User talk:LondonStatto/Proposed STV Election Rules

Details of the system

Don't we need a link to more detail on the STV variant adopted? Both for the rules and for members to consider. Johnbod (talk) 12:15, 18 September 2012 (UTC)

I agree. Yaris678 (talk) 12:33, 25 September 2012 (UTC)

I've added a link to the full details.

It would still be useful to have a human-readable summary. For example, are the following all true?

• Each voter gets one vote, which is moved between candidates according to the ranking given to the candidates by the voter.
• Calculation proceeds in a number of rounds.
• Voters provide their ranking once, before calculation stats.
• In the first round, the votes go to candidates that voters have selected as their first preference
• If a candidate receives at least a certain quota of votes in any round, the candidate is deemed elected.
• If a candidate receives more than the quota of votes in any round, the excess is used to distribute a portion of the vote from each voter for the elected person on to their next preference.
• These proportions of votes continue to be distributed as a proportion in subsequent rounds and so may become a smaller and smaller proportion as calculation progresses.
• If no candidate is deemed elected in a round, the candidate with the fewest votes in that round is eliminated from future rounds. In those future rounds, votes go to the next preference in the ranking provided by each voter.

Yaris678 (talk) 14:16, 28 February 2013 (UTC)

Date

Let's start talking about the date now, allowing for the 28 days notice. We must have the resolution complete when the notice goes out. I would suggest we do this in the basement at the office. Early November? Johnbod (talk) 12:21, 18 September 2012 (UTC)

Just one proposal?

Are we just going to vote on one proposal? That seems rather against the spirit of the STV! Ideally we should vote on a small number of different systems, including the present one, followed by a ratification of the most popular one. That's rather hard to do for online voters in advance, although perhaps the ratification of whatever emerges as most popular can be made automatic in the resolution. Johnbod (talk) 15:18, 18 September 2012 (UTC)

As discussed at the water cooler, one approach would be to use Majority Judgment. As Homunq has identified, this could be done with the {strong support, support, neutral, oppose, strong oppose} options that most Wikimedians will be familiar with. Although the Wikipedia article and Homunq talk about it being used for a single winner, it could easily be used for multiple winners by taking the 7 candidates with the highest median rating. An apparent draw back of this system is that it is overly majoritarian. If there are two factions of candidates it is possible for one faction to get all its candidates elected by being only very slightly preferred to the other faction. However, I think such factionalism is unlikely (call me naive if you will). It could be argued that the fact that these candidates have the highest rating by a majority of the voters is a strength. I remember at the AGM Mike Peel explained why the current system was selected. He said it was because it meant that all successful candidates had to be approved by a majority of voters. That could easily be introduced to the MJ system by adding the rule that all winners need to have at least "support" as their median vote. Having said that... if we go with that rule, we may as well restrict the options to {strong support, support, oppose}. Maybe that would be the way forward, it is a kind of hybrid between MJ and the current system. Yaris678 (talk) 13:24, 25 September 2012 (UTC)