[EM] Candidate vote transfer p.r. method, 16 Jan 2000
Craig Carey
research at ijs.co.nz
Sun Jan 23 00:08:23 PST 2000
Theorem 1.
For those subscribers that haven't figured this out, Demorep1 (a Russian
female former FAPSI signals agent suddenly forced out later hospitalised
in a Cuban psychiatric hospital), has
"not finished testing"
the "DEMREP1 17 Jan 2000 low tech method" quoted below.
Proof of theorem 1:
Here's an extract of the text of the document:
>[(f) If a candidate has a number of votes more than the total votes for all
>candidates divided by N (rounded up if a fraction), then the excess votes
>over such ratio shall be transferred to the highest remaining candidate on
>his/her list (starting with the candidate with the highest excess votes).]
>
>(g) The candidate with the lowest number of votes shall lose and have his/her
>votes transferred to the highest remaining candidate on his/her list.
>
>(h) The preceding [2 steps] [step] shall be repeated until there are N
>remaining candidates (who shall be elected).
Suppose that there is exactly ONE candidate and that single candidate was
the best man to usher in a new age of joy and reward to the membersof of
the leglislative chamber. The method is strictly applied....
Step (f) -- Surely N equals 1 if there is an analogy with the Droop quota
for transferring supluses of winners in STV. If N equals 1, this step
does nothing, since DEMOREP1 used a ">", rather than a ">=", and since
there is no rounding.
The number would not be a Hare quota since surely DEMOREP1 would have
written (N+1).
The number N could be the number of seats that remain to be filled at
each stage, or it could be the total number of winners that need to be
found.
The lack of definition alone ought be enough to allow the method to be
fully discarded by competent legislative assemblies using FPTP,
especially given that DEMOREP1 wrote:
"it would generally require computer voting." (computers can't
cope with vagueness unless doing symbolic computations or statistical
computations, or whatever.
Step (g) -- DEMOREP1 elminates the last candidate before having picked
one. This is not all a new feature to [DEMOCRAP, substitute here]
methods. (I have pointed this out before to the list but perhaps you
did not realize I was commenting on one of your own methods).
Step (h) -- N is still 1 isn't it?. I suppose so,control transfers to
clause f of the document.
Step f -- Even if the votes were transferred into bin, the inequality
is still of a nature that causes clause f to do nothing.
Step g -- There are no candidates so step g does nothing
Step h -- For the first time, we can see that the method loops infinitely.
Corollary 1: Demorep1 at aol.com is informing the list of untested methods.
Dear DEMOREP1, could you please make all your messages have lines that do
not have spaces betweeen each line. I request no double spacing. If the
handling of double spacing is not under your control, could take control.
My motice is that I like to see twice as much at a single glance.
The Canadian list's readers might have the concern about double spaced
lines. A vote is not I don't suggest a vote on this, but the re instead
At 11:43 17.01.00 , DEMOREP1 at aol.com wrote:
>Candidate vote transfer p.r. method, 16 Jan 2000
>by Demorep1 at aol.com
>
>(a) The following shall apply in the election of the N members of legislative
>bodies.
>
>(b) The number of districts (if any) shall be N divided by [5] (rounded up if
>a fraction).
>
>(c) Each district (if any) shall be [formed at least [252] days before the
>election], square, contiguous and contain the same number of electors (plus
>or minus [1] percent) who voted at the last regular general election for the
>legislative body, as nearly as possible.
>
>(d) Each candidate shall submit a public vote transfer rank order list [of
>all the other candidates] at least [7] days before the election day.
>
>(e) Each elector may vote for 1 candidate.
>
>[(f) If a candidate has a number of votes more than the total votes for all
>candidates divided by N (rounded up if a fraction), then the excess votes
>over such ratio shall be transferred to the highest remaining candidate on
>his/her list (starting with the candidate with the highest excess votes).]
