What Don calls "bottom methods" are methods that look at more than 1st choice preferences. "Bottom", then, being used to mean anything but "top". So, while IRV & Plurality look only at 1st choices, Approval, Borda, & Condorcet look at overall high-ratedness. Of those 3 methods (Don used to call them the ABC methods), Borda has unique strategy problems that prevent it from being acceptable for public political use. But, under conditions where voting is sincere, Borda would arguably be the best rank-count, if we want the greatest good for the greatest number. Though those conditions don't exist in our political elections, Borda's ideal merit is still useful to mention in discussions like this. What makes IRV & Plurality different from those 3 methods is that IRV & Plurality only look at one small part of the overall set of voters' ratings of candidates. It's not surprising that that causes IRV to screw up big-time. Borda (if voting is sincere), Approval, & Condorcet look at the whole situation together. You see, the problem of looking at it piecemeal is that when you look at one little piece, you can do something that doesn't make sense in terms of the whole picture. So IRV is acting ignorantly when it does its eliminations & transfers. How can anyone say that IRV can do the right thing when it's only looking at a little piece of the overall situation? IRV's overall procedure can't be expected to be any good if it's doing irrevocable actions based on looking only at a small part of the overall situation. That should be obvious. Needless to say, what I say in this letter is in addition to my other objections to IRV, which mostly involve criterion failures. Criterion failures caused by IRV's ignorant piecemeal way of looking at the ballots. Mike Ossipoff _____________________________________________________________________________________ Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com
