[EM] Voting reform statement - new draft, please give opinions

[robert bristow-johnson]

On 8/19/11 12:22 PM, Jameson Quinn wrote:
>
>     These are better than Plurality:
>
> Plurality has big problems. Any of these would solve most:
>
>         * Approval
>         * Bucklin
>
> / (Majority Judgment)
>
>         * Condorcet
>         * Range
>         * SODA
>       Approval is ideal as a first step in voting reform.
>
>

alright.  may i offer some perspective here?  this is a little more of 
what i've been thinking in the past couple of days:  what have we 
learned from (or should be learning from) the ongoing IRV experiment and 
Fairvote and such?  i remember hearing or reading a little concession 
from some knowledgeable IRV advocates (will not mention names) that 
Condorcet was better than IRV and Burlington 2009 sorta epitomized how 
and why it's better.  but they said the same thing, that IRV was a good 
"first step".  something like "let's get Ranked Choice Voting in first 
and optimize later."

now here's the problem: if a voting reform has some anomalous result or 
doesn't exactly deliver on its promise, the reactionary opponents of 
that reform will be happy to point that out and this makes the rest of 
the electorate suspicious or skeptical of the next reform effort when it 
comes up.  in Burlington 2005, we adopted IRV with 65% mandate and it 
was narrowly repealed in 2010 with less than 52%.

the problem is that if Approval is adopted and later disliked (it might 
not be an anomalous result, but might be that voters tire of having to 
decide whether or not to approve of their 2nd-choice candidate) they 
will be disinterested in any second step.  there is a finite number of 
times that voters are willing to try something new.  (see 
http://vtdigger.org/2010/03/05/vermonters-should-consign-irv-to-the-ash-heap-of-electoral-history/ 
)  i'm afraid that we'll have to wait for another generation (and i hope 
that we continue to have 3 or more competitive parties in Vermont) to 
revisit the issue of a better method than FPTP or TTR.  the anti-IRV 
crowd likes to think that the traditional vote-for-one ballot is handed 
down by God and the other people that voted against IRV (and had voted 
*for* IRV in 2005) just didn't like how it turned out and will be more 
skeptical of the next reform than they were in 2005.

so besides Aspen CO, Cary NC, Pierce Co WA, maybe Ann Arbor MI (can't 
remember who else, Burlington VT, of course), other towns, perhaps 
Cambridge or Mpls/St.P. or SF will also have a problem and revisit the 
IRV issue, and with that, other ranked-choice systems like Condorcet. 
other methods of voting reform get stained (from the POV of 
traditionalists, and this seems to be close to religion for them) by a 
failure of one method.

so, i think we should learn from FairVote's error(s).  and i think we 
really should be careful and aim for the *final* step, rather than the 
first step.

> Gerrymandering and safe seats are also problems. Proportional
> representation would solve it. There are many good options, including
> some with geographical aspects, but closed party list is not good.

i'm definitely in favor of mapping algorithms that, given a few 
parameters from humans, sorta "blindly" draw legislative districts 
according to mandated rules (like equal population districts, statewide 
proportionality regarding groups and class of voter, community centered, 
and competitiveness - not to ensure someone's safe seat).

-- 

r b-j rbj at audioimagination.com

"Imagination is more important than knowledge."