[EM] Alan Greenspan, Bucklin, and IRVing:

[Bart Ingles]

I Like IRVing quotes Rob Ritchie:
>
> This system is the "Bucklin" system that has an
> interesting history in American elections. In public
> elections, the problem has been that many voters stop
> ranking  candidates, as they don't want a second-choice
> to count against the chances of their first choice.
> 
> [Don here: It is good if the voters of the larger factions only make one
> choice when voting in a Bucklin election.  This will cause Bucklin and any
> of the other weird methods to become more like Instant Runoff Voting.]

True enough.  Many or most major party supporters will decline to rank a
second choice even under IRV, 
 (1) partly because there would be no reason to bother with a second
choice (since their second and lower choices will never be counted
anyway), and
 (2) partly to make it clear to supporters of any potential third-party
spoilers that their best strategy will be to rank one of the top two
candidates first, if there is any chance that the third party candidate
will be in a close contest with one of the frontrunners.  In other
words, Gore supporters should vote only for Gore, in order to make it
clear to Nader supporters that Nader has no chance to win the final
round even if he gets that far.

But at least Donald now acknowledges that voters of the largest factions
_should_ only make one choice, at least IRV, Approval, Bucklin, and most
of the other commonly discussed methods.  His previous examples seemed
to claim the opposite.