[EM] Papers are voters: asserting aspects of voters

[Craig Carey]

I ask subscribers to regard this as a puzzle that I haven't understood.
Yet there seems to be little stated concern that the list is less than
what subscribers might want or hope for.

This may contain errors. Please imagined corrected as reading. I can
correct errors in private too.


At 01:06 08.10.00 +0000 Sunday, MIKE OSSIPOFF wrote:


>> > Any BC complying method meets GSFC, because, since B is majority-beaten
>> > from the sincere Smith set (this time I mean sincere Smith set), and,
>> > since no one falsifies a preference, it's impossible for any member
>> > of the sincere Smith set to have a majority defeat from outside that
>> > set. That means that there can't be a majority beatpath from B to the
>> > sincere Smith set candidate who beats B. That means that B's defeat
>> > from the sincere Smith set can't be the weakest defeat in a cycle.
>>
>>Which properties are necessary and sufficient for an example for being
>>an example showing that a given election method violates GSFC? Which
>>properties are necessary and sufficient for an example for being an
>>example showing that a given election method violates SDSC?
>
>Again, I'm not quite sure what you're asking. I've demonstrated that

Didn't you just argue with Mr Schulze in the "ambiguous" argument, and
claim that the omission of a "For All" was not significant?. Mr Schulze
was plainly enough asking you to write briefly and test an infinity of
election examples (points) for all preferential voting methods.

You understood when you were commenting against Mr Schulze over nothing
more than a mistake in choosing words. You fail to understand when the
issues is that of pretending quite undefined ideas are suitable for
the international public that browses to the Russ Paielli website and
that also has the misfortune of reading messages from this mailing
list. Of all your examples, GSFC is instantly detectable as being a
criteria that is TOTALLY undefined. Here the definitions that Mike, seems
to like to not read

http://russp.org/ElectionMethods.org/criteria.htm#GSFC

I note to Mike that Mr Schulze's intent is rather plain. He knows as
most subscribers do, that you evade the question and fail to make any
response. Yet the intent of the question is clear.


I shall ask a much narrower question than Mr Markus Schulze. This
question may be harder to answer, since I am requesting a Boolean
truth value rather than an existential (For All, There Exists) logic
expression.


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

QUESTION THAT IS ABLE TO PROVE THAT MIKE OSSIPOFF DOES NOT KNOW WHAT HIS
OFTEN WRITTEN ON GSFC DEFINITION IS:

Does this 4 candidate 7 paper 1 winner perfectly defined voting method
pass GSFC ?, as defined at the Russ Paielli website?:

Ossipoff, who can evaluate the contributions):

Demonstrate the details of a proof showing that GFSC passes or fails this
following preferential voting method:

At 13:39 03.10.00 +1300 Tuesday, Craig Carey wrote:
...
  |  VBCD =       A one winner election method
  |      A    a0
  |      AB   ab
  |      AC   ac
  |      AD   ad
  |      B    b
  |      C    c
  |      D    d
  |
    (B wins VBCD)
  |           = (a<b) . (a+c<b+d or a+c<2*b) . (a+d<b+c or a+d<2*b) .
  |            (Tacb or (2c<a+b or 2c<a+d)(c+d<a+b or b+c<a+d)) .
  |            (Tadb or (c+d<a+b or b+d<a+c)(2d<a+b or 2d<a+c))

    (C wins) and (D wins) are given by swapping B with C, and B with
    D, respectively, in the (B wins VBCD) equation.

    (A wins) = (b<a) & (c<a) & (d<a)

    Note 1: ab is a single variable and not a product of two variables.
    Note 2: a = a0+ab+ac+ad.
    Note 3: while that is not a method that would be used, it is a method
       that finds exactly 1 winner in all non-infinitesimal regions.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

The presenting of false information to this mailing list by
Mike Ofssipoff reduces the greatness and accomplishments of this mailing
list. It wsa Mr Steve Eppley who started this, by sending in a message
defining GSFC that was to a causal glance, mathematical. It has other
bad definitions too. It is not a message that is free of very bad ideas.
I presume people at this list have very little ability to follow bad ideas
of others. The Steve Eppley GSFC definition

http://www.egroups.com/message/election-methods-list/5014

Mr Ossipoff's page listing undefined rules:

http://russp.org/ElectionMethods.org/criteria.htm#GSFC

I would join another list that discussed methods that indeed were
preferential voting methods: under the principles of STV, if there were
one.