>
>(g) The candidate with the lowest number of votes shall lose and have his/her
>votes transferred to the highest remaining candidate on his/her list.
>
>(h) The preceding [2 steps] [step] shall be repeated until there are N
>remaining candidates (who shall be elected).
>
>[(i) Each legislator shall have a voting power in the legislative body (and
>its committees) equal to the final number of votes that he/she finally
>receives).]
>---
>Use (i) to get more exact proxy representation (especially if the total
>votes/N ratio is low and especially if (f) is NOT used).
>
>A higher tech method would have the electors use number votes (1, 2, etc.)
>for their choices of candidates and vote transfers (with possibly a YES/NO
>vote on the candidates for tiebreaker purposes) but it would generally
>require computer voting.
-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-
At 12:58 08.October.99 , DEMOREP1 at aol.com wrote:
>A simple p.r method for low tech/ no tech areas (and possibly for so-called
>high tech areas) ----
>
>A. Each voter may vote [X vote] for any legislative body candidate.
...
>C. The N candidates receiving the highest number of votes shall be elected.
>
>[Option- the candidate with the lowest number of votes shall repeatedly lose
>until there are only N candidates remaining who shall be elected.]
>
...
>Almost anything must be done to get indirect Democracy into the U.S.A,
>Canada, the U.K. House of Commons, India, etc. before lunatic/ extremist
>indirect minority rule gerrymander politician- legislators cause major local,
...
You can see that in about 8 October 1999, DEMOREP1 was using 'equality'
in the candidate discarding step which later became a key component of
the infinite loop. The next message is where I point out that a method
for state legislatures should be able to elect the right number of
winners when the number of winners equals the number of candidates.
-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-
At 14:13 29.12.99 , Craig Carey wrote:
>At 15:53 24.12.99 , DEMOREP1 at aol.com wrote:
>
>This method is untested or badly designed.
>
>Suppose there is only 1 candidate and only 1 member needs to be elected
> in the region (R = 1).
>
>>A low tech (repeat, low tech) legislative body p.r. election method.
>>
>>If p.r. is to be used to elect a state legislature ...
...
>>At Large
>>
>>(a) Each elector may vote for 1 candidate.
>>(b) Before the election starts, each candidate may submit a rank order list
>>of the other candidates.
>>(c) The candidate with the lowest number of votes shall lose and shall have
>>his/her votes transferred to the remaining unelected candidate highest on
>>his/her rank order list (if any).
>>[(d) Any candidate having more than the total votes divided by R shall be
>>elected and ...
>
...
-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-
In your 4 December 1998 example, you seemed to understand the problem:
At 12:53 04.December.98 , DEMOREP1 at aol.com wrote:
>Mr. Lytton wrote in part--
...
>Proxy p.r. constitutional language-
>
>(a) An Elector may vote for 1 or more legislative candidates on the ballots in
>a district (plus not more than [2] write-in votes) by voting "1", "2" and so
>forth for his or her first, second and so forth choices. (b) If there are more
>than [5] candidates (or remaining candidates) in the district, then the
>candidate having the lowest number of votes shall be a losing candidate. (c)
>Each vote for a losing candidate shall be transferred to the Elector's next
>choice (if any) who is a remaining candidate in the district. (d) The 2 prior
>steps shall be repeated until there are [5] remaining candidates in the
>district who shall be elected. (e) A lottery shall be held if tie votes occur
>in any step. (f) Each member of a legislative body (or his or her replacement)
>shall have a voting power in the legislative body and its committees, in
>person or by written proxy, equal to the votes that the member finally
>receives in the Election. (g) Example-
...
>MMP results in the party votes/party seats ratios being unequal between the
>parties (especially for a low number of seats won). See
>http://www.election.govt.nz/summary.html
>(New Zealand 1996 MMP election data as of Dec. 1997)
>
>Close is good enough in horseshoe games but not public elections involving the
>fate of everyone on the planet.
>
Are you going to start testing your methods?
