[EM] Second Opinion please

[Blake Cretney]

Donald E Davison wrote:

> Don's Web Site Text:
>      There are other methods proposed for electing single-seat positions -
> other than Choice Run-off. Three of the proposed methods are the A-B-C
> methods, Approval Voting, Borda Count, and Condorcet.
>      In these three methods, as in Choice Voting, the voter is allowed to
> select a number of choices. And, also as in Choice Voting, if there is a
> majority candidate on the count of the first choices, there is a winner -
> the election is over - no math of any method is needed.
> 
> Blake wrote:
>      That isn't true of approval, as I define it. There may be multiple
> candidates with a majority approval. There is no way to tell which, if
> any, would have a majority of first-choice votes.
> 
> Donald writes:
>      I am willing to be corrected, but I am not yet willing to accept your
> defination as correct. You seem to be defining Plurality-at-Large.

At the risk of inciting further outrage, I would define
Plurality-at-large as the method where each voter may vote for, or
withhold a vote from, each candidate, but may only vote for a maximum
number of candidates equal to the number of positions to be filled. 
This method ensures that if people vote along party lines, the
plurality party will win.  In an approval-at-large, you would be able
to vote for as many candidates as you like.  This would have effects
similar to single-winner approval.

>      Of all the persons, other than you, that have written on Approval
> Voting(AV), they have all been more than willing to have AV step aside if
> there is a candidate with fifty percent plus one or more on the count of
> the first choices.

In the hopes of suggesting that it isn't unanimous, I submit these
pages:

http://bcn.boulder.co.us/government/approvalvote/center.html
http://renoir.vill.edu/faculty/bartlow/html/mat1220/approval.html

BTW, in my definition, as well as theirs, there is no way to indicate
a first choice on an approval ballot.  There is no "count of the first
choices."

>      One of the flaws of AV is that it will not always elect the majority
> winner of the first choices. 

What?  Aren't you saying the exact opposite?  Either AV is the way I
describe it, and has this flaw, or it is the way you describe it and
doesn't.  

> If I read you correctly, you are taking the
> position that this is the way it should be. If so, you certainly are
> daring. The supporters of a majority winner may want to "hang your ass" if
> you tell them their candidate is a loser.

I have no position on how approval voting "should be."  I have
attempted to define it as most others do.  You seem to draw a
distinction between how AV is and how it "should be."  I think if we
are going to describe AV, it makes more sense to describe it as it is.

>      You should rethink this majority postion of yours.

Later on in the same email, I state that I do not advocate Approval
voting.  It seems strange then to assume that my definition of it
would encapsulate my beliefs about the desirability of various
properties of voting methods.

---
Blake Cretney
See the EM Resource:  http://www.fortunecity.com/meltingpot/harrow/124