Puerto Rico election

[Mike Ositoff]

Yes, that sure does sound like an instance where our
methods might be welcome.

Whom should we e-mail in Puerto Rico? Or are the decisions
on the voting made in the U.S.? If so, whom should we e-mail
here?

The Condorcet advocates here should have a named organization,
like Condorcet Society or something. I suggest that we send
an e-mail, from the members of that society (maybe admitting
that membership so far may be less than 10, so as not to
be accused of trying to deceive about that).

Maybe SD is the best thing to propose. Simplicity &
easy explainability likely would outweigh Schulze's merit
improvement of meeting Independence From Clones with no
exceptions.

Would it be assumed that the proposal would be SD, or
is that something to vote on?

Choosing between SD & SC is, for me, mainly a matter of
explainability, though SD certainly seems better.

Maybe we should start drafting a letter.

Maybe something like:

We've heard that you're going to hold a public election
to choose between 4 alternatives, on a crucially important
issue. The issue is too important to use the old "Plurality",
or "First-Past-The-Post" system, where people vote for 1
alternative, and the one with most votes wins. The problem
with Plurality is that it often violates majority rule
and elects something that has a majority against it, and
defeats something that would beat each one of the other
alternatives in separate 2-way elections.

How would it be if the alternative that won your vote
had a majority preferring one of the other alternatives to
it. For the sake of public satisfaction with the choice,
it's obviously desirable to avoid that.

Here's how to avoid it. The founder of voting theory,
in the 18th century, by the name of Condorcet, suggested
the solution:

* Let voters rank the alternatives in order of preference.

* If more voters rank A over B than vice-versa, than A beats
B.

* If one alternative beats each one of the others, then it wins.

* If no 1 alternative beats each one of the others, it's
  because of a "cycle" in which A beats B beats C beats A.
No problem! Starting with the weakest defeat (as measured
by how many voters rank the defeater over the defeated),
sequentially drop defeats that conflict (by forming a
cycle) with stronger defeats, till there's an undefeated
alternative.

***

Any suggested modifications, or alternative letters?

As Norm pointed out, SD has the advantage of being a strong
candidate for being exactly what Condorcet meant (but
with votes-against).

Also, its definition, while necessitating the mentin of
cycles, avoids the definition of the Smith set. 

Would this be less accepted?:

Use the Least Beaten rule to choose from the Smith set.

Least Beaten rule:

If no 1 alternative beats each one of the others, then the
winner is the one whose greatest defeat is the least, as
measured by how many voters rank the defeating alternative
over it.

Smith set:

The Smith set is the smallest set of alternatives such that
every alternative in the set beats (as defined above) every
alternative outside the set.

***

Which of these definitions is more likely to be accepted,
SD or SC?

***

For maximum simplicity, the proposal could just be 
plain Condorcet (PC), which could be called the
Least Defeated rule (LD). As defined above.

***

Any suggestions about whether to propse SD, SC, or PC?

Take a vote? Or just propose SD?

***

Mike