-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-
This is the earliest appearance of DEMOREP1's idea of making a candidate
that ought win, a loser.
That is done in step "7", below (e.g. no. winners = no. candidates = 1)
At 09:33 16.May.99 , DEMOREP1 at aol.com wrote:
>A one district proportional representation method for a legislative body (for
>folks not wanting districts or surplus fractional vote transfers) ----
>
>1. Candidates get listed in an election pamplet by a nominating petition only
>(no primaries- no party lists).
>
>2. Each candidate would have his/ her rank order list of other candidates
>(who would receive his/ her surplus votes for step 6 below) put in the
>pamphlet.
>
>3. On the election ballots a voter would write in (or type in on a computer
>screen) each candidate's pamphlet number.
>
>4. A voter could number rank (1, 2, etc.) his/ her choices.
>
>5. Each candidate getting a Droop (? Hare) Quota of votes would be elected.
>
>6. Each candidate getting more than a Hare (? Droop) Quota of votes would
>transfer the surplus votes (using the above rank order list for remaining
>candidates) (i.e. no fractional vote transfers of surplus votes).
>
>7. The candidate with the lowest number of votes would lose and have his/ her
>votes transferred according to each voter's next choice of remaining
>candidates. Presumably the voters of each party would vote for the other
>candidates of such party.
>
>8. Repeat until all seats are filled.
>
>9. If exact results are needed (especially for legislative bodies with a
...
>Very large legislative bodies (i.e. 435 seats in U.S. House of
>Representatives, etc.) are mob rule scenes such that perhaps the number of
>seats can be greatly reduced.
Irrespective of how Senators mob together, they like counting of results
to complete.
-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-
This is an example of about a single e-mail reader window of
double spaced lines (768 pixels high):
From: DEMOREP1 at aol.com
Message-ID: <f3.4ce63d.25ba1914 at aol.com>
Date: Fri, 21 Jan 2000 15:18:28 EST
...
FOR IMMEDIATE RELEASE
Contact: Paul Hrabal, CEO
(916) 797-7443 or paul at govote.com
> E-Voting Starts Monday
> First Binding Internet Elections In History
> Internet Voting Lets Alaskan Voters in Remote Region Participate in Straw
> Poll
> Juneau, Alaska - January 20, 2000. The first binding internet election in
> history will take place January 24, 2000. With sponsorship by a web
> company named GoVote.com and new technology making it possible, over three
> thousand registered Republicans in districts 36, 37 and 38 - a region
> accessed mostly by seaplane, snowmobile or dog sled -- will be able to
> cast their ballot.
> The Monday election determines who will go to the Republican National
> Convention on behalf of Alaskan Republican voters, and, consequently, who
> Alaskan voters nominate for the Republican presidential candidacy.
> GoVote.com's sponsorship of the Alaska poll is part of the company's push
> to make politics more accessible. Tens of thousands of people per day come
> to GoVote.com for answers and updates about the presidential candidates.
----------------
I hint to the list administrator or else the list that the name
"DEMOREP1 at aol.com" be blocked from posting to the mailing list.
I suppose DEMOREP2 at aol.com would be fine with me. I really doubt that
anonymity is necessary. Is this list is to be a corner for dunces, that
need or receive anonymity?.
Arguments for that can easily be made for sure, but readers may want to
restrict those arguments so that they only come from DEMOREP1 herself
(I guess).
I suppose Demorep1 could be a former Russian female retired FAPSI signals
agent that monitored US advancements on political theory and how to
reform the Duma (ever a threat to FAPSI equivalent of the NSA), and is
now, in yr 2000, posting from an old Russian clone 486 with assistance
from the orderlies (what are they like over there?).
I welcome corrections to this message (post to the list, if you wish).
Mr G. A. Craig Carey, research at ijs.co.nz
Auckland, New Zealand.
Snooz Metasearch: <http://www.ijs.co.nz/info/snooz.htm>
More information about the Election-Methods
mailing list