[EM] CVD wants Alt.V to be fairer but it isn't: misleading website

Craig Carey research at ijs.co.nz
Tue Dec 17 23:58:56 PST 2002


Here is a correction:


At 02\12\18 18:41 +1300 Wednesday, Craig Carey wrote:
...
 > >> Markus wrote (4 Feb 2002):
 > >> > In so far as IRV meets majority for solid coalitions and independence
 > >> > from clones, IRV can hardly be called "erratic" compared to primary
 > >> > with runoff.
...
 >A better rule is proportionality (it is in the style of STV/IRV) but
 >unlike the clones rule, it does not (as was alleged) reject "FPP"
 >(First Past the Post).
 >
The paragraph can be now improved to conclude "that only proportionality"
of the set that is around, rejects First Past the Post.

So, the clones rule ?:
  * worse than proportionality though similar to it?,

Normally there is a problem with bad definitions that are not really
repairable, but here I can't find defining text.

To eliminate PR-Squared, a meta-rule or rule, prohibits arbitrariness.
I guess that Mr Schulze's "clones" rule has a trait of being arbitrary.

Proportionality is not arbitrary but widely liked ("a vote for each
preference, i.e. each candidate chosen" [for preferences, not papers,
as I wrote in the last message]), and pairwise comparing certainly
seems arbitrary. Contact with politicians presumably be brief and
unfortunate if the method is arbitrary. [STV is like pulling a truck
through a clay hill with 40 chains, and the final stopping position of
the truck is not really arbitrary at all.]


Here is correction of a mistake where a word is missing:

 >It seems novel and uniquely remarkable and in my view, opposed to all
 >that the Alternative Vote stands for (at least once extrapolated out
 >to STV),
 >to have some test called "independence of clones" be both:
 >  * worth using, and
=  * which fails First Past the Post.

Obviously an authoritative website on "independence from clones" would
list a quantifier logic equation defining the idea, and I never saw
any weebpage doing that. Full generality if you are able, and check the
0 winner case (unless you have some reasoning for not doing that).




G. A. Craig Carey


----
For more information about this list (subscribe, unsubscribe, FAQ, etc), 
please see http://www.eskimo.com/~robla/em



More information about the Election-Methods mailing list