Thursday, February 14, 2013

The CSM Voting Algorithm and Reform

Until CSM8, the voting algorithm has always been first-past-the-post. It's not a great system. It's not terrible. But it does ensure that organized groups can easily trounce unorganized groups in getting their candidates through. Which is something CCP and the CSM want to move away from. They want the less organized to have some hope of getting through their candidates. Which is fine. As long as a proper voting system is to be implemented, one that doesn't disenfranchise votes of the organized. (Players organizing should always be encouraged.)

There's been some call for a single transferable vote system. Twice, though, CCP Veritas has mentioned (enthusiastically) the Schulze method (or perhaps the Schulze STV, where multiple winners are determined, as in CSM elections.)

Hans seems suddenly concerned and is asking the community to post in the following thread, their thoughts on a new voting system. I'm assuming because CCP is going ahead with voting system reform, without CSM or player input.

I can't say I'm particularly surprised. I would be very surprised if it wasn't the Schulze system being implemented. Veritas seemed super keen on Schulze; it's a non-trivial algorithm, and Veritas is attracted to non-trivial problems.

I don't think we're going to get a lot of say in the matter. As fussed as I probably should be, I'm not feeling all that fussed about the prospect of a new voting system. I just hope that CCP is not going to assume this is their panacea, and that increased voter awareness and education is no longer needed.

***

What I would like to see, is all the voting results released after the election. CCP releases each vote, not who made the vote. It allows independent verification of the results. It ensures the algorithm was coded properly, and that the system is implement properly. Voting systems require oversight. This is certainly no skin off CCP's back and has no privacy concerns. Each vote is anonymous. And releasing the data is not much more than a dump of the voting tables, a few hours work at most to retrieve and release the data. One more piece of transparency.

(Hell, you can even get people playing with the results using different voting systems, to see how different systems capture different results.)
Posted by at 9:41 PM
Labels: ,

7 comments:

  1. This topic needs to be settled. It seems to occupy half the energies of the CSM at times and push more important issues off the agenda.

    Seriously it's a system for selecting a bunch of nerds to help brainstorm. Any 14 nerds will do, they're a fungible species.

    Move on, fix sov, fix tech.

    ReplyDelete

  2. Schulze is actually pretty good.

    I approve of a change. It's certainly a huge improvement over the current system (which by voting system standards, is TERRIBLE).

    ReplyDelete

  3. CCP releases each vote, not who made the vote. It allows independent verification of the results.

    No it doesn't. The most that will show is what data is purported to create the result, and allow independent confirmation that the released data agrees with the result. A big list of votes can easily be forged, and unless a significant proportion of people admit to how they voted you can't show that the list wasn't forged.

    Of course, releasing votes against voters can solve that issue, but introduces privacy concerns, which is, I'm sure, what you were looking to avoid. Independent vote confirmation is a complex subject, but one that, I believe, has solutions. Merely publishing a big list of supposed votes is not one of those solutions.

    ReplyDelete
    Replies

    1. Without a voter verified paper trail any digital method will be flawed from a tin-foil hat point of view. We had quite the fuss about it in the netherlands a few years back. (which is why we still vote with a piece of paper and pencil.)

      Not to say it isn't technically possible, but the effort required to make it a sound system is probably not going to stack up against the importance.

      T20 references aside I think CCP doesn't really care who gets voted in as long as they are serious.

      Having said that, getting the actual vote table would be interesting.

      Delete

    2. But, some groups, such as Goonswarm, perform some detailed exit polling of their members. If there were statistical anomalies between how their members said they voted, and what the voting data shows ... well ...

      The data can be faked ... but it has to line up with the statistical expectations of exit polling and the like.

      Delete

  4. WTF? I thought we were playing an internet spaceship game.

    ReplyDelete

  5. One interesting emergant behaviour in Australia, which uses a STV type system is that each party will release a suggested ordering of all candidates (as a handout, outside polling stations), calculated by them to best 'play' the system, and get them what they want.

    If CCP were to release voting information, I'd be interested to look at the numbers of identical choices in it, seeking similar behaviour.

    ReplyDelete

Subscribe to: Post Comments (Atom)
EVE Online and the EVE logo are the registered trademarks of CCP hf. All rights are reserved worldwide. All other trademarks are the property of their respective owners. EVE Online, the EVE logo, EVE and all associated logos and designs are the intellectual property of CCP hf. All artwork, screenshots, characters, vehicles, storylines, world facts or other recognizable features of the intellectual property relating to these trademarks are likewise the intellectual property of CCP hf. CCP hf. has granted permission to Poetic Discourse to use EVE Online and all associated logos and designs for promotional and information purposes on its website but does not endorse, and is not in any way affiliated with, Poetic Discourse. CCP is in no way responsible for the content on or functioning of this website, nor can it be liable for any damage arising from the use of this website.