[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Comment on Results for Ca¢s 2016 Primary Election (fwd)
- To: noelle
- Subject: Re: Comment on Results for Ca¢s 2016 Primary Election (fwd)
- From: robert <http://dummy.us.eu.org/robert>
- Date: Sat, 11 Jun 2016 08:08:45 -0700
- Keywords: my-Oakland-voicemail-number
The Debian project uses https://simple.wikipedia.org/wiki/Schulze_method which
is a variant of the Condorcet system.
https://en.wikipedia.org/wiki/Voting_system has a bunch of systems. The
one that Doug describes here is "approval voting". I think he's
referencing https://en.wikipedia.org/wiki/Approval_voting#cite_note-28
which says that, in most cases, it elects Condorcet winners.
> From: "Disqus" <http://www.disqus.net/~notifications>
> Date: Wed, 08 Jun 2016 17:42:34 -0000
>
> Doug Meserve wrote, in response to Noelle:
>
> That might be ideal, except that there are still unresolved issues about the
> mechanics. From what I've read, the popular "instant runoff" method leads
> to chaotic results once any 3rd-party candidate starts getting enough votes
> to matter, including cases where ranking a candidate higher can actually
> hurt that candidate's chances in the final result. The better method, "
> Condorcet," gives a much cleaner picture of voter preferences, but also has
> the possibility (though unlikely) of winding up with "cycles," where
> candidate A wins over B, B wins over C, but then also C wins over A! Such a
> result basically means a tie, which again may be a true representation of
> voters' intent, but then the question becomes how to break such ties. There
> are a variety of methods, and therein lies the hangup that is stalling
> implementation. Also, Condorcet requires some significant computation to
> tally up correctly, meaning that it's very hard to do by hand in any kind of
> recount situation.
>
> The "vote for two" idea is much, much simpler for everyone involved, and yet
> will give just about as good results as Condorcet, from what I understand.
> It also has a big advantage of being a very small tweak of the current
> system.
>
> Link to comment:
> https://disq.us/url?impression=619fdf28-2da0-11e6-9327-002590f38886&experiment=digests&behavior=click&url=https%3A%2F%2Fww2.kqed.org%2Fforum%2F2016%2F06%2F07%2Fresults-for-cas-2016-primary-election%2F%23comment-2719144098%3AyPJNb85dz9_Ac6kMDlnRXIssjiA&type=notification.post.registered&variant=active&event=email
>
> Noelle wrote:
>
> how about ranked choice voting?
>
> -----
> Options: Reply with "Like" to like this comment, or respond in the body to
> post a reply comment.