>Tideman(wv) meets GSFC & SDSC, and that Smith//PC doesn't. Any examples
>that comply with the example outlines that I posted for Smith//PC
>will be sufficient to show that Smith//PC fails GSFC & SDSC.

You were asked about all methods, and not asked about Tideman. Why not
just answer the question?. Mr Schulze did not request a pretence of a
response, and no that you gave one, I ask: why was no reply made?.

I ask and appeal to all members to consider not replying to anything that
Mr Ossipoff wrote. I suggest that they have a separate folder for messages
from Mike Ossipoff and remove the messages from their Election Methods
List folder. This can cover all that promote poorly defined ideas listed
at the Russ Paielli website.




>In general, any example that complies with the premise of a
>criterion, and in which the method being tested doesn't meet the
>requirement of the criterion will be sufficient to show that that
>method fails the criterion. Also, such an example is necessary to show

If you do not refer to the definition then you are not responding to
the request. Why not refer to your actual rules?. Is the online
representations of your GFSC undefined because of mistakes you put into
it, and so very wrong you will not quote it (in an unaltered but more
lucid form)?.

>that. In fact, that's what it takes to show that any method fails
>any criterion. For more detail, I refer you to the definitions of
>the criteria.

If you have the definitions in the form requested somewhere, i.e. in a
form where they use as input, all of the properties of a voting method,
then do tell us where to look. The question is: are your rules defined.
Your reply appears to be: refer to the rules. That is not a response
to the question which you appear to have not read closely:
Mr Markus Schukze used the words "a given election method".

You pages has many errors and the wording makes it probable that Mr
Shulze wish to avoid a seeing the falsity of the page reproduced as
a way to discard the questions. I.e. the false idea that papers are
voters. There are not: one voter can vote with 12 quite contrary papers.

I note to Mr Ossipoff the detail, that referring me or Mr Schulze to
somewhere else, is not a response to the questions.

Why is Mike Ossipoff telling people to go elsewhere, without giving
information on where to go to?. Can that be fixed or withdrawn?.

-----------------------------------------------------------------------
                     The first appearance of GSFC.

The GSFC definition seems to have been first sent to this mailing list
by Steve Eppley in Feb 2000. I quote it below.



http://www.egroups.com/message/election-methods-list/5014

At 12:37 23.Feb.00 -0800 Wednesday, Steve Eppley wrote:
...
 >I've been using a different criterion, which dispenses with
 >the "absolute majority" requirement, and so is more general:
 >
 >   Beatpath Criterion (BC)
...
 >   General "Strategy-Free" Criterion (GSFC)
 >   ----------------------------------------
 >   If no voter ranks any alternative ahead of a more preferred
 >   or equally preferred alternative, and an absolute majority
 >   rank at least one member of the sincere top cycle ahead of
 >   an alternative X not in the sincere top cycle, then X must
 >   not finish first.
 >
...

This rule is dumb.

The word "majority" could make the rule powerless (if it were somehow
defined).
The rule confuses voters with papers. I note to Steve that that is in
effect, an extremely hostile thing to do here.

What is an "equally preferred" alternative?. Neither Mr Ossipoff nor
Mr Eppley defined a real valued function that evaluated preferences.
This is not in any sense a small mistake in multiwinner elections, and
note that Mr Ossipoff still has refused to add a constraint on the
number of winners. I wrote to him privately. It is not so clear to
me that Mr Eppley has made the same mistake (since it followed from a
commenting on what could be 1-winner concepts).

