Mike,
You wrote:
What's important about strategy is the mininimization of _need_ for
strategy. That's how strategy is important.
In wv or MMPO, if you truncate to deter offensive strategy, the only
importance of that deterrence is that it keeps you from needing more
drastic defensive strategy (as you would need in Margins, for instance).
I notice that Jobst, Kevin, and someone whose initials I didn't
recognize, strongly disagree with having AERLO as an option.
In reference to strategy, you are always throwing around these IMO far
too strong and emotive words "need" and "forced". Why not simply refer
to "incentives"?
Regarding AERLO, I agree with Jobst in being opposed to *explicit*
strategy devices. But I'm not in principle opposed to allowing the
voters to enter an approval cutoff in their
rankings which might, for "defensive strategists", perform a similar
function. But if there is such an option, it should be something that
makes sense to voters innocent of strategy and
it should preferably have some up-front function so that voters see it
as important. It shouldn't be the case that "this is just in case
there's a top cycle" and then there isn't one for a
very long time so most of the voters either don't bother or don't take
it seriously.
Since Mike has stated that the purpose of AERLO is for
"acceptable/unacceptable" voters to enter below the acceptable set of
candidates, I suppose there's no reason not to
call it an "approval cutoff". But I see a problem in justifying how it
works. If we say "This is so that if some voters don't like the result
of the (first,'provisional') election, then they
can change their vote (for the second, 'final' election)" then this just
prompts the natural question "How is that fair? If some voters can
change their votes after the 'first election',
why can't other voters change their votes after the 'second election'?
Why stop at only two elections?". I don't know any good answer to that.
Chris Benham
----
Election-methods mailing list - see http://electorama.com/em for list info
