RE: Let's found an organization to oppose IRV

[Rob Lanphier]

On Wed, 15 Nov 2000, LAYTON Craig wrote:
> Incedentally, have detailed vote counting rules been developed for any of
> the major Condorcet systems?  I mean, the type of rules that could be
> followed by vote counters doing a manual count, covering all contingencies
> such as dead heats &c?  I would like to see any.

Yes...though nothing spelled out in rigorous legal language.  There's been
enough of a definition that I was able to write software that determines a
winner in any condition.  You might want to play around with it here:

http://www.eskimo.com/~robla/politics/condorcet-front.html

I wrote this with a lot of help from Mike and Steve Eppley, who helped me
work out the corner cases.  The source code is available, so if you read
Perl, you can figure it out.

This implements what we now refer to as Minmax(wv) (I think), as well as
Copeland's method.  It also computes the Smith set, which isn't much good
as a method in isolation, but is a good indicator of if you've found a
really gnarly corner case with Minmax or Copeland's method (i.e. if they
don't match).

Rob Lanphier
[EMAIL PROTECTED]
http://www.eskimo.com/~robla