[EM] Fw: Invitation to join politicians-and-polytopes

[Norman Petry]

Mr. Carey,

I consider the creation of a second voting methods discussion list to be
pointless.

Rob Lanphier has always done an excellent job of maintaining the current
list.  It is working fine, and I don't see any obstacle to you posting
messages or responding to items of interest here, and ignoring those posts
which are irrelevant to your goals (that's what I do).

I am not surprised that you have been unable to attract participants to your
new list, since I think most of the members of EM would share my sentiments.
However, I am surprised that you are now actively trying to "poach" EM
subscribers by directly soliciting their involvement in your new list.
Going beyond your *TWO* earlier announcements here on EM (which all
subscribers will have received, and could have freely acted upon) is in
extremely bad taste.

With regard to your specific goal of attempting to patch single-winner STV
(AV/IRO), so that it becomes a reasonable voting method, I have no desire to
assist with the project.  Although multi-winner STV is an excellent system
of proportional representation, it is highly defective when used as a
single-winner method.  I think that attempting to fix the severe problems
inherent in AV is like trying to build a house on sand, so I have no desire
to pursue that particular dead-end with you.  This is particularly true if
the fixes you are attempting can only be described in mathematical terms.
Any method which cannot be described in a natural language to an informed
layman is also a dead-end.  Voting methods ultimately must be considered and
adopted by a polity which understands and therefore, trusts the results that
the method yields.  Voting is not a black art (or an exercise in
geometry) -- it is simply a tool that democracies can use for making
decisions.  Mystifying the decisionmaking process by contriving a voting
method which almost no-one can understand is fundamentally undemocratic.
Only mathematical societies would be willing to adopt a method whose
rationale can only be described in mathematical terms, and this is too small
a constituency for me to care what methods of decision-making they might be
willing to adopt.

Fortunately, it isn't necessary to build on STV to create good single-winner
methods.  There is a large class of pairwise methods, *any* of which is
superior to AV in terms of results, and many of these methods are very
simple to describe in ordinary, natural language, and therefore have some
hope of adoption.  One of the things I've been working on is a simulator to
empirically test a variety of proposed methods (I can send you a copy of
this if you like).  These simulations show that pairwise methods
substantially outperform all other single-winner methods under most
reasonable assumptions.  Furthermore, the better single-winner methods
satisfy a wide range of useful criteria, which provide important guarantees
to voters who might be willing to consider adopting a new voting method.  In
contrast, AV fails many valuable criteria, and its performance is
particularly bad.  For example,

(a) In electorates with 1-dimensional voting patterns, it is substantially
outperformed by simple runoff voting (the idea that IRO might "reform"
American politics is clearly someone's idea of a bad joke).
(b) In elections involving many candidates, the winner is practically
random.  Accuracy declines drastically as the number of candidates increase.

For these reasons, my intuition is that when the holy grail of the "perfect"
voting method is found (if it hasn't been already), it will be some type of
Condorcet-completion method.  Therefore, this is the area in which I will
continue to devote my energies, and this list seems like the best place to
do it.

Sincerely,

Norman Petry


-----Original Message-----
From: politicians-and-polytopes Moderator
<politicians-and-polytopes-owner at egroups.com>
To: npetry at cableregina.com <npetry at cableregina.com>
Date: April 6, 2000 9:55 PM
Subject: Invitation to join politicians-and-polytopes


>
>Hello,
>
>This is an invitation to join the politicians-and-polytopes group, an email
group
>that I moderate at eGroups, a free, easy-to-use email group service.
>By joining this group, you'll be able to easily send messages to
>fellow group members using just one email address. eGroups
>also makes it easy to store photos and files, coordinate events
>and more.
>
>Here's my introductory message for you:
>------------------------------------------------------------------------
>
>
>A new mailing list on the mathematics of preferential voting has
> appeared. The discussion list is named the "Politicians and Polytopes"
> mailing list.
>
>If this message is not of interest, then please note that none other
> will follow unless you subscribe.
>
>Politicians and Polytopes has a home here:
>      http://www.onelist.com/group/politicians-and-polytopes
>      <mailto:politicians-and-polytopes-subscribe at egroups.com>
>
>
>Also I have a page on my ideas on keeping the best of STV (the Single
> Transferrable Vote) while doing away with a lot of its particulars:
>      http://www.ijs.co.nz/ifpp.htm  (axiomatic voting theory)
>
>Maybe some children might want to join and guess at the final fix
> to the problem of repairing the Alternative Vote (1 winner STV). The
> list can test their suggestions. In STV, a vote for a candidate
> will make that candidate lose, sometimes, so there is a quite a
> change that the list could interact with political issues in
> retarding the progress of electoral reform, under the name of the
> special interest group, known as pure mathematics.
>
>However, the list has 3 subscribers apart from me.
>This message is going to 10 people: some important names miss out
> on getting this message. Two people getting this already got a
> handwritten invitation.
>
>The PaP list will have low activity for quite a while. The list is
> not moderated.
>
>These are pages of the other list, the EM list:
>     http://www.eskimo.com/~robla/cpr/election-methods.html
>     http://www.egroups.com/group/election-methods-list
>
>
>
>Craig Carey
>research at ijs.co.nz
>7 Apr 2000
>
>------------------------------------------------------------------------
>
>TO JOIN THIS GROUP, simply choose ONE of these 2 options:
>
>1) REPLY to this email by clicking "reply" in your email program
>
>-OR-
>
>2) Go to our site at
>   http://www.egroups.com/subscribe/politicians-and-polytopes
>   and click the "JOIN" button
>
>
>If you do not wish to join the politicians-and-polytopes group, just ignore
>this invitation.
>
>Regards,
>
>Moderator, politicians-and-polytopes
>
>
>SPECIAL NOTE FROM eGroups:  Because eGroups values your privacy,
>it is a violation of our service rules for moderators to add subscribers
>to a group against their wishes. If you feel this has happened, please
>notify us at abuse at egroups.com
>
>
>
>
>
>