Quote Originally Posted by Jasdoif View Post
Because I didn't get around to this last thread....

(...)

All that said...Section 4b describing the voting process could benefit from stating some of these things.

  • That "A, B, C" represents a vote for the combination of A & B & C, and only the combination of A & B & C, should be called out. While it certainly makes sense, it isn't intuitive; my assumptions would have been that it represented giving equal weight to A/B/C individually, or that all the subsets (ABC/AB/AC/BC/A/B/C) are equally weighted.
  • Assuming only considering options appearing on a majority of ballots (should any exist) is in fact standard for votes, it should definitely be mentioned as it's not standard for IRV.
  • Mentioning that votes decide which single change is done next certainly wouldn't hurt, to mitigate confusion over votes attempting to reflect all the changes together.
I'd go further than that. IRV doesn't work well for the application at all. To top it off, the fact that the curator finds running votes to be "exhausting" means the situation is biased towards inertia.

Every proposal that gets seconded by two or three or four other participants should simply be the object of a binary vote where the two options would be "implement proposal" and "don't implement proposal". (The proposal can be a mix of things, of course. So if you want A + B + C and only want them if they're together, you suggest voting on it, if four others agree, then a binary vote takes place between "implement A + B + C" and "no changes".)

Elegant, simple, and perfectly functional.

(I'd raise the support for a call for a vote to four instead of two because anything that has any chance of passing should have no problem at all getting the support of four people; raising the bar there has the effect of reducing the likelihood of a proposal that won't pass of reaching the formal voting stage, which would be an useless exercise.)