[EM] Markus wants to justify my remarks.

[MIKE OSSIPOFF]

EM list members:

I'm grateful to Markus for this posting. Apparently Markus wanted to
show you why I said the things about Markus that Blake quoted. Markus,
I really appreciate that.

Yes I've been known to lose patience with some twit like Markus, and
say something uncomplimentary. Check the following message to judge
if I do unjustifiably.

Another thing that this helps clarify: Sometimes I've said things
like "Either this person is an idiot or is arguing dishonestly."

Anytime I say that someone might be dishonest, Blake calls that paranoid.
No, regrettably dishonesty isn't at all unusual. Statistically, it's
common human nature. Notice that the above sentence doesn't "ascribe"
anything to dishonesty, but merely mentions it as one theory, the
alternative to idiocy. Blake always seems to miss that.

Well, the letter that I'm replying to here is one about which I
could say that either its writer is an idiot or is arguing dishonestly.
No, I don't say that because I'm mean. I bring it up because it
relates to Blake's charges.

I'd said:

>Since I didn't say anything about
>how the others vote, and since I said that given a certain premise
>a certain requirement must be met, then any conditions that isn't
>in the criterion's premise is not part of the criterion's premise.

Markus said:

(1) If I understand your comment correctly, then "it ensures that B
cannot win" simply means: "B doesn't win."

I reply:

In general, I'm not interested in hearing various of the innumerable
possible rewordings of my criteria, with a question about whether
the rewording means the same thing.

But this time, I think we can say that your wording above could be
substituted in the criterion if one wanted to, though my wording
is clearer.

Markus continues:

Then your WDSC looks as
follows:


   If a majority of the voters strictly prefers A to B then there is
   always a situation where B doesn't win and where no member of that
   majority votes a less-liked candidate over a more-liked candidate.

I reply:

Correction: It may look that way to you.

I don't know what your point is. Now we know how WDSC looks to you.
Are you asking me if your way of looking at WDSC is what I mean?
It isn't what I mean. I meant what I said. There are innumerable
things that you could write, to ask me if they're the same as WDSC.
Obviously I don't have time to deal with every question of that type
that you could ask.

I must admit that I _don't_ notice a similarity between how WDSC looks
to you, and the actual WDSC that I wrote.

You're saying that if a majority prefer A to B, then there's always
a situation where B doesn't win. I don't know what it means to say
there's always a situation where B doesn't win. Maybe you mean it's
possible to contrive a set of preferences & ballots for which B doesn't win 
and
for which the members of that majority don't reverse a preference.
If so, then your criterion is passed if we can find any configuration
of the all the votes that makes B lose without that majority reversing
any preferences. Ok, say a majority prefer Gore to Bush. With Plurality,
can we contrive a ballot-set in which Bush loses and the none of the
Gore>Bush majority reverse a preference? Sure. Write an example in which
all of that Gore>Bush majority have Gore as their favorite. The example
writer can configure any variables that aren't specified in the criterion's 
premise. In fact, we could even say that in our example,
100% of the voters have Gore as their favorite, and they all vote for
Gore. And so Plurality meets your criterion. We have an example in which
a majority prefer Gore to Bush, and Bush loses, and no member of that
majority reverses a preference. So Plurality passes your criterion.

But I'm not going to keep evaluating your every rewording of my criteria.
If you think you can find a problem, then tell me what it is. But
don't keep wasting my time with your rewordings.

By the way, it should be obvious to anyone that your rewording
doesn't have the same meaning as WDSC's actual wording.

Markus continues:

(2) You wrote that you "didn't say anything about how the others vote."
Does that mean that it is sufficient when there is at least one (not
necessarily sincere) voting behaviour of the voters of the minority so
that B doesn't win and no member of the majority votes a less-liked
candidate over a more-liked candidate?

I reply:

No. It means that I didn't say anything about how the others vote.

Or maybe your asking if what you're saying is a consequence of
the fact that I didn't say anything about how the others vote.

I could answer that better if I knew what you were trying to say
in that paragraph.

I don't know if what you say is a consequence of WDSC not saying
anything about how the others vote. But let me tell you what the
consequence is: Because the criterion doesn't mention how the others
vote, then, no matter how the others vote, if the criterion's premise
is met, then the criterion's requirement must be met.

Let me say that for you again: Because the criterion makes no
mention or specification or stipulation about how the others vote,
then, if the criterion's premise is complied with, then the criterion's
requirement must be met in order for the criterion to be complied with,
regardless of how other people vote.

In other words, if we find an example, with a particular method, in
which the criterion's premise is met, and the criterion's requirement
is failed, then the method fails the criterion, no matter how the other
voters voted. No matter how the other voters voted, and no matter what
sign of the zodiac the moon is in, and no matter whether your dog
has fleas, etc. That's what it means to apply a criterion. The premise
tells the conditions under which the requirement must be met. If
the premise is met and the requrirement is not met, that's what it
means to fail a criterion. Do you understand now?


If you still don't know what it means to meet a criterion, then you should 
be
asking some more elementary questions about that before asking about
specific criteria.

But let's take a closer look at Markus's question, and try to
guess what he means:

(2) You wrote that you "didn't say anything about how the others vote."
Does that mean that it is sufficient when there is at least one (not
necessarily sincere) voting behaviour of the voters of the minority so
that B doesn't win and no member of the majority votes a less-liked
candidate over a more-liked candidate?

I reply:

My best guess is that Markus is asking if WDSC is met as long as
it is possible to write some configuration of all the votes of
voters outside the majority referred to in the criterion, such that
B doesn't win and no member of the majority reverses a preference.


Again let's test Plurality. A majority prefer Gore to Bush. Only
one voter isn't part of that majority. He votes for Bush. The majority
have Gore as their favorite. They vote for Gore. Bush loses. The
majority hasn't reversed any preferences. Plurality passes.

In fact, this is pretty much the same asinine question that Markus
asked before, isn't it.

And Blake wonders why I become impatient with someone like Markus,
with his inane questions that he sometimes continues & repeats
on & on.

Now, I should tell why Plurality fails WDSC: Say a majority prefer
A to B, as the premise stipulates. 60% prefer A to B. But, 5/6 of those
have C as their favorite. Say the other 40% of the voters vote for B.
Can the A>B 60% make B lose without reversing a preference? If the
5/6 vote for A, they're reversing their preference for C over A.
And if they don't vote for A, then A only gets the votes of 10% of
the voters.

So even if a majority prefer A to B, we can devise an example in which
B wins, with Plurality, unless members of the A>B majority reverse a
preference. Plurality fails WDSC, though it meets Markus's criterion.

Mike Ossipoff








_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com