[EM] David Gambld reply, 25/1/04, 0620

[MIKE OSSIPOFF]

I'd said:

>The CW is the social utility maximizer.

You replied:

Not always. Please justify this statement

Mike replied in summary:

With 1 issue dimension, the CW is the SU maximizer, if disutility is
measured by distance.

Thankyou for this clarification. Issue space in most political systems is
principally but not entirely one-dimensional (IMHO).

I reply:

"Principally" means that it's the best guess. There's a strong tendency for 
issues to be linked, correlated, which makes for a nearly 1-dimensional 
issue-space.Sure there are exceptions.

But, as I said, Merrill's multidimensional spatial simulations also showed 
Condorcet and Approval doing much better than IRV, by SU.

I don't know whether or not the CW is always the SU maximizer with more than 
1 dimension. There'd obviously be a tendency in that direction. But either 
way, Merrill's simulations suggest that IRVists would do better to not bring 
up SU.

You cointinued:

There can be other types of
disutility.

I reply:

Yes, and I answered that in the message to which you're replying.   The 
voter median position won't be occupied only by an unknown, noncommittal, 
evasive candidate, under the conditions that would result from the adoption 
of Condorcet.

You continued:

For example in the recent election for Governor of California
there was probably a great deal of disutility of knowledge.

I reply:

What's that??

You continued:

Due to the number of
candidates most voters would probably been unable to tell you the names of 
more
than a handful of candidates let alone what they all stood for.

I reply:

That isn't the same as saying that the only candidate near the voter median 
will be an unknown.
Known candidates will compte there too, and some of those will be sincere in 
their advocacy of voter median positions.

You'd said:

A common way for somebody to
promote a 'pet' electoral method is to do the following:

1/ Find a set of criteria that your system meets and state them.

2/ Dismiss the criteria your method doesn't meet as irrelevant or even
better
just don't mention them

I'd  replied:

During the time that EM has been in existence, we've had to answer this
abyssmal ignorance many times. There should be a FAQ with answers to all the
stupid questions like that. Or else David should check the archives better
before presuming to tell us how it is.

David apparently is using his ESP here.  Because, without ESP, how does
David know that someone started by choosing a method, and then searched
around for criteria that the method would meet.

You  replied:

A way people often come to advocate or like a thing is like this. They look
at something superfically think "this is a good idea" and agree with it.

They
then examine their decision in more detail and look for strong reasons to
justify their choice.

.
I reply:

ESP again? Or did someone tell you that that's what they do? If so, then who 
told you that they do that?

You continued:

Sometimes people change their minds. Many people on the EM
list originally thought IRV was a good idea but on later consideration 
switched
to either supporting Approval or Condorcet. The initial opinion often comes
before the detailed knowledge and may change as a result of gaining that
knowledge.

I reply:

Another amazing discovery by David. But it isn't quite clear how you can 
tell when someone is misrepresenting their reason for liking a method. Are 
you the arbiter of that?

You continued:

As regards the criteria to which to evaluate elections methods by at
electionmethods.org we have the following:

Monotonicity.

Two Condorcet criteria, which surprise, surprise only Condorcet meets.

I reply:

So now are you also saying that Condorcet first decided that he liked his 
method, and then contrived a criterion that it would meet? I suggest that 
all that deviousness is in your mind.

You continued:

Five strategy criteria (SFC, GSFC, SDSC, WDSC,FBC).

Also it is worth pointing out that only one effective strategy is needed to
manipulate an election method if offensive burial is effective (as it is in 
WV
Condorcet methods) second,third,fourth and fifth strategies are unnecessary.

I reply:

Apparently David believes that each criterion is for a different offensive 
strategy :-)

David, it's a good thing for strategists that only one offensive strategy is 
necessary, because only one is available in wv Condorcet.

I've said many time, but apparently must again repeat, that, though IRV 
doesn't have offensive order-reversal, it requires, without any offensive 
strategy being done, a much more drastic defensive strategy than Condorcet 
wv does. IRV will make the need for defensive favorite-burial. That need 
happens without anyone using offensive strategy. That's the mark of a really 
poor method. IRV and Plurality have that in common.

