Assembly Hall

Intro

Given that the current CSM is reviewing the White Paper that governs its functions, now is an excellent moment to review the CSM election process.

One major idea is that each vote receives a fair share of representation. 44296 votes were cast in candidates that were elected in the last election and 14813 in candidates that were not.

The later did not receive their fair share and even if this doesn't mean that they are not represented now they still account for more than the first and second place together. If some or all these votes were not 'wasted', then the election result could have been somewhat different. Maybe not so much of whom was elected, but who would be the top seven.

Two improvements regarding this were mentioned in the last december's meeting minutes, namely the nomination process and implementing the STV system, but I'd like to propose an alternative that is simpler, accomplishes roughtly the same result and is significantly better on a few aspects:

Two Turns for CSM Election

The idea is to split the election process in two turns. The first one is to prune the candidate list, thus replacing the nomination phase, and to gauge the support of each candidate. The second is to allow the voters to make an informed decision of whom can best represent them that still have a good chance of being elected, thus reducing the percentage of wasted votes.

The whole process would go like this:

1. Candidacy application period opens.
2. Publication of the list of approved candidates; first turn begins.
3. List of candidates who made to the second turn is published, along with the full results; second turn begins.
4. Results are announced.
5. The elected members take office.

The top N candidates are selected for the next turn on step 3. Since there are 14 seats in total, I recomment N = 25, but any number between 21 and 28 chosen a priori could do the trick.

EDIT: as stated below, some people would feel more comfortable if there is an abstain option in each turn.

What I don't like about the changes proposed on December's Minutes

• There is no commitment in the like-based nomination. Two candidates received more 'likes' than votes (112% and 59% more likes than votes), and it is very likely that many players didn't vote for the candidate they supported or, in fact, gave 'likes' for more than one.


• People that want to make an utilitarian vote have no accurate way of knowing which candidate can represent them well that still have a reasonable chance of being elected. Again, the like-based system doesn't accomplish this very well. The 2nd most 'liked' candidate ended up being the 11th most voted one, while the 24th most 'liked' managed to climb to the 14th place.


• The STV implicitly assumes that each voter has perfect information about every candidade. This could not be further from the truth in an environment where 83% of the people don't even vote. Even among the ones that do vote, it is unlikely that most know well more than a handful of the candidates.


• The STV makes the action of casting the vote a lot more complicated. It is up to each voter to decide how much time to spend learning about the candidates, be it 30 seconds or 30 days. Once one is chosen, however, casting the vote should be the most simple action possible, as to not be a factor influencing the result. For someone that spends 30 days choosing a candidade, spending 5 minutes to cast a vote may be only a small annoyance. For someone that spends 30 seconds, 5 minutes to cast a vote is unacceptable and may aswell turn some voters off (and yes, these too are valid votes).

Optional extension: make the second turn optional

The main drawback of splitting the election in two turns is, of course, that people need to vote twice (although that's not much worse than having to nominate your candidate and then vote too later).

One way to deal with this is to make the second turn optional for all the players that voted for any candidate selected for the second turn. In this case, if any of these players decide to vote again then it is the final vote that counts, otherwise the vote from the first turn just carries over.

Optional extension: each candidate chooses a number of substitutes

This is more geared to the second turn and is sort of a reverse STV. It is each candidate, and not each voter, that chooses a number of substitutes to whom his/her votes will pass on in case of not being elected.

The direct advantages are:

• Even less wasted votes.
• More visibility of each candidate aligment, since they will need to announce who they believe are other likeminded candidates.
• Since they are immersed in the process, they are in a better position to evaluate who else to choose.

In its most simple form each candidate needs to choose exactly one othe candidate. Due to pigdeon-hole principle, each of them needs to choose N-14 substitutes to garantee that not a single vote is wasted. An intermediate solution (for instance, allowing each of them to choose three to five substitutes) has an extra benefit of providing support, without enforcing it, for players to form political parties, if so they wish.

The main reason is to reduce the percentage of wasted votes.

The other reason is to eliminate the 'like'-based nomination phase or, more accurately, to promote it to a first-pass election.

I'm not sure the "two turn" thing I like.

If a voters candidate didn't make the list, there is the "I didn't vote for them so if things go to hell in a hand basket its not my fault."

I like my "write in" votes when voting and don't want to think i have to vote for someone i don't necessarily want to vote for.

Your idea actually makes sense and would work far better than anything we have right now. For this reason alone...most people will not like it.

I, however, like it.