The word "sincere" is a not-defined property. The word sincere seems to
be a code-word meaning "voter" rather than sincere. The idea of voter
is a it different from the idea of the will of democratic choosing, but
instead it is a receptacle for the motives and thoughts that Mike Ossipoff
asserts into it. I imagine it is something that absolutely no subscriber
agrees Mike is right to do.



Check this out:

At 00:55 07.10.00 +0000 Saturday, MIKE OSSIPOFF wrote:

>Markus quoted me:
>
>> > Why doesn't that work when A is replaced with {A1,A2,A3}? Because
>> > they're in a cycle of majority defeats. Their majority defeats could
>> > be greater than that of B. B's majority defeat could be weaker than
>> > any other defeat in the election, and B could win by having the weakest
>> > maximum defeat.
>> >
>> > Of course if B is in the sincere Smith set,
>
>Sorry--I meant "the voted Smith set".

There seems to be a belief that Mike has a facility and competence in
writing about voters. Of course, most of the time he sees ballot paper
counts, he has no knowledge of the voters or how many papers each did
cast, or whether there were voters.

Here is the latest Annotated Ada 95 Reference Manual. Ada 95 is the most
well defined language known:

http://www.ada-auth.org/~acats/arm.html  Ada 95 Ref Manual: latest version

At what data after Mike Ossipoff wrote his online definitions, did they
become able to be ignored by Mike?, e.g. when preparing comments for this
list?. Mr Russ Paielli doesn't understand the rule. Mike complained that
someone sent a message to Russ but not to Mike. I wrote at least 2 very
intelligent messages to Russ and he never replied. I have no evidence to
show that Russ Paielli can read writing that comments on even the simplest
ideas of preferential voting. Maybe Mr Ossipoff has to pay US$3/yr to keep
online the GSFC rule?.

--------------------------------------


I request under the rules of the list, answers to these questions
These 2 questions are easy. I ask Mike to list all the complex evasive
responses he thought of penning prior to deciding to let the list know (one
way or another) that neither of the two yes/no responses would be sent to
the mailing list. I note to Mike that the number of voters is reasonable
and in the dictionary sense, they votes seemed to be "sincere".

Does method H21 pass GSFC ?.
Does methods H21 pass SDSC ?.

Only the online definitions of Mike Ossipoff are to be regarded as the
official definitions. A corollary of this is that explanation from Mr
Ossipoff is not good enough.

---------------------------------------------------------------------------
Method [H21], devised to allow GSFC to be tested.
   A   a0
   AB  ab
   AC  ac  : a = a0+ab+ac, etc.
   B
   BC
   BA
   C
   CA
   CB

(A wins) = (b + 0.2*cb < 1.1*a + 0.1*ca)
(B wins) = (b + 0.2*cb > 1.1*a + 0.1*ca)
(C wins) = False

---------------------------------------------------------------------------

If you have some insights into the effect of altering the "1.1", and the
"0.2", then please post that in.

A mathematical response is the preferred form for the answers to my
requests.

I ask Mike Ossipoff to translate his online rules into existential logic
expressions. That seems to be what Mr Markus Schulze was hoping for. Any
such form will exhibit a parameter that represents the function that is
the method.

... this mailing list can't even get Mr Mike Ossipoff to count the number
of parameters to the functions that are rules. I give up.

I just don't agree that Mike is right in calling papers voters, then
asserting attributes about voters, and expecting me to share a common
belief that I comprehend the reality of how Mike can in truth assert
that properties of voters (like their FBC-likes-mores or their
SDSC-should-acts' or their GSFC-absence-of-false-acts') as exactly as
stated. The word sincerity means that the aspects are not about papers
but voters. It is as if 99% of what Mike writes is dedicated to framing
a belief that ballot paper counts should be ignored. I come and go, in
this mailing list, but the subscribers love to read what Mike has to
say.

Someone has to protect the list from leaking subscribers to other lists
that could have even lower academic standards. [or is that not true?.]




G. A. Craig Carey

http://www.ijs.co.nz/ifpp.htm  The Politicians and Polytopes mailing list.