You are not logged in. Login Now
 0-24   25-49   50-74   75-99   100-124   125-149   150-174   175-199   200-224 
 225-249   250-274   275-299   300-324   325-349   350-374   375-399   400-424   425-449 
 450-474   475-499   500-524   525-549   550-574   564-588   589-594    
 
Author Message
6 new of 594 responses total.
scg
response 589 of 594: Mark Unseen   Sep 23 20:25 UTC 2003

Electronic voting machines should be pretty easy to test beforehand --
simulate lots of elections, keep very close track of how many votes were cast
for which candidates, and audit the results.  My understanding is that the
Diebold machines aren't quite so easy to test, as part of the Diebold election
program involves Diebold having remote access to the voting machines while
the election is going on, so they could at least in theory be issued commands
during the real election that wouldn't have been issued to them during the
tests.

That said, there are plenty of rich and/or influential people who make
promises to deliver a state to a certain candidate, generally through the use
of endorsements or fundraising, who have no control over the voting process.
Sometimes such people actually succeed.  Sometimes they fail (John Engler's
promises to deliver the Michigan Republican Primary to George Bush in 2000,
for example).  While Mr. Diebold is in an interesting position to fulfill his
promises through sinister means, I don't think we can conclude that his intent
is to do so.  On the other hand, the lack of an audit trail also opens up lots
of possibilities of fraud by local election officials.  Reset the machine in
the middle of the day, input the right number of votes to match the number
of voters who came through that morning, and you won't have any risk of
somebody noting that more ballots were used than counted, or of your ballot
boxes being found floating in the Bay (San Francisco, a few years ago).

It seems to me that all the usability of the touch screen voting machines
(ignoring that they're not actually all that easy to use) could be gotten
along with an audit trail, by having the voting machines print out a completed
ballot that could be looked over and placed in a ballot box, after each voter
finished voting.  If nobody was suspicious of the results, the counting could
be done easilly via the machine's internal counting features.  If the results
looked wrong, or somebody wanted to challenge them, a handcount of paper would
still be possible.  If what was being printed on the paper was different than
what was being entered into the voting machines, presumably somebody would
notice.
scott
response 590 of 594: Mark Unseen   Sep 23 20:43 UTC 2003

if (today == ELECTION_DAY) {
        mode = REPUBLICAN_UPSET;        
}  else {
        mode = HONEST;  
}
gelinas
response 591 of 594: Mark Unseen   Sep 23 21:21 UTC 2003

Which is why I like optical scanning of paper ballots marked by the voter.
But this is a long way from the "I'm Bummed Because . . . " topic.
gull
response 592 of 594: Mark Unseen   Sep 24 00:56 UTC 2003

http://www.salon.com/tech/feature/2003/09/23/bev_harris/index.html is 
another interesting article about this.  It talks about the general 
problems with electronic voting machines, and about a specific problem 
with Diebold machines that allows the results to be easily altered with 
no audit trail.  (Diebold was made aware of this problem and responded, 
essentially, that it was a feature, not a bug.)  It also points out 
hints of touch screen vote fraud in Georgia and King County, Washington. 
 Finally, it notes that in a 2001 MIT/Caltech study the only voting 
system that had a worse error rate than punch cards was...touch screens.
richard
response 593 of 594: Mark Unseen   Sep 24 02:50 UTC 2003

IBB it is officially fall today and we are still in summer agora.  will this
summer ever end?!  bring on the fall!
gelinas
response 594 of 594: Mark Unseen   Sep 24 04:06 UTC 2003

Oh, relax, richard.

IBB I've not yet registered Sol II and now it has expired.  (Sol II is a Palm
application that computes sunrise, sunset, BMAT, BMNT, BMCT, EECT, EENT, EEAT,
moonrise and moonset.  "BMAT" is "Beginning of Morning Astronomical
Twilight.")
 0-24   25-49   50-74   75-99   100-124   125-149   150-174   175-199   200-224 
 225-249   250-274   275-299   300-324   325-349   350-374   375-399   400-424   425-449 
 450-474   475-499   500-524   525-549   550-574   564-588   589-594    
Response Not Possible: You are Not Logged In
 

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss