Stephane wrote: >I do not say that a truncated preferential ballot could be used to >generalize an approval method. Maybe it could with a kind of Borda count. >All I say is why restrict the voter input to 2 or 3 categories, when some >would gladly give more?
What do you propose doing with those categories, though? Cardinal Ratings has been invented and discussed, and it's very similar to Borda Count except you can put as many people as you want in each category. i.e. Borda gives n points to first pick, n-1 to second, etc. but with CR you can give n votes to as many candidates as you like, and n-1 to others, or no votes to some, or just 1 or 2, or whatever you like. Most people here argue that strategically it's just like Approval (give every candidate all or none to have maximum effect) but it gives greater freedom to those who don't want to follow that strategy for whatever personal reasons. The only other way I can think of to use those extra categories is something like 3-level approval, where second-tier votes only count if nobody has a majority of first-tier votes. I suppose you could go to 3rd tie, and so forth, but the key difference from Bucklin is that you can put more than one candidate in each category. So, my question remains, what do you propose doing with those extra categories if we include more than 2 or 3 categories in approval extensions? Alex ---- For more information about this list (subscribe, unsubscribe, FAQ, etc), please see http://www.eskimo.com/~robla/em
