(I sent this yesterday morning, but unfortunately I cut and pasted [EMAIL PROTECTED] instead of [EMAIL PROTECTED] Anyway, it still seems on-topic.)
I read through "Rivest-TheThreeBallotVotingSystem.pdf," and I was wondering one thing. It says: To vote FOR a candidate, you must fill in exactly two of the bubbles on that candidate's row. You may choose arbitrarily which two bubbles in that row to fill in. (It doesn't matter, as all three ballots will be cast.) To vote AGAINST a candidate, you must fill in exactly one of the bubbles on that candidate's row. You may choose arbitrarily which bubble in that row to fill in. (It doesn't matter, as all three ballots will be cast.) Why not have ballots pre-printed with randomly filled bubbles (or pre-punched chads), where there is exactly one filled bubble in each row? Voting would be simply finding an empty bubble next to a candidate's name and filling it in. Voters could easily verify that there is only one vote next to each name if they wished (going down the row with a piece of paper would be sufficient to tell at a glance), but they wouldn't have to spend all their time in the voting equivalent of an SAT test. In addition to randomizing the filled bubbles, you could ensure that each ballot could be uniquely identified (no two ballots have the same bubble order). This order could replace the ballot ID number, and a person could use this to search through the list on the internet to verify their ballot was counted. In a plurality election, mandating that there were at least three differences in bubble order for each ballot would ensure that no two ballots were identical when turned in, and if there are too few candidates to do this you could add a randomized ID number to the bottom to ensure uniqueness. In an Approval election, there would be no way to ensure each ballot is unique without such an ID number, but you could still make certain the starting bubble order was not only randomized, but also started out as different as possible from all other ballots. I might be overlooking something, but I thought I'd toss out this suggestion and see if it would work. Michael Rouse [EMAIL PROTECTED] ---- election-methods mailing list - see http://electorama.com/em for list info
