we handle votes on contentious issues—only how we handle the votes on motions to close and motions to table. I'm not sure anyone really has been working with a static and fully-formed idea of how these votes work. The new text is an attempt to document what I believe is a workable framework. We haven't needed it, really, so far. As I've noted, votes tend to be lopsided. But we might need it one day and it will be best to have an explicit and unambiguous process. (This text all really only becomes relevant when the Consensus-Seeking has failed.)

Also, I might as well pre-empt what might seem like an obvious objection: "But if it takes more than half of the total CTC to pass a motion to vote, then wouldn't all the problems described apply to that vote?" My answer would be: A lot of them, yes, but there are some very important ones that wouldn't. Specifically, people who will abstain from voting on the actual contentious issue are very likely to vote for closing discussion and moving to a vote. So there's less danger for gridlock there.

I see several problems with this approach, please see #7813 (comment).

In short: I'm -1 for codifying quorum (in such or similar way) — it just introduces more issues.