[EM] Russ, 2 March, 1952 GMT

[MIKE OSSIPOFF]

I´d said:

    No, the Condorcet's Criterion that was at the website at the technical 
evaluation page was a Mike style criterioni entirely. It referred to the CW 
(which you called the IDW, defined as I define the CW), and it stipulated 
sincere voting. Both the CW and sincere voting are defined by me in terms of 
preference, and were defined at your website in terms of preference.


Russ replied:

If you stipulate "sincere" voting, then you are eliminating the voting
strategy and stipulating that the actual votes cast are identical to the
"sincere" or true preferences.

I reply:

Wrong. You need to read my definition of sincere voting before you try to 
quote it.

Since you don´t, or shouldn´t, have it at your website now, I´ll write it 
for you here:

A voter votes sincerely if s/he doesn´t falsify a preference, or fail to 
vote a preference of his/hers which the balloting system in use would have 
allowed him/her to vote in addition to the preferences that s/he actually 
does vote.

[end of sincere voting definition]

The wording isn´t as smooth as one might like, but it´s unambiguous.

Russ continues:

In that case, a Mike-style criterion is
equivalent to a standard tally-rules-based criterion.

I reply:

...except that no-preference ("tally-based) criteria don´t make reference to 
preference, and my sincere voting defintion refers to preferences, and so 
any criterion that stipulates sincere voting refers to preference and is 
therefore a Mike-style citerion.

Further, the my Condorcet´s Criterion defintiion additionally mentions te 
CW. The CW is defined in terms of preference. So my Condorcet Criterion, the 
one that you had at your website while you were permitted to, is entirely a 
Mike-style criterion.

Russ continues:

You can take any standard criterion, play this trick, and call it a
Mike-style criterion.

...by adding to it a stipulation of sincere voting and mentioning the CW? 
You catch on fast :-)
Then it would indeed be Mike-style criterion.

Russ continues:

You will only be obfuscating the issue and adding
nothing of any value, however.

I reply:

You´re all befuddled and confused about what you mean by a Mike-style 
criterion.

That's exactly what we did with the old
Condorcet criterion on our former website

I reply:

Congratulations, that´s correct. We posted a Mike-style criterion when we 
posted Condorcet´s Criterion.

Russ continues:

, and looking back at it, I
realize it was an embarrassment.

I reply:

You suddenly realized that my criteria weren´t any good a few days after I 
told you that you no longer had permission to have them at your website :-)

Russ continues:

The Condorcet criterion can and should
be defined in terms of actual votes only.

I reply:

Yoiu´re saying that it _can_ only be defined that way? Well, if Condorcet 
himself said it that way, then one could argue that it wouldn´t be 
Condorcet´s Criterion any other way. But then you have the choice of saying 
that Plurality passes, or specifically saying that only rank methods pass, 
or specifically saying that Plurality fails, or saying that the method only 
applies to rank methods. Then you have a rules criterion, a criterion that 
makes requirements or premise stipulations that mention a method´s rules. I 
prefer Mike-style criteria to rules citeria. Mike-style criteria are about 
results only, not rules.

Anyway, as I was saying, if Condorcet stated a rules criterion, then you 
could say that, strictly speaking, CC must be a rules criterion. But I call 
my version Condorcet´s Criterion because it´s the Condorcet´s Criterion 
version that is not a rules criterion, and which acts as one would expect CC 
to act, anc whilch doesn´t have the problems that CC, and some other 
criteria, would have if they don´t mention preference.

Now, as for "should", if all you´re saying is that CC _should_ be defined 
without mentioning preference, then of course you´re merely expressing a 
subjecive preference of yours, and of course you have a right to your 
(newly-acquired) opinion.

I´d said:

No, that Condoret's Criterion was thoroughly a Mike style criterion, for the 
reasons stated above.

Russ said:

Then "that" Condoret's Criterion wasn't the standard Condorcet
criterion.

I reply:

Why don´t you advocate that we keep the current "standard" voting system, 
Plurality?

I don´t know if there´s just one standard CC definition. It seems to me that 
I´ve noticed a few non-preference versions that did silly things, such as 
say that Plurality passes CC, or make rules stipulations in their premise.

Which of those silly versions do you prefer?

But no, my definition is not the former standard definition, if there is a 
standard definition.

But my non-rules criterion Mutual Majority has gotten quite a bit of use.

Russ continued:

I could just as well define my own version of monotonicity
too, but what would it accomplish other than making me look like a fool?

I reply:

I don´t doubt for a minute that that´s exactly what it would do. But, on the 
other hand, that has never stopped you.

But if the traditional Condorcet Criterion has a fault, or if the various 
"standard" versions all have their various faults, then hopefully you´ll 
forgive me if I propose a different one. I use CC very little. It was you 
who wanted to have my CC version at your website. I wanted to mention at the 
website that it isn´t the standard version, and you opposed mentioning that.

My CC is so similar to the standard one(s) that it´s well justified to call 
it CC.




I´d stated:

If no one falsifies a preference, and if a majority prefer the CW to 
candidate Y, and vote sincerely, then Y shouldn't win.

Russ replied:

Actually, I think Approval passes this criterion. If that majority draw
their cutoff between X and Y, then Y can't win.

I reply:

Look, imbecile, what if I write an example in which they don´t do that? 
SFC´s premise conditions stipulate that no one falsifies a preference, and 
that a majority prefer the CW to Y and vote sincerely. Those are the only 
premise conditions, conditions that I have to abide by when writing a 
failure example.

The whole point of SFC is that all those voters need do is vote sincerely. 
You think that the criterion is complied with because you can find a sincere 
way they can vote that will ensure that Y won´t win. But, with a complying 
method, all that´s required of them is that they vote sincerely.

You want Approval to pass SFC because if those voters guess where the cutoff 
should be, they can keep Y from winning. As I said, you´ve completely missed 
the point of SFC, which is that the only requirement on them is that they 
vote sincerely.

Russ continues:

I realize that some criteria are defined for ranking methods only, but I
clearly recall that Mike intended for his criteria to apply to all
methods (as implied by the compliance table at the top of the old
"Technical Evaluation" page).

I reply:

You´re quite right. My criteria apply to all methods. (But FBC needs a small 
fix, which I´ll state soon, and SARC needs some work too, if they´re to 
serve the spirit and intent of FBC when the method is RC or something like 
it).

SFC applies to Approval, and approval doesn´t pass. Neither do Plurality, 
IRV, Borda, or margins Condorcet.

Russ continues:

The question is then whether rating candidates equally in Approval
constitutes a "sincere" vote when the voter doesn't really consider them
equal.

I reply:

Yes it does. The definition of sincere voting was at your website for a long 
time.

Russ continues:

I say Approval passes SFC as stated above.

I reply:

What was that you said about making a fool of yourself?

Mike Ossipoff

_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/