[EM] Markus, 16 March, '05, 0603 GMT

[MIKE OSSIPOFF]

Dear Markus--

You quoted me:


http://lists.electorama.com/pipermail/election-methods-electorama.com/1997-February/001295.html
>GMC: Never elect a majority-rejected candidate (a candidate
>over whom someone else is ranked by a majority) unless
>every candidate in the set from the method is to choose
>is majority-rejected.

But now you wrote (11 March 2005):
>I've used that term [= "majority-rejected"], but it has
>no role in defining my criteria, and so It gives you no
>justification for your claim that my criteria apply only
>to MinMax.

I reply:

I'm going to repeat this: "Majority-rejected" has no role in defining my 
criteria. GMC is not a criterion of mine. GMC was a criterion of mine. But 
it hasn't been for a long time. As I've said, I no longer use GMC, because 
other criteria better measure for the standards that I consider important. 
"Majority-rejected" has no role in defining my criteria.

You continue:

Only the MinMax(winning votes) tie-breaking strategy satisfies
your "Generalized Majority Criterion" (GMC).

I reply:

Markus, in the message to which you're replying, I told you that PC meets 
GMC, but the Smith-complying Condorcet versions don't meet GMC. You're 
basically just repeating what I've just said.

But you're all confused when you think that that means that GMC applies only 
to PC. GMC applies to all methods (though I don't use GMC).

But even if GMC applield only to PC, you're again all confused when you 
think that that would then mean that my other criteria apply only to PC. If 
you say that you didn't say that, I'll post the recent message in which you 
said it. Not that it's important. I suggest you drop it.

You continue:

Between 1996 and 1997, you used your concept of "majority-rejected"
candidates very frequently to define your "Generalized Majority
Criterion" (GMC)

I reply:

As you always eventually do, you've again begun resorting to repetition of 
statements that have already been answered. I don't guarantee that I'll keep 
answering your repetition.

Though it isn't clear what that's relevant to now, sure, your quote shows 
that I used the term "majority rejected" in a definition of GMC, at the time 
when GMC was a criterion of mine.

You continued:

...and to motivate the MinMax(winning votes) tie-breaking
strategy.

I reply:

It's a fair justification of PC, though I discontinued use of GMC because 
other criteria more usefully measure for the standards of majority rule and 
getting rid of the lesser-of-2-evils problem. Feel free, however, to keep 
promoting GMC if you really want to.

You continue:

It is clear why you now claim that you have never used this
concept to define your criteria: You want to claim that all your
criteria apply to all "winning votes" methods.

I reply:

Let me try to explain this to you again:

1. I said that my criteria aren't defined in terms of "majority-rejected". 
That isn't the same as saying that no criterion that was once a criterion of 
mine, , but which is no longer a criterion of mine, was defined in terms of 
"majority-rejected". If I said that I have never used the term 
"majority-corrected" to define a criterion, then I said it in error, not 
because I'd bother trying to deceive you about a long-disused criterion.

2. GMC is not one of my criteria. It hasn't been for a long time.

3. Do I want to claim that all my criteria apply to all winning-votes 
methods? I do claim that all of my criteria appy to all winning-votes 
methods. As for wanting to, I can't say that I want to keep repeating that 
to you.

Not only do all of my criteria apply to all winning-votes methods, but all 
of my criteria apply to all methods.  Even to methods that aren't 
winning-votes methods.

In contrast, Blake's Condorcet Criterion, and probably some of his other 
criteria, apply only to rank-balloting methods.

4. But if GMC didn't apply to all methods, or even to all winning-votes 
methods, and even if I still called GMC a criterion of mine, that wouldn't 
mean that my other criteria don't apply to all methods.

5. Even though I now don't consider GMC to be a criterion of mine, GMC 
applies to all methods, not just to all winning-votes metods, and not just 
to PC.

You continued:

You want the readers
to mistakenly believe that you proposed "winning votes" methods in
general and not only MinMax(winning votes).

I reply:

At the time when I proposed wv, PC & Smith//PC had been proposed by me on 
EM.  I didn't mention  BeatpathWinner, SSD, or Ranked-Pairs using wv,  
because BeatpathWinner, SSD, and Ranked-Pairs weren't known on EM at that 
time.

But I clearly and unmistakeably proposed wv as a measure for the strength of 
pairwise defeats, and advocated wv as the most useful measure for the 
strength of pairwise defeats. And I told of the strategy advantages 
resulting from wv.

The difference between BeatpathWinner and the other wv methods is 
insignificant compared to the difference between wv and margins.

By the way, you've been making some pretty silly mis-statements for some 
time now, mis-statements that would be obvious to anyone but you, and you've 
begun repeating your mis-statements that have already been answered. It's 
around this point that I usually begin suggesting that there's something 
wrong with you, or suggesting that maybe you're an idiot.  But I'm not going 
to do that this time. But this time I won't keep replying to statements that 
you keep repeating.

Mike Ossipoff

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