At worst, in a really devious electorate, where there are many who'd like to 
do offensive order-reversal, Condorcet then begins to have a strategy 
problem. But that strategy problem isn't like IRV's strategy problem. IRV 
requires defensive favorite-burial. But offensive order-reversal can be 
deterred by mere defensive truncation.

You continued:

Finally we have the summability criterion which isn't a proper election
method evaluation criterion at all.

I reply:

I guess David isn't going to let us in on how he knows that.

You continued:

In the context of British General elections in
which all the ballot boxes are transported to a single central location to 
be
counted it looks pretty irrelevant.

I reply:

David, it isn't entirely clear how the transportation of ballot boxes makes 
the Summability Criterion irrrelevant. Did you think that the problem with 
non-summable methods is the infeasibility of carrying the ballots to the 
central counting location? :-)

The website describes the serious problems of nonsummable methods.

You continued:

It is also worthy noting that Australia
managed to conduct IRV elections in 1919 over some very large and sparsely
inhabited constituencies

I reply:

Yes, low voter population definitely helps IRV.

You continued:

This is a criterion to do with different electoral administration
arrangements in different countries,

I reply:

It's a criterion to do with the need to do one big central count, which 
hugely increases the ease of fraudulently altering the election results. 
Check the website again. In Condorcet or Approval, simultaneous local counts 
can be done (You know, parallel processing), while in IRV, it's necessary to 
do one big national count that stores and keeps referring to each of the 100 
million rankings in a U.S. presidential election.

You continued:

it is not a valid criteria to assess an
election method against.

I reply:

Dave, "criteria" is a plural. The singular is "Criterion".

You make a lot of statements that you don't suport because you can't.

You continued:

Where is latter-no-harm?

I reply:

I refer you to my posting entitled "Woodall's Whacky, Zany Criteria".

You continued:

, what about participation?

I reply:

I make no secret of the fact that Condorcet (like IRV & all rank methods 
except Borda) doesn't pass Participation. Of the methods proposed on EM, 
Approval and CR are the ones that meet Participation. I like Approval and 
CR. So what's your conspiracy theory for why we don't list a criterion that 
favors Approval & CR?

Even though Participation favors Approval & CR, in comparison to IRV, we 
don't list Partilcipation because we don't consider it to be one of IRV's 
worst problems. It would be desirable for a method to meet Participation, 
but it doesn't seem important compared to IRV's other failures. And, though 
Participation compliance would be nice, we don't consider it as important as 
the things that wv guarantees.

We're not obligated to list every criterion. Websites put up what the owner 
considers important. It would be prohibitively cluttering and confusing to 
readers if we listed every criterioni that we don't consider as important.

Though wv & IRV both fail Participation, the difference is that wv offers 
what we consider some valuable guarantees that (we claim) more than offset 
the Participation failure. A great many agree with us about Participation, 
because lots of people advocate rank methods that don't meet Participation.

David will say that we're being high-handed or dictatorial by not listing 
every criterion that anyone has ever proiposed. We want to keep things 
simple and uncluttered. Uncluttered with criteria that we don't consider as 
important.

Yes, David's opinion of what's important may be different from ours. But 
until David is appointed Fuhrer, he will just have to accept that our 
choices for our website are based on what we, not David, consider important.

You continued:

A longer and more objective set of criteria used to be found at Blake
Cretney's Condorcet.org ( which appears to have vanished).

I reply:

Poorly defined criteria. For instance, Blake's Condorcet Criterion is met by 
Plurality. Or would be, except that Blake stipulates that Condorcet's 
Criterion only applies to rank methods. A bit useless, that. The criteria at 
our website, including Condorcet's Criterion, apply to every proposed 
method.

You continued:

Mike Wrote:

So perhaps you'd be so good as to give us an example of someone making their
criteria sound as good as possible, and really building them up, by saying
something that you can show to be unjustified.

Or is it also bad to discuss a criterion's value even what one says is
justified?

electionmethods.org provides an example of making your preferred method look
as good as possible by failed to mention criteria it doesn't meet.

I reply:

You mean because we didn't include the sillly Later-No-Harm? Or 
Participation, along with every criterioni that we consider less important?

