[Election-Methods] Partisan Politics + a method proposal

[Dave Ketchum]

  On Sun, 01 Jun 2008 19:05:14 -0400 Fred Gohlke wrote:
> Good Afternoon, Dave
> 
> I did a very poor job of describing my intentions when I started the 
> outline based on Juho's comments.  It struck me it would be a good idea 
> to encourage a joint effort to create a sound electoral method.  Several 
> ideas are regularly discussed on [Election-Methods] and, although I'm 
> not intimately familiar with most of them, they seem to favor fixed 
> approaches.  Since I don't think any have gained general approval, I 
> thought it might be worthwhile to seek a more flexible approach in the 
> hope of combining the best elements of all of them.
> 
> The statements in the outline are not intended (or expected) to remain. 
>  They should be replaced by more definitive statements as various people 
> challenge this or that assertion and help mold a clear, sound method of 
> electing our public officials.  My role in the process is that of a 
> clerk.  I fully intend to voice my opinion, but the outline must be what 
> others want it to be, not my impression of what they want it to be.
> 
> There is the obvious difficulty of properly expressing the views of 
> others, so, my preference is that contributions be written to replace 
> statements in the outline.  I am concerned about the handling of 
> divergent opinions, but will cross that bridge when I come to it.
> 
> Ideally, the outline would be in a fixed location where it could be 
> maintained, but I've no idea of the practicality of that notion.  Unless 
> and until we can made such an arrangement, I will append the outline, in 
> it's then-current form, to some of my posts.  I'm not certain I'll be 
> available to continue the process, but feel confident that, if the idea 
> has merit, someone will find a way to make it work.
> 
> I'm writing all this explanation to you because I'm hoping you will 
> restate some of your observations in a way I can copy into the outline. 
>  I'd much rather not try to restate your intent.  Here are a couple of 
> the comments you've made that I don't know how to handle in their 
> current form:
> 
> re: "Depends on race - even one such page would be overkill for some 
> local races."
> 
> I suspect our best course would be to select one race (you've already 
> mentioned 'governor') and build up a method around that.  Once the 
> method for one race is clearly defined, it should be straightforward to 
> modify it for other races.
> 
When I say "race" below it will be because requirements are stiffer for 
governor than for village trustee - and even for governor of a large state 
than for a  small state.  It varies based on importance of office, size of 
electorate, and even on experience with the electorate.

Examples may be offered for starting thoughts such as for governor of a 
medium state.
> 
> 
> re: "Degrees are not always the best evidence of ability."
> 
> How should the requirement be stated?
> 
race - degrees are sometimes important, but understanding of task is 
sometimes more important.
> 
> 
> re: "Internet web pages are, more and more, the best choice."
> 
> Can you make this an assertion I can include?
> 
Idnhahr - I do not have a handy reference.
> 
> 
> re: "There are STRONG arguments against alphabetical order - particular 
> list positions attract voters."
> 
> The arguments should be presented in a way they can be examined.
> 
Idnhahr - but some claim different orders should be used on different 
ballots to even out benefits of being first.
> 
> 
> re: "Equal approval rating should be permitted (IRV chokes on such, but 
> IRV should be rejected for other reasons)."
> 
> This needs exposition, examination and, perhaps, challenge.
> 
When ranking candidates, voters can desire to express equal liking for two 
or more.

Condorcet has no problem with honoring such ranking.

If IRV is presented with equal rankings of which one is to be discarded as 
least liked, fair treatment of remaining such candidates is a challenge.
> 
> 
> re: "Ranking is appropriate, but do it more like the weighting described 
> above." ... and ... "Anyway, while bullet voting should be permitted, 
> there should not be more than one other method, such as ranking or 
> weighting."
> 
> These need itemization and exposition so they can be enhanced.
> 
Context was permitting voters to expresses thoughts fitting incompatible 
election methods.

While a new method might be devised using some combinations of such, just 
letting voters express such without preparing for counting is destructive.

For example, ranking for IRV is incompatible with rating for Range.

I do believe bullet voting fits with most anything else - though some, 
such as IRV, may not permit it.
> 
> 
> re: "If IRV tempts, join me on Condorcet, which uses the same ballot but 
> does not have the same failure.  As an example A is popular below (and 
> Condorcet would see A's popularity), but IRV would not elect A without 
> more first-place votes:
> 
> 28 B>A>C
> 25 D>A>E
> 24 F>A>G
> 23 A>B>C
> 
> This may be profound but it can not stand without information to support 
> the assertion.  I lack the knowledge to flesh it out.
> 
IRV would discard least liked in order: A, FAG, DAE, and declare B winner.

Condorcet will discard in order: FGDE, and see A deserving.

Try, instead, 28 A>B>C and 23 B>A>C - could happen on a different day with 
the same voters:
      Condorcet will see no difference.
      IRV will now see A as deserving.
> 
> 
> re: "You describe Range rating here.  Ranking is a method worth mentioning."
> 
> At the risk of showing my ignorance, I wasn't aware there is a 
> difference ... or why it might be important.
> 
Ranking A>B>C in IRV or Condorcet calls A top, B middle, and C bottom, but 
says nothing about how much this voter likes A better than C.

Rating A=9, B=2, and C=1 says this voter would agree as to ranking, but 
likes A MUCH better than the disliked other two - and Range, based on 
ratings, will count these numbers - would take 7 voters rating A=4 and B=5 
to make B tie A after above single vote.
> 
> 
> re: "... should there be a (length of) residency requirement?
> 
> Sure."
> 
> What should it be?
> 
race.
> 
> 
> re:  "Yes, population counts.  Experience also counts - if unreasonably 
> few or many candidates happen often, adjust."
> 
> How can this be presented so those reading the outline can agree or 
> provide additional insights?
> 
race.  History can also help:
      If there are too many candidates, make it harder next year.
      If too few, go for easier.
> 
> 
> re: "Candidate lists NEED establishment X days before election to allow 
> planning and thinking."
> 
> We need to say how many days so others can provide reasons why the 
> number should be changed.
> 
race.  Tend to want more campaign time for governor.
> 
> 
> re: "Then unplanned events can create need for changes (e.g., candidate 
> dies)."
> 
> To the maximum practical extent, we should identify the possible 
> unplanned events and a manner of dealing with each.
> 
NOT LIKELY - context was write-ins, for which each voter decides need for 
write-ins.
> 
> 
> re: "My words about a write-in method were perhaps too detailed to 
> bother with here."
> 
> It is more likely the details need to be listed so they can be examined 
> and challenged (if appropriate)
> 
Context was optimizing this part of counting - which we had best ignore here.
> 
> 
> re: "I want a set of nominators, which really attends to this problem, 
> though whether the count should be over or under 100 is controlled by 
> other needs."
> 
> How would we create a set of nominators (for governor?)?
> 
We do not - those wanting a particular individual as candidate see to that.

Concerned with quantity of candidates, we make the task easier or harder 
to attend to that goal.
> 
> 
> re: "(a nominating mechanism or an election mechanism) Actually they can 
> get mixed."
> 
> Perhaps, but our outline must have clarity.
> 
> 
> 
> I hope you don't consider it an imposition for me to ask you to revisit 
> this matter.  Indeed, if you think the whole idea silly or a waste of 
> time, I'll drop it.
> 
> Fred
-- 
  davek at clarityconnect.com    people.clarityconnect.com/webpages3/davek
  Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
            Do to no one what you would not want done to you.
                  If you want peace, work for justice.