10/9    "approval voting *eliminates* the spoiler problem, how
        do you feel it fails to address it?  IRV,
        however, partly because it is not monotonic,
        and due to several other side effects risks
        *severe* spoiler effects. -dans
        \_ I don't think IRV's shortcomings in voting theory really
           are a practical concern. In common cases the results seem
           acceptable to me. How does approval voting eliminate spoilers?
           If there are A,B,C and I prefer A>>B>>C, do I vote for B or not?
           If I want A to win, I could not vote for B, but then I risk
           C winning which I really don't want. It depends on the perception
           of how likely each one is to win. It doesn't appear that different
           from the winner take all thought process.