Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Appendix B: Single Transferable Vote #18

Draft
wants to merge 4 commits into
base: main
Choose a base branch
from

Conversation

adambirtles
Copy link
Member

This adds Appendix B, which defines Single Transferable Vote.

The definition is based on my STV counting code that was used to provide the initial count for the previous AGM. The method was agreed upon by the committee prior to the AGM.

There are a few things that I think we should consider before putting this to the Committee for a vote:

  • What happens in case of a failed count (due to an unresolvable tie)? I have added a proposal to the appendix.
  • Should we specify that the vote may be intially counted by computer but must be counted by hand for verification?
  • Should the appendix state what should be included in the results? I have only explicitly stated that the invalid ballot count should be included.

Fixes #12

@markspolakovs markspolakovs modified the milestone: EGM 2021/10/29 Oct 21, 2021
Copy link
Member

@sersorrel sersorrel left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

personal opinions:

  • h
  • if that's a thing that people think is necessary, yeah
  • potentially? maybe worth documenting somewhere, here seems like an ok place, or maybe we just leave it as the implicit "look at last year's agm news article", idk

i do think it would be good to consider how RON is handled, i don't know if inserting a single RON candidate into an election with multiple seats is sufficient if RON is just treated like any other candidate (i.e. candidates who reach quota after RON could still get elected in subsequent rounds)

in any case, probably worth specifying how (or whether!) RON is handled here, imo

appendices/B-single-transferable-vote.md Outdated Show resolved Hide resolved
@adambirtles
Copy link
Member Author

From discussion on Slack last night:

There is a flaw with this method of counting because, as @sersorrel has already brought up, it only allows for voters to choose to reopen nominations (RON) for at most one seat in multi-seat elections. There are a few solutions to this:

  1. Have as many RON pseudo-candidates as there are seats. (Apparently URY's solution)
  2. Stop the counting if RON gets elected, leaving any remaining seats unfilled. (Suggested by @sersorrel)
  3. Don't have a RON option, disallow defaulting, and count spoiled/empty ballots when calculating the quota. "Disallow defaulting" means making it so candidates absolutely have to reach quota to be elected and seats can be left unfilled by the end of the counting.

It was widely agreed that solution 1 would be hacky and confusing to voters.

I don't recall if YUSU require that we provide a RON option. Regardless, we should keep it. Voters expect it from other society and YUSU elections. Also, it is a more intuitive way for voters to express dissatisfaction with candidates (either a subset or as a whole) than solely relying on not ranking candidates and spoiled/empty ballots. As such, solution 2 is probably the best option.

For the most intuitive solution for voters, it could also be useful to incorporate parts of solution 3: count spoiled/empty ballots for the purposes of the quota and disallow defaulting. This would make not ranking candidates basically equivalent to ranking them after RON. Meanwhile it would make spoiling a ballot and casting an empty one equivalent to only voting for RON. However, these additional changes aren't strictly necessary and have some problems of their own, so I don't feel too strongly about them.

All in all, I recommend stopping counting if RON gets elected. Optionally and additionally, we could disallow defaulting and count spoiled/empty ballots in the quota.

@adambirtles
Copy link
Member Author

I'm not likely to be able to get this updated, let alone passed by the committee, by the EGM this evening. I have, however, decided on what changes and clarifications I plan to make to the proposal. I plan to use this proposed system with said changes as Returning Officer this evening. For documentation and transparency here are the changes:

RON and spoiled/empty ballots

RON will get a single entry on the ballot. If, at the end of a round, RON has been elected then the counting will end. Candidates will not be allowed to be elected by default. That is, they must meet the quota to be elected. The droop quota will be counted using all ballots, including spoiled and empty ones.

Election rounds

This mainly clarifies how ties are broken.

If any candidates reach quota, the candidate with the highest score will be elected. If multiple candidates have the highest score and there are no more of them than seats remaining, then they will all be elected. If there are more of them then counting will end.

Unfilled seats

I would just like to clarify that I plan to update the proposal such that it doesn't talk about the counting "failing", only saying in certain circumstances that it will "end" for whatever reason (including all seats being filled).

If there are unfilled seats when counting ends, all the candidates that the counting elected will be considered elected to the position. As per the constitution

Computer counting

I will state explicitly that counting may initially be done by computer but must be done by hand for verification. (Although I don't know if I'll be able to update the program in time to use it this evening. 😬)

Results

Still not entirely certain whether the appendix should state what should be included in the published results. But I don't think it could hurt so I probably will include it if it's not too much hassle to figure out.

@FromAnkyra
Copy link
Contributor

Another thing that needs to be dealt with: what do we do if a candidate drops out mid-race?

@adambirtles
Copy link
Member Author

Another thing that needs to be dealt with: what do we do if a candidate drops out mid-race?

Assuming they drop out before the counting has started, my suggestion is that their first-choice ballots should instead go to their second choice during the set-up (and considered empty if there's no second-choice). After that they can just be treated like any other eliminated candidate during ballot redistribution (i.e. they get skipped over).

If they drop out after the counting has been started but before results have been confirmed/announced, then I think there are two possible solutions:

  1. We can either rerun the counting with them excluded like above
  2. We just go ahead with the results as they come out and, if they win, they can choose to resign immediately

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Appendix proposal/request: STV
4 participants