Re: Robonson wins; Debian v monotonicity (was Re: Some analysis of DPL 2003 results
On Mon, Apr 21, 2003 at 09:46:06PM +1200, Craig Carey wrote:
> This is a lengthy argument against the current Debian problem of
> wrongly rejecting Mr Branden Robinson who would almost certainly be
> the winner if the method of the last election was maximally proportional
> (and passing P2) and monotonic. I.e. the method then is almost the
> smallest adjustment that 'debugs' the unfair Alternative Vote method.
"argument against .. Debian problem". By your argument Debian has no
I'm still trying to decide whether the first two lines above can be read
two ways or not. I'll tentatively congratulate you on a well constructed
troll, but rise to the bait anyway. :)
This DPL election will probably make an interesting case for the election
reform movement: where a strict IRV vote (discarding some valid Condorcet
votes in favor of only linearly-ranked votes) would have selected one
candidate (B) when the majority would have preferred another (M). Having
three strong candidates, and no clear linear relation between them all is
very real-world, reflecting candidates and an electorate who have more than
a single issue in mind. The first time IRV selects a candidate who does not
have popular (if not extreme) support will probably also be it's last time,
and probably be such an ugly mess that election reform in the U.S. will be
set back a generation or two, and (if not already at rock bottom) further
erode the public's faith in the electoral system. In the case of a
volunteer association, it would probably have a negative impact on the
Those who argue that IRV encourages voter turnout in Australia may be
unaware that election officials go door to door after major elections
handing out $100 citations unless you can come up with a really good reason.
"Not Australian" works. Turnout under duress isn't all that interesting a
figure on the merit of an election system. More than half of the
Australians I knew were pretty unhappy with their elected representatives.
I'm not sure that's even better than plurality. None were suspicious of
their electoral system, but I suspect that's more a difference between
Australians and Americans rather than a difference between IRV and plurality
"smudging the weight of the paper". This is like the essay grading
technique - throwing them down a staircase and grading based on how far they
go? I really like the arguments like "the bad election method is bad".
If this was a real post and not a troll ... you need to be a *bit* less
emphatic with your opinions and more clear with your justifications. The
modification of IRV to handle equally ranked ballots was clear, but pretty
irrelevant with respect to the validity of Debian's Condorcet method.
[just a Debian voter who was satisfied with the results and saw no problem]