[EM] Various Criteria and the 3 matrices

[MIKE OSSIPOFF]

[list of apparently correctly-defined criteria]

>If there are other criteria, they should be added to the criteria.html.
>

There are other criteria. Among others, there are a number
of other majority rule criteria that a number of us judge
methods by. They include SFC, GSFC, NDDC (WDSC), SDSC, MDC (SrDSC).
They're all met by any method that meets BC. Other criteria
that are related, though they don't mention majority, are
FBC & SARC.

As for those criteria that you mentioned, that change ballots,
or add or remove voters or candidates, or count the ballots
in separate subsets for separate elections, all the distinctions
between good methods & inadequate methods can be made without
using any of those criteria that change ballots, remove or
add voters or candidates, or count ballot subsets separately.
I don't deny that those criteria that you didn't like don't have
the degree of importance of the majority rule criteria. But
I must say that I question the value of Mutual Majority. It's
about a fortuitous special case. A method that carries out
majority rule, as that criterion describes, in the special
fortuitous situation that MMC describes, may not do well by
majority rule under more general conditions. Also, every example
that shows IRV passing MMC is also an IRV badexample, if you're
one of the voters not in that fortunate majority. Then, it becomes
an example of IRV strategically forcing you to completely abandon
your favorite by voting someone else over him/her.

MMC can be used as an attempt to make it look as if IRV complies
with majority rule. Don't fall for that.

Mike Ossipoff

Mike Ossipoff

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com