Steve's vote for our ER sw recommendation

[Steve Eppley]

Here's my ranked ballot for our single-winner recommendation to the
ER list:

1 Smith-Condorcet
2 WB/Smith-Condorcet
3 plain Condorcet
4 WB/Condorcet
5 none of the other methods
6 the other methods

I reserve the right to change my ballot anytime.  There's plenty more 
for me to learn.

I considered trying to rate (weight) the methods on a -100 to +100
scale, but my error bars would swamp the differences in the numbers:
my top four choices cluster very close together.  (Maybe I should 
rank the four equally 1st?)

I'm unsure about my rankings.  I'd appreciate feedback from the
public on weighted ballots vs ranked ballots, and whether they'd be
turned off by a complex tally algorithm even if reassured by examples
and by debate among experts that the algorithm is sound.

I don't believe that the complexity of the definitions of Condorcet
and Smith//Condorcet will be as important a negative factor in
getting public approval as the technical flaws of the other methods
would be in getting public approval of the other methods.

---Steve     (Steve Eppley    seppley at alumni.caltech.edu)