Re: University Researchers Challenge Bush Win In Florida

["Jason Coombs" <jasonc () science org>]

As for source code or other security vulnerabilities in closed- or open-soure vote tabulators, there is little point in 
rigging such schemes, and less point in exploiting them. Good old fashion statistical abberations exploited for the 
benefit of the party that finds them first will win every time.

In principle, all voters have roughly the same risk of their vote not being counted under any electoral system. This is 
called 'equitable risk'.

If, through testing of electronic voting machines, statistical anomalies can be detected that favor the candidate that 
is entered into the database third (or whatever, take your pick, and it would be different for different voting 
machines and maybe in different regions, say, because Florida is full of elderly) then you can 'rig' an election in 
your favor simply by having a non-random selection for the order in which the candidates get listed, and a failure to 
properly distribute that randomness across precincts.

If anything, that is what I believe is most likely to have happened in 2004. Bush elected through the (fair ?) 
exploitation of statistical anomalies tied to misbehaving or ill-conceived electronic voting equipment. Teamed with the 
fact that partisan, interested voters are in charge of the process this is very plausible...

Sincerely,

Jason Coombs
jasonc () science org

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.netsys.com/full-disclosure-charter.html