[Foundation-l] The average voter and voting systems (was Re: Notice of the results of the WMF Board of Trustees election)

[Ray Saintonge]

Jussi-Ville Heiskanen wrote:

>On 7/14/07, Alison Wheeler <wikimedia at alisonwheeler.com> wrote:
>  
>
>>Re:
>>On Fri, July 13, 2007 08:27, Peter Halasz wrote:
>>    
>>
>>>I agree. STV is the best system.
>>>      
>>>
>>>>On 7/13/07, Jussi-Ville Heiskanen <cimonavaro at gmail.com> wrote:
>>>>        
>>>>
>>>>>While I think the Single Transferable Vote is the best voting system,
>>>>>          
>>>>>
>>STV (in all its various actual forms*) is great at producing a 'best
>>supported' result where it seeks to select a single winner from the range
>>of candidates, however it is quite poor at selecting multiple winners from
>>the same range. This is - primarily - because the transfers take no
>>account of removed candidates such that someone who receives few first
>>preferences might get removed and reallocated early on in the count yet
>>over all the voters they actually have good support where the 'winner' had
>>solely a sufficient number of first preferences but no wider base of
>>support.
>>    
>>
>This is inaccurate. The way to get past this is to first find out who is
>first.
>Then eliminate her votes entirely from the count, and redistribute them,
>iterating the same process as before, but her removed from the figuration
>entirely. Then who comes first in this iteration is considered to have won
>second place. Then he in turn is removed from the process, redistributing
>his votes etc. etc.
>
Possibly.  The one thing one had better be sure of before implementing 
an STV system is that the bugs have been worked out before it's used in 
an important election.  I have seen people attempt to use it without 
having thought it through, and it can easily become a total fuck-up.

Ec