[Election-Methods] a story for christmas

[rob brown]

On Dec 26, 2007 2:11 PM, Jan Kok <jan.kok.5y at gmail.com> wrote:

(sorry i forgot to reply to your other points)

Your railing against RV is like lobbying against the unsecured
> vestibule idea.


I don't see any sort of analogy with the unsecured vestibule idea.  Maybe
you can have small groups of trusted people....your friends and family and
neighbors....that can combine their vote via range voting, then the combined
vote gets applied with maximum strategy?

Are you suggesting that?  I have no problem with it in theory, in practice
it sounds ridiculously complex and impractical.

If that is not what you are proposing, then the vestibule concept is
meaningless and irrelevant to elections with large numbers of people that
don't know and trust each other.


> Why are you so keen on FORBIDDING people from casting
> weak votes / making some of their property available for others to take?


Because the design actually encourages them to do so and makes them feel
dishonest if they do otherwise (this is an opinion, but from someone who
designs user interfaces for a living and therefore has some experience in
how a UI is interpreted).  You've gone to significant extra effort to allow
people to, in your words:

weaken your vote (let others "take advantage" of you)

and that is the only differentiator of Range from another method
(Approval).  And I don't like Approval either, since it's only
differentiator from Condorcet and DSV is that it inserts randomness into it
(due to people's inability to perfectly guess how others will vote).

I'm sorry if my locksmith story didn't map one to one on that particular
issue.  The main point of the story might be that a system that punishes
those that are honest, tends to promote dishonesty (especially in the long
term).

Regardless, the story wasn't intended to be a direct analogy (never claimed
it was), it was simply a way of illustrating the absurdity of several of the
Range Voting people's arguments.

Such as:

1) since everyone is allowed to defect if they choose, the system is "fair"
2) summing short term tangible utilities is a good indicator of long term
utility
3) simulations based on the short term are meaningful for the long term
4) you can form valid mathematical proofs that have variables that are
determined by psychology,
 and of course my favorite....
5) eusocial animals are good indicators of how humans will behave in
situations that involve strategy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20071226/741d35c4/attachment-0003.htm>