>I've heard people refer to that
> >kind of an IIAC, but I've never heard a precise definition of it.
> >Nor have I heard a claim that any method could pass such a criterion.
>
>First, when I said that IIAC was meaningless for non-ranked ballots,
>it's important to understand that I meant that in the sense of "an
>election between two government appointees is meaningless", not in the
>sense of, "the colour of sound is meaningless".
>
Whatever you meant by "meaningless", it isn't at all clear why
you think that it's meaningless that Approval and Plurality
pass IIAC.
On the contrary, it's very relevant, because IRVies, as I said,
always bring up Arrow. It's relevant, then, that Approval, but not
IRV, passes Arrow's outcome-criteria.
>With that said, can you explain why anyone would care about whether or
>not IIAC is passed by a non-rank ballot method, like plurality or
>approval?
Yes. Because Arrow's criteria are brought up so often. Because they're
brought up often by people who advocate IRV against Approval.
As I said, that makes it relevant to point out that Approval, but
not IRV, passes Arrow's outcome-criteria.
IIAC isn't one of the criteria important to me, but it's quite relevant
because Arrow is brought up so often.
You seem to be saying that the relevance of a method passing IIAC
depends on whether that method is a rank method, and it isn't at
all clear why you believe that. Criteria are written because someone
is saying that they're desirable to comply with. If a method complies,
it complies, regardless of its balloting or other procedural details.
I notice that you didn't give a precise definition of the version
of IIAC that you vaguely referred to, and that you didn't state what
method you think would meet that criterion.
>Whether approval will do better seems to depend a lot on the
>intelligence of the electorate, and the availability and accuracy of
>polls. With enough information, and proper strategy, any method will
>give the sincere CW.
Yes, that's common knowledge here. The methods differ, however,
in what those people will have to do in order to accomplish that.
IRV will often require people to dump their favorite by voting
someone over him/her. Approval will never do that.
>Approval allows the strategies involved to be less
>extreme. However, the prevalence of strategy in approval makes any
>outcome seem possible,
I understand your dislike of strategy. Then advocate Condorcet.
Condorcet is the Cadillac of voting systems. But don't criticize
Approval because it isn't Condorcet. I'm the first to agree that
Condorcet gets rid of strategy-need better than Approval does.
However we're now comparing Approval to IRV, not Condorcet. IRV
requires more drastic strategy than Approval does. In IRV, as in Approval,
people's use of strategy, informed or uninmormed, will
affect the outcome.
and allows for some very optimistic approval
>examples.
I've answered many postings here that gave what were intended as
pessimistic Approval examples. I don't think I've posted Approval
examples here.
>Consider that both you and Bart advocate approval, and both
>of you give examples showing how well approval does, but your examples
>show how a CW would be elected where it wouldn't in other methods. His
>examples show how approval can keep CW's out if their average ratings
>aren't high enough.
In Bart's example, the low-rated CW won't win in a 0-info election,
if people use their best strategy, but the CW wins if voters have
sufficiently good information about eachother's preferences, as
it always would, as you said. What's the problem then?
Rather than giving Approval examples, I've replied to other people's
Approval examples, it seems to me. I don't ask as much of Approval
as you do. I don't want people to have to dump their favorite.
Approval acts on candidates' overall high-ratedness, unlike IRV,
which only looks at first choices. Approval quickly homes in on the
voter median position and stays there. I could go on. I will within
a few days.
>
>On the other hand, it seems like many Americans (and I'm not implying
>that Canadians would be better) seem to have enough trouble with a
>plurality ballot. I'm not convinced that the general public will
>understand approval well enough to use it effectively. Although they
>probably won't misunderstand it quite so much as Mr. Davison has.
They know whom they want to vote for in Plurality, though I don't
agree with them. They'd know that nothing is stopping them from
voting for Nader in addition to Gore. That's good enough.
Will they use Approval's mathematical strategy? No, just as they
don't use Plurality's mathematical strategy. But they know that
when they vote for the person they'd vote for in Plurality, there
is nothing preventing them from voting for everyone whom they like
more.
>
> >>I certainly wouldn't argue that IRV is the best possible. On the
>other
> >>hand, I wouldn't try to organize against the efforts of people trying
>to
> >>implement it.
> >
> >But I bet that would depend on how honest those IRVies are, and
> >how sleazy their tactics are. Sure, if the IRVies were honest, I
> >might not oppose them in the way that I do.
>
>I'm not sure about "sleazy tactics. "Sleazy tactics" sounds like smear
>campaigns, or brief cases full of money going to pay off congressmen.
You mean like CVD giving lots of money to California LWV, and then
having a majority on the study committee, and being able to
exclude Approval from the LWV website where IRV is promoted, and
being able to write the study materials on both IRV & Approval,
with no input allowed from Approval advocates in LWV?
>Why are you worried that plurality might meet the criterion.
When Bruce Anderson proposed MMC, he didn't consider Approval and
Plurality to pass it. I'm not worried that Plurality will pass.
I merely would like the criterion to comply with the intent of its
proposer. Likewise the Condorcet Criterion, for instance, isn't
intended to be a criterion that Plurality passes, or which no method
passes. That's why I wrote a better wording for it that lives up
to the intent of people who use CC.
>Are you
>concerned that someone might later take your criterion out of context,
>and use it to defend plurality, even where this would be meaningless?
I suppose that's possible. Mostly I want MMC & CC, etc. to live up
to the intent of the people who use them.
>Despite its pleasant name, I don't care about the "defensive strategy"
>criteria.
If you don't care about LO2E, then we're not going to reach any
agreement.
However, in this example
>
>46 D A B C -- A B C much lower rated
>19 A B C D -- D much lower
>18 B C A D -- D much lower
>17 C B A D -- D much lower
>
>IRV picks B. Unless the A/B/C voters really get their act together, or
>have some kind of primary, D will win. If one of A/B/C does win, who
>will decide which one? A very small group of voters who were counting
>on other people to make sure D lost.
You seem certain that the A, B, & C voters will bullet-vote.
Why should they? They despise D. Surely beating D is more important.
If it's 0-info then more than their favorite is probably above their
mean rating. If they use strategic value, they're unlikely to
bullet vote.
Anyway, I'll tell you one thing they won't do: They won't vote anyone
over their favorite.
>
>Picking D is a failure of MMC and GITC (for a ranked method). Is
>picking one of A/B/C a violation of DSC.
I must admit that I'm not quite sure why you think that picking from
{A,B,C} violates WDSC or SDSC.
Mike Ossipoff
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
Share information about yourself, create your own public profile at
http://profiles.msn.com.