[EM] Herve Moulin's proof not really a proof

[Nicholas Buckner]

> Date: Mon, 11 Jun 2012 14:15:05 +0200
> From: "Markus Schulze" <Markus.Schulze at alumni.TU-Berlin.DE>
> To: election-methods at lists.electorama.com
> Subject: Re: [EM] Herve Moulin's proof not really a proof
> Message-ID:
> 	<89c17a23209e71506bfe9d539ad08852.squirrel at mailbox.alumni.tu-berlin.de>
> 	
> Content-Type: text/plain;charset=iso-8859-1
>
> Dear Nicholas,
>
> who is elected by your method in these 7 situations?:
>
> http://lists.electorama.com/pipermail/election-methods-electorama.com/2007-February/019497.html
>
> Markus Schulze
>
>
>
> ------------------------------
>
> Message: 5
> Date: Mon, 11 Jun 2012 16:02:13 +0200
> From: Kristofer Munsterhjelm <km_elmet at lavabit.com>
> To: Markus Schulze <Markus.Schulze at alumni.TU-Berlin.DE>
> Cc: election-methods at lists.electorama.com
> Subject: Re: [EM] Herve Moulin's proof not really a proof
> Message-ID: <4FD5FA65.2020001 at lavabit.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> On 06/11/2012 02:15 PM, Markus Schulze wrote:
>> Dear Nicholas,
>>
>> who is elected by your method in these 7 situations?:
>>
>> http://lists.electorama.com/pipermail/election-methods-electorama.com/2007-February/019497.html
>
> It appears that his method elects D. Adding 4 DABC voters then makes A win.
>
> I'm using the following CSV files:
>
> 3,1,4,2,3
> 3,1,4,3,2
> 4,2,3,1,4
> 5,4,2,3,1
>
> and
>
> 3,1,4,2,3
> 3,1,4,3,2
> 4,2,3,1,4
> 5,4,2,3,1
> 4,4,1,2,3
>

According to an older Droop quota I do my best to try to discredit
using inside just my program, you would get Candidate D first.
However, I did update my quota to votes/2+.0001 as a placeholder (as
the votes increase +.000001 would be needed). With the new quota,
Candidate A wins:
First round:
A: 2.375025
B: 1.700000
C: 2.437513
D: 1.416683
C is removed.
Second round:
A: 1.375025
B: 1.700020
D: 1.416683
B is removed.
Third round:
A: .750010
D: 1.250010

Final order: A>D>B>C

Adding those four votes doesn't change the order, as during the final
round A>D has 10 votes and D > A has only 9 votes.

Sorry that the source code for the new quota has been up for only a
short while. I still believe it complies with both.

Sorry for any confusion,
Nicholas