On 5/21/2017 7:10 PM, Armando wrote:

Thank you for letting us know that our answers were useful.

Thank you for spreading useful voting knowledge.

Based on your current thinking it's clear -- to me -- that you
understand what we wrote.

You, and we, are exploring frontier territory, so there's not a lot of
formal writing about "proportional multi-winner Condorcet" methods
beyond what we've told you about.

If you, or anyone, has specific questions about what I wrote regarding
this topic in "Ending The Hidden Unfairness In U.S. Elections", just ask.

Richard Fobes

Dear Richard Forbes,
a special thank to you for your ready answers. Actually I already met your website years ago, and I found it one of the most fascinating tool for internet voting.
I’m not an expert, so I’m not sure to have well understood, correct me if I mistake:
your systems seems to fit specially context like USA, where traditionally the relationship within the constituencies and their representatives is harder than in non-anglosaxon countries. It seems that they are the best solution for countries who don’t prioritize proportionality, who look for local representation, but however want some pluralism.
Since in my view and for my needs my priorities are perfect proportionality and political representation (instead of local representation), so nation-wide district are not a problem, I’m correct in believing that your systems don’t fit to me more than a proportional condorcet?

Thank you again

Actually VoteFair ranking would work quite well for electing members to
the Norwegian parliament, or any other parliament/legislature where
highly proportional results are important.  Doing so simply involves
choosing the right numbers.

Specifically, for the case of Norway where there are 19 districts
("constituencies"), and which has a unicameral (rather than bicameral)
parliament, I suggest:

Category 1: Each district ("constituency") would elect 4 members using
VoteFair representation ranking. Typically the two dominant parties in
that district would win 2 seats each.  (If a nation had lots of
districts or states, then only two members would be elected this way.)

Category 2: Each district would also elect an additional 1 or 2 members
using VoteFair "districtwide" partial-proportional ranking (using only
the ballots in that district).  Typically these seats would be won by
members of that district's non-dominant parties.

Category 3: An additional 30 (or 20 or 40 or whatever) seats would be
elected using VoteFair "nationwide" partial-proportional ranking (using
all the ballots across the nation).  Although this method includes the
word "partial" in its name, the method becomes fully proportional simply
by choosing to use this method for lots of seats.  These seats would be
won by whichever parties did not win "enough" seats at the district level.

Yes, as explained in "Ending The Hidden Unfairness In U.S. Elections"
for use in the United States there would be only a few (of what I'm
calling here) Category 3 seats.  That's because in the U.S.
proportionality by state is as important as proportionality by political
party.  That's also why I included the word "partial" in the name
VoteFair partial-proportional ranking, namely because in the U.S. case
it does not cover enough seats to ensure full party-based proportionality.

Getting back to the case of Norway, or elsewhere where party
proportionality is more important than location proportionality, simply
filling lots of nationwide seats using VoteFair partial-proportional
ranking achieves a high degree of proportionality.

As part of this suggestion, VoteFair party ranking would also be used.
It eliminates the need to set a quota for the minimum percentage of
votes needed (by a very minor party) to win a single seat.  Of course
only a voter's top-ranked party would be used for the proportionality
calculations.  (The full ranking would be used to limit how many
candidates each party can offer in that district's election.)

Most importantly, VoteFair ranking elects the entire parliament without
allowing political-party leaders to choose which candidates fill that
party's seats.  Why is this important?  It allows the voters to control
the parties.  This greatly reduces the corruption that inevitably occurs
when insiders have control of which candidates win that party's seats.

As a reminder, VoteFair ranking eliminates the need to ask candidates to
choose which kind of seat they are competing for.  Every candidate is
competing for the Category 1 seats, but they have an opportunity to win
a Category 2 seat in their district if their party is medium-popular in
that district.  And every candidate who fails to win a Category 1 or 2
seat has another chance to win a Category 3 seat if the candidate is
popular in his/her district and is in a party that failed to win enough
of the Category 1 and 2 seats.

The situation in the United States is quite complex, which is why it was
challenging to design VoteFair ranking to meet the needs here.  In
contrast, smaller nations -- that have only a few big metropolitan areas
(instead of hundreds as in the U.S.) -- can use a simplified version of
VoteFair ranking.

Thank you for asking for clarification instead of just assuming that
VoteFair ranking is only suitable for use in the U.S.

And thank you for asking questions because I'm sure that lots of other
readers have the same (or equivalent) questions.

Richard Fobes

On 5/22/2017 5:44 PM, Armando wrote:

On 05/22/2017 08:18 PM, VoteFair wrote:

It's not even all that clear what "proportional multi-winner Condorcet"
means, independent of actual implementations. The lower bar is "reduces
to Condorcet when there's only one winner", but how could we generalize
Condorcet beyond that point? Hard to tell.

Yes, "proportional multi-winner Condorcet" has no clear, unambiguous
meaning beyond the criteria for identifying the winner of the first seat.

In the case of the full VoteFair ranking system, VoteFair popularity
ranking -- which is mathematically equivalent to the Condorcet-Kemeny
method -- is used to identify the most popular candidate (or party)
after calculating the "remaining" influence of each ballot.  It produces
results that are as proportional as desired.  Consequently it qualifies
as a "proportional multi-winner Condorcet" method.

After I published the VoteFair ranking system, Markus Schulze published
his "Schulze STV" method, which also fits within the "proportional
multi-winner Condorcet" category.

Are there any other such methods?

Richard Fobes

On 6/2/2017 12:23 AM, Kristofer Munsterhjelm wrote:

Yes, it is not easy to say which methods should fall in the "proportional multi-winner Condorcet" category. I also note that even if it would be a requirement that the first seat shall go to the Condorcet winner, if one exists, it is quite possible that the Condorcet winner would not be elected if there are two seats. (e.g. when there are two big parties, left and right, and one small centrist party with a Condorcet winner)

Juho

On 06/04/2017 10:07 AM, Juho Laatu wrote:

The LCR example is a concrete example that giving the first seat to the
CW makes the method fail Droop proportionality. E.g.

43: L>C>R
41: R>C>L
6: C>L>R

number of voters = 90, Droop quota for two seats = 30, so both L and R
should be elected, but C is the CW.

For larger assemblies, it might still be a good idea to give a few of
the seats to winners chosen by a multiwinner method with few seats, or a
single-winner method. Doing so would make centrists the kingmakers in a
kingmaker scenario, rather than minor parties on one wing.