[EM] fix to Rivest 3-ballot flaw! very simple

raphfrk at netscape.net raphfrk at netscape.net
Wed Oct 4 03:23:41 PDT 2006


 > From: wds at math.temple.edu
 > I posted it on http://groups.yahoo.com/group/RangeVoting/ 
 
 <Summary for those that don't want to go to the RangeVoting List>
 
 The method amounts to allowing the voter to vote
 FOR two candidates and AGAINST 1 candidate.
 
 The AGAINST vote is a FOR vote for all the other
 candidates.
 
 Also, the AGAINST candidate has to match one of the FOR 
 candidates and both FORs have to be different.
 
 The voter can get a copy of any of the 3 votes.
 
 (If that is an incorrect decription, please correct me)
 
 However (from the RV list), Warren Says:
 
 > Well... slight flaw. If I look at your FOR vote (either for vote)
 > then I have no idea who you voted for. *But*
 > if I look at your AGAINST vote,
 > then I know who you did NOT vote for reducing the set from N to N-1.
 > That is a bummer, I must admit. However, not a very big bummer?
 
 <The proposed soltion to this uses 5 ballots instead of 3>
 
 However that assumes that the following is invalid:
 
 1) FOR: A
 2) FOR: A
 3) AGAINST: A
 
 This would be the case using Warren's stencil system as 
 if a voter uses his FOR A stencil on ballot 1, the 
 voter cannot use it for ballot 2.
 
 However, if the mechanism just ensures that the AGAINST 
 ballot matches one of the FOR ballots, then the problem 
 disappears.
 
 I would suggest that there are 2 sets of stencils. Each 
 set contains a FOR/AGAINST vote for each candidate.
 
 There would be 4 ballot slots. The voter places 4 ballots
 in the slots. The voter then applies a stencil for each pair. 
 This marks the ballots so that each pair has opposite votes. 
 The voter then selects one ballot as his receipt and the other 
 3 are cast.
 
 The receipt will contain the opposite vote to the vote it
 is a receipt for. However, that's not a problem.
 
 Also, I wonder if it would be a good idea to have 2 types
 of ballot. A FOR ballot and an AGAINST ballot. It seems
 pointless to mark the AGAINST ballot by marking FOR all
 the other candidates on a FOR ballot. The voter would
 receive two pairs of ballots.
 
 This removes the need for stencils entirelly. A slider
 would be moved up and down to select a candidate. The
 machine will then stamp both ballots at that place.
 This automatically creates a FOR/AGAINST pair. The
 2 ballots would have to be shaped slightly differently
 so that the voter can't place two FORs in the slots.
 
 One issue is getting the ID numbers to match, for the 
 receipt and its ballot and not for the other pair. I am
 not sure how easy that would be to accomplish.
     
________________________________________________________________________
Check Out the new free AIM(R) Mail -- 2 GB of storage and industry-leading spam and email virus protection.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20061004/9a524c35/attachment-0003.htm>


More information about the Election-Methods mailing list