[EM] James, 5/3/05, 2135

[MIKE OSSIPOFF]

I don't agree that copying the thread-title makes the most useful subject 
line. If I entitled this "Re: approval doesn't meaningfully meet Majority or 
Mutual Majority", that wouldn't tell which posting by that name I'm replying 
to.

James--

You said:

[This is a slightly improved version of the posting that I made 50 minutes
ago. -JGA]

I reply:

Good, that's good that you're improving your postings. But I suggest that 
you should improve them more before posting them.

You continue:

Mike,
	I am not making errors

I reply:

You have been making errors. Lots of errors.

Your posting that I'm replying to now might not have the simplest kind of 
factual errors that your previous ones had, but it still has incorrect 
statements.

You continue:

Main point: The way you word your PMC definition is not meaningful.

I reply:

That's your main point? Your main point doesn't make any sense. In what way, 
for what reason, do you think that the wording of PMC isn't meaningful? Do 
you know what meaningful means? True to form, you repeat that statement 
throughout your posting, without at any time telling why you don't think 
that wording is meaningful. Instead of lreplying to each repetition of that 
statement in your posting, I've replied to it in this paragraph.


You continue:

You
should word it more like your MMC and CC definitions.

I reply:

Again, you apparently forgot to tell us why I should word PMC like MMC and 
CC. PMC isn't MMC or CC, and so unless you can suggest a reason, there's no 
particular reason why it should be worded like them. You repeat that 
statement too, in throughout your posting, and there's no need to reply to 
each repetition, and so this paragraph is my reply to it.

But I'll comment some. You keep saying that, to be meaningful, PMC should be 
worded like MMC and CC. My CC, MMC, and PMC were all written as they were in 
order to comply with the intent of the original criteria that they're 
versions of. None of those criteria is original with me.

So that's why I word them as I do. Predictably there are differences in how 
they're worded, because they're intended to comply with the original intent 
of different criteria.

The other Condorcet versions were around for a long trime before I wrote 
mine, and so was the traditional votes-only MC. MMC is based on a criterion 
that Bruce Anderson proposed in the early 90s, and my MMC is intended to 
comply with the intent of Bruce's criterion.

You post "CC" and "MMC" versions that are worded more like PMC. But those 
criteria can't accurately be called "CC" or "MMC", because that wording 
change makes them too dilfferent from the intent of their original versions. 
You say that your modified versions aren't meaningful. Yes they're 
meaningful. They just aren't what I've sought to test for, and they just 
don't bear any resemblence to the criteria whose names you give to them.

In a previous posting you said that preference criteria should stipulate 
that everyone votes sincerely. I thought that was just a careless statement, 
but apparently you believe it. As you said, preference criteria must 
stipulate how voting is constrained by preferences. But who says that there 
must just be one such constraint, for all preference criteria?

And, by the way, your comparison would be more meaningful (if I may borrow a 
word from you) if you compared PMC to SFC and GSFC instead of to CC. CC 
speaks of everyone voting sincerely, but PMC (like SFC & GSFC) speaks only 
of members of the referrred-to majorilty voting sincerely.

SFC and GSFC are intended to require a lot from a method. MC is one of the 
weakest and most easily-met criteria. They're opposite in that regard. 
Therefore it's odd that you think that PMC should share the voting 
stipulations of SFC & GSFC.

You continue:

Note: In this posting I will refer to your MMC and CC definitions as PMMC
and PCC, because they are preference criteria.

I reply:

Then do you refer to the other CC versions as NRMEECC 
(Nonrank-Methods--Explicitly-Excluded Condorcet Criterion, because it 
explicitly doensn't apply to nonrank methods) or NRMEFCC (Nonrank 
methods-Explicitly-Fail Criterion, because it explicitly says that nonrank 
methods fail it)? Or PPCC (Plurality-Passes Condorcet Criterion, because 
Plurality passes it)?

Actually there's no need for saying PCC and PMMC, because preference CC and 
preference MMC accord with the original intent of those criteria. The 
difference, from NRMEECC and NRMEFCC is that my CC seamlessly applies to all 
methods, with no mention of methods' rules (including balloting rules).

You continue:

[...]

Summary: Points 2 and 3 lend support to the argument that your wording of
PMC is not a meaningful one

I reply:

No. They don't. You don't at any time tell why you don't think that PMC's 
wording isn't meaningful.

You continue:

, i.e. not in accord with the intent of the
majority criterion.

I reply:

You don't at any time justify that claim. Where PMC differs from FHC is when 
FHC says that Plurality passes and Approval fails, due to something that no 
one could seriously call an advantage of Plurality over Approval. PMC 
differs from FHC when FHC malfunctions. If you propose FHC, you're the last 
person who should be using the word "meaningful".

You continue:

	Take your PMMC definition, your PCC definition, and your current PMC
definition. It's kind of like Sesame Street:
"One of these things is not like the others / One of these things just
doesn't belong / Can you tell which thing is not like the others / By the
time I finish my song?"
	The one that doesn't belong is your current PMC definition.

I reply:

It's good that you've learned something from Sesame Street. But you've 
missed their point. They were merely trying to teach you that things can be 
different some particular way even if they might be similar in other ways. 
They weren't trying to teach you that things should be the same. Things are 
often different. That's why we call them different things, instead of the 
same thing. You've failed to understand what Sesame Street was trying to 
teach you.

Say you were presented with a pancake, a square cloth pot-holder, and your 
brother. Which one of those things is most different from the others? Which 
of those three things "doesn't belong"? By your reasoning, you'd believe 
that  it's somehow incorrect for your brother to be different from those 
other two things, and so you'd want to make your brother more like those 
other two things, by making your brother flat and not alive.

But you don't have to do that. It's ok for things to be different from 
eachother. It's alright.

I take the voting systems topic seriously, and so I reply conscientiously to 
people, even to you. But you don't. You're using EM as a role-playing 
playground. If you want to play the role, without doing the work (the work 
of verifying what you say), then I suggest that you take a drama class 
instead.

Mike Ossipoff

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