3 ms·
I'll start off with my prototype: https://elegant-shaw-2cb49a.netlify.app/votevote https://elegant-shaw-2cb49a.netlify.app/votevote Anyone into electoral meth
by culi 5y ago
I'll start off with my prototype:
https://elegant-shaw-2cb49a.netlify.app/votevote https://elegant-shaw-2cb49a.netlify.app/votevote
Anyone into electoral methods or game theory? If you've ever watched CGP Grey's videos on electoral methods or seen Nicky Case's *To Build a Better Ballot*,[1] it's very much in the spirit of that. It's mostly a toy that will run a single election in a large number of different electoral methods.
Currently I've implemented ~26 different electoral methods which is pretty neat. However, I've shied away from the methods that voting theory nerds love the most: Condorcet methods. I've implemented Copeland (although there's a mistake I haven't fixed yet) as well as Kemeny Young. However, Kemeny Young is implemented in the most brute force way possible and is O(n!) right now.
Mostly I think I'm just looking for people with a math/game-theory background that would be interested in helping out, but I'm open to anyone interested. I know there's a lot of optimizations I could make to the Kemeny Young algorithm, but I've been hesitant to do so because I can't find the right theorems to make sure certain assumptions are true. For example, I'm quite sure that a subset of weak Condorcet winners would be at the front of the winning path, but I haven't found a textbook that's stated this explicitly and it's been a long time since college so I'm kinda iffy on proving it to myself. I also don't think having to learn the proof everytime just to make sure my assumptions are mathematically sound is the best way to develop such a tool. Thanks
It's all open-source and won't have ads and will hopefully just be a static site so it's really just a hobby project.
[0] https://www.youtube.com/watch?v=s7tWHJfhiyo https://www.youtube.com/watch?v=s7tWHJfhiyo
[1] https://ncase.me/ballot/ https://ncase.me/ballot/
- iamwil 5y agoThis is pretty interesting. Have you connected with people doing token economics design or DAO governance? That would be pretty applicable. Where did you source all the different electoral methods?
- culi 5y agoThanks! The only people I've really connected with are other voting theory nerds. There's a voting theory mailing list where a lot of academics hang out. Some of them are leaders in the field like Forest Simmons and some others I haven't interacted with as much. Maybe once I have more to show I could reach out to wider audiences The electoral methods are sourced from various places. When I get the new site up and running I want to have a spec source for each of the methods' implementations since some of the names can be confusing. A lot of the methods were found on the electowiki[0] which is run by the same people who do the mailing list. Many others I just found from reading articles, publications, textbooks,[1] and even wikipedia. I've kept a list of ones I wanna implement so I've just been slowly working through them. But there's a strong bias towards simple to implement ones so far. The much-loved (by voting theorists) ones won't be implemented for a while until I figure out how to do things more efficiently [0] https://electowiki.org/ https://electowiki.org/ [1] https://www.maa.org/press/maa-reviews/the-mathematics-of-elections-and-voting https://www.maa.org/press/maa-reviews/the-mathematics-of-ele...
- iamwil 5y agoWhat are the 26 different electoral methods? I only see three simulated.
- culi 5y agoI haven't yet added visualizations for all of them. I realized a lot of optimizations that could be made while working on this so I decided to hold off on that until I rebuild it for the new site (gonna be at votevote.page) But I do at least show the resulting winners of all 26 methods underneath the input section.