I vote for Condorcet's own proposal

[Mike Ositoff]

I'm not proposing LOP or FOP as public proposals, due to their
wordiness, complicatedness, and the possible academic criticism
of their Pareto violation. They both need rewording, which
could complicate them further. I won't bother you with the
rewording, since I don't suggest them as public proposals.

Of all the methods we've been looking at that meet Condorcet,
GMC, Pareto, & Independence From Clones, the one described
by Professor Condorcet, in the body of his book, the
one I've been callling Sequential Dropping, is the only one
that's never been undecisive in the examples I've tried the
methods on.

That, and its _great_ brevity & simplicity make it very likely
the best public rank-method proposal.

This is the same as what Condorcet said, right?:

Starting with the weakest defeat, sequentially drop the
defeats that conflict (by forming a cycle) with larger
defeats, till there's an undefeated alternative.

For me, its rival is SC (Smith//Condorcet(EM) ).
SC's rule doesn't talk about cycles or conflicts. True,
Sequential Dropping (which I'll start callling Condorcet1)
is briefer. Difficult to choose between, in regards to
simplicity of explanation to the public or to a Congressmember.
In a way, Condorcet1's wording is better, because it only refers
directly to the problem: the cycle(s), the conflicting
defeats. In a way, SC could be better, because it doesn't
mention them, if people don't want to hear those words.
But when I defined the Smith set on newsgroups, I was accused
of trying to teach set theory, an impression that turned
people off. 

So maybe Condorcet1 is the most explainable, in addition
to its Clone Criterion compliance, which is a nice extra.

***

Actually, I may personally prefer SC's results, though I
consider public acceptability to be much more important than
tiny differences in what we like.

Though SC violates the Clone Criterion, it has some kind of
justification when it fails to choose from a subcycle that
contains winners with respect to the main cycle. I don't
like what Condorcet1 does in my indecisiveness example
(though it isn't indecisive in that example).

It seems to me that both Condorcet1 & SC are both simple
& decisive & the top contenders for a public proposal. Most
likely Condorcet1 is a little better in that regard. If so,
that's what I vote for as a public proposal.

***

Maybe "Condorcet" should refer to Condorcet1.