By the way, let me say that I do consider Participation to be a valid way of 
comparing Approval and IRV, because, as I said, IRV doesn't have the 
desirable propoerties by which Condorcet outweighs its Parttilcipation 
failure. We'd have included Participation except for the desire for brevity.

But I'm going to tell Russ that people (or at least one person) are 
clammorning for Partilcipation, and suggest the possibility of listing it. 
I'll go by Russ's judgement about whether that would overly clutter the 
criteria list and compliance table.

You'd said:

4/ Describe your selection of criteria as 'objective' ones that have been
selected by 'experts'

I'd replied:

Again, who has said that their preferred criteria are the objective ones
that have been selected  by experts?

You replied:

I quote from electionmethods.org

"The Election Methods Education and Research Group (EMERG) was founded by
Russ Paielli and Mike Ossipoff. Mike is an expert on election methods"

I reply:

You'd seemed to be talking about claims that _someone else_ has selected our 
criteria. It wasn't evident from what you said that you were referring to 
someone's own selection of their criteria.

You continued:

I appears that you did!!!!!!!

I reply:

If you're trying to imply that we use the word "expert" to justify our 
claims, that's rubbish. I've always said that methods, criteria, claims, 
etc. are to be judged on their own merits. Instead of speculating about 
motives, it would be better if you tried to understand criteria better. The 
merits of the criteria and methods that we propose don't depend on a 
characterization of me. If you want to find fault with the criteria or 
method recommendations, then talk about them.

Unlike Rob Richie, of CVD,  I've never called myself an expert. Russ said 
that. But Russ didn't say that that's how we justify our criteia and our 
method recommendations. So you're saying that we're using a dishonest or 
unethical tactic, because Russ referred to me as an expert? Are you saying 
that he was misleading readers? Are you aware that now you're really getting 
down in the mud? First you're telling us your theories about the order in 
which people choose standards and methods, implying that we're 
misrepresenting that order. Now you suggest (if that's what you're trying to 
say) that Russ has misrepresented my qualifications.

The person who can't make his case on the merits of facts about the topic, 
will often do what David is doing: Speculate about the motives of others. 
Theorize that they're misrespresenting their reasons for their advocacy. And 
now, though David's point here isn't at all clear, he seems to be finding 
fault with Russ's reference to me as an expert.

I'll tell you one thing for sure, Dave: Someone like you isn't qualified to 
discuss whether or not I'm an expert.

If an expert is someone from whom you need to learn, then yes Dave, to you 
I'm an expert.

You continued:

"EMERG is anon-partisan organisation. We believe that election methods 
should
be evaluated objectively and mathematically, without regard to which
political parties or ideologies might benefit."

Though not objectively against a complete set of criteria it appears.

I reply:

I've told you that it wouldn't make any sense to list every criterion that 
has ever been proposed.

You continued:

"The choice of an election method should not be based on subjective notions"

" It should be based, rather, on set of strictly objective technical 
criteria
"
Again a partial selection of criteria based on subjective notions about what
you consider important is not in my opinion the selection of an election
method based on strictly objective technical criteria.

I reply:

I've always said that standards are an individual, subjective, relativist 
matter. But the choice of criteria to measure for a standared isn't as 
subjective.

Balancing brevity against someone's request that we list every criterion is 
unavoidably a subjective choice. We jusge how many criteria we can list 
without causing confusion among readers.

Finally Mike said:

David, yoiur participation is making this mailing list have to act as an
elementary school. It would be better if you would start reading through the
list's archives, so that you wouldn't repeat elementary questions that have
been answered many times in the archives.

Dave's reply:

As regards the kind of behaviour you might find in an elementary school
remember the Floyd Algorithm.

I reply:

I was referrting to all the elemenatry answers that are needed when replying 
to you, answers that we've given so many times in the past.

If you believe that I was unreasonable to lose patience with Markus's 
endless  repetitions of replies to things that I'd never said, and 
mis-statements of what I had said, that's your subjective judgement.

Mike Ossipoff

_________________________________________________________________
Check out the coupons and bargains on MSN Offers! 
http://shopping.msn.com/softcontent/softcontent.aspx?scmId=1418