How to vote for the Wikimedia Foundation board - be negative!

The Wikimedia Foundation is - said roughly - the organisation running the infrastructure behind Wikipedia. The Foundation is governed by a Board of Trustees comprised of volunteers. Out of the 10 board members 3 are elected by the community of the Wikimedia projects. This election is now.

There is an election page and several posts about who to vote. I will focus on "How to vote" or: how to maximize impact with your votes..

In general the system used is a form of approval voting. As such it is similar used to the voting system used by Wikimedia Deutschland for its board and as already written in this blog (and agreed upon by Wikipedia): there are two ways to maximize impact in such as system:

When you really want somebody to be on the board only vote for this person (or these people) = Bullet voting

When you really want somebody not to be on the board vote for everybody but this person. = Compromise voting.

But the WMF system does not offer two choices (yes or nay) but three (yes, neutral, nay) and even more interesting. The different votes are weighted differently. In the most likely outcomes the nay votes will have way more impact than the yes votes. Neutral votes are not counted in any way:

How are the winners determined, and what is the process once voting is completed?

Voters submit votes using a Support/Neutral/Oppose system. The votes will be tallied and the candidates will be ranked by percentage of support, defined as the number of votes cast in support of the candidate divided by the total number of votes cast for the candidate ("neutral" preferences are not counted, so this is the sum of support and oppose votes) - Support/Support+Oppose. The candidates with the highest percentage of support will be recommended to the Board of Trustees for appointment, which occurs once additional verification of requirements is completed.

Why nay votes count

To give you some examples with small numbers:

Person A who receives 9 support votes and one oppose vote has an approval rate of 90%.

Person B who receives 17 support votes and two oppose votes has an approval rate of 89%.

In this case person A would be the winner although in total numbers he has only half the support of person B. In this scenario the candidate needs 9 support votes to make up for one oppose vote.

This of course works for bigger numbers as well:

A  90- 10 gives an approval rate of 90%
B 900-101 gives an approval rate of 89,91%
A wins

A. 900 - 100 gives an approval rate of 90%
B: 908 - 101 gives an approval rate of 89,99%
A wins

The higher the approval rate for the top candidates is in general, the higher is the impact of a negative vote.

With an approval rate of 90% for the top canditaes voting nay has 9 times more impact than voting yes. Around 80% approval rate the impact of a nay is 4 times as big, around 70% approval rate the factor is below 3 and around 50% the factor is just one. Below 50%, the support votes become more influential than nays.

Given normal circumstances in such an election the approval rates for the top three candidates should be over 70 or 80%; so a negative given has three to four times than impact than a positive vote given.

For you as tatical voter this means: don't waste nay-votes. Don't vote neutral. 

What does that mean?

First, let's hope not everybody votes strategically - the likely outcome would be that all of the candidates will receive an approval rate of less than 50% which would really deligitimize the whole system. But probably this won't happen.

What will likely happen is another scenario: not the people who have the most support in the movement will win, but those who have the least opposition. This of course favours a different kind of candidates: people more keen on compromise, without strong opinions and more middle-of-the-road-types.

On the good side: the probably will make the board in general more efficient. Middle-of-the-road candidates need less time and attention of the other board members. For an international board comprised of 10 volunteers this is important. Coordinating 10 people in this setting is even strenous enough when all of the are great teamplayers.

On the good side: if gives a fighting chance to people from smaller projects. They may not be as well known internationally as those from the english Wikipedia but in this system it's not really the total number of people supporting one candidate that counts. Coming from a smaller project and having a big majority there supporting you could be more helpful.

But, seen from a community standpoint, it favours candidates who are less likely to stand for their points, who are less assertive and are less ready that may hurt or are not popular. Or, said differently: it favours candidates who just may get lost inside the board and never be seen again after their election.