On Sat, Dec 07, 2002 at 10:36:45AM -0500, Raul Miller wrote: > > > Proof that A.6 draft is undemocratic. > > I'm ignoring your proof because you've left out something I consider to > > be significant out of your axioms. [see above.]
On Sun, Dec 08, 2002 at 02:10:18AM +1000, Anthony Towns wrote: > I'm not sure that's fair, since quorum doesn't come into play in the examples > given. If you prefer, just multiply by 10, or 100, or whatever you feel is > appropriate. I'm critiquing the axiom, not the example. By his rules some elections with quorums do not have a democratic outcome. > In more detail, and multiplied, the example is: > > D defeats A by 60:50 (20:50, but multiplied by 3:1 > supermajority requirement) > A defeats B by 60:10 > B defeats D by 50:20 > > (simplifying the votes slightly gives: 50 ABD, 10 DBA, 10 DAB; which > is to say most people prefer A to B, most people find A acceptable but > not enough for supermajority, and most people find B acceptable and it > doesn't require a supermajority) > > ...and the question is which defeat to eliminate: B v D because it's > weaker, or A v B, because A can never win anyway. Given that A can't > possibly win by assumption (it failed its supermajority requirement), > I'm not seeing why it's not sensible to say "well, most people would > rather `B' than further discussion, so let's go with `B'". The issue here is the impact of transitive defeats. Should options whith supermajority requirements have their votes scaled with respect to options in general or only with respect to the default option? I suppose I should dig through the archives from a couple years ago for the examples you brought up when I was proposing this kind of scaling and when you were objecting to it. > Which is to say that "if option X doesn't defeat the default option by > its supermajority requirement, it is ignored" seems to be fairer than > considering defeats by the default option as especially strong. Define "fairer" -- the definition of "fair" is the crux of this issue. -- Raul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
