User:Omegatron/Consensus-building vote

On Wikipedia, we try to settle all disagreements with consensus decision-making. Ideally, this would mean that we have calm, civil, open-minded discussions about things, make an honest attempt to understand each others' viewpoints, and are swayed by each others' arguments until we come to a unanimous agreement on the best way to proceed.

In reality, the best way is often not clear, there are too many people involved to have a meaningful discussion, or people are just stubborn. In situations like these, it's tempting to call a vote and just let majority rule. But Wikipedia is not a democracy, and votes like this are considered evil for a variety of reasons. (See Polling is not a substitute for discussion.)

A new method
When editors resort to polls or votes, the processes they use and the outcomes they get vary wildly from one instance to the next. Sometimes a poll is used to gauge popular opinion, and then a discussion tries to attain consensus from this. Sometimes a vote with two options is created, and the side with more votes wins. Sometimes a poll with multiple options is created and a strong majority for one option is interpreted as "consensus".

This page is an attempt to describe a new method that might solve some of the problems with the ones that, despite being recommended against, are in common use.

This is not the same as a majority-rule vote, and is not the same as an opinion-measuring poll. This is actually a consensus-forming process in disguise. (Creating a new name for this process that doesn't involve connotations of democracy, voting, and majority rule would probably help it to run more efficiently and be more acceptable to editors. Referendum?  Ballot?  Election?  Judgment?  Determination?  Votescussion?  Pollscussion?)

Process

 * 1) Look at the discussion so far, and try to find at least three(?) different proposals that have been made so far.  List each in a neutral fashion under separate subsections.
 * 2) Editors then sign their names under any that they would be happy with.
 * 3) * "Support" votes are forbidden, though. Your "vote" is a statement of your rationale for why you support that particular proposal.  (Of course it's ok to borrow someone else's rationale with "as per User:X".)
 * 4) * Rationales should not be part of the proposal itself; every editor has their own rationales.
 * 5) Anyone can come up with a new proposal and add it under a new section, and votes can be retracted by striking them out if a better proposal comes along.
 * 6) Discussion about editors' rationales or new proposals should be confined to a separate Discussion section, so the "vote" can act as a short summary of everyone's viewpoints.  If viewpoints change (which they should, ideally), editors can cross out their rationales or update them.

Ideally, this will lead to a new proposal that everyone would be happy with, they all vote for it, and the matter is settled. In reality, a compromise version will probably be reached that a large majority will be happy with.

It's like discussion, but in an orderly fashion; editors' voices don't get buried in the mud, and the same arguments aren't repeated over and over again for weeks or months.

Examples

 * Commons:Commons:Images on normal pages or categories:Vote
 * Talk:Peta-
 * First requested move provided only two options, support or oppose; a false dichotomy. This resulted in mostly "oppose"s, thought for varying reasons.  When an alternative option was suggested, it was supported by everyone.  This should have taken place in one single vote/discussion instead of two separate proposals.  Then again, this could be seen as a success for the method of using consecutive support/oppose votes.
 * Talk:Floor and ceiling functions
 * First rename proposal provided only two options, support or oppose. When reformatted in this manner, the consensus was more clear.
 * Village pump %28proposals%29
 * Tried to use this method to decide a user interface change, but no one could figure out how it was supposed to be used.
 * Several people voted for one option, and said in their comments that they supported other options, but didn't actually sign under them.
 * Several people voted against a change, claiming the change was too long, ignoring several shorter options.
 * Several people made alternative proposals in their vote, but didn't actually add them as options.