This is a fascinating discussion, and Yet Another Reason I Am Grateful
that Udhay Invited Me to Join Silklist.
On 12/20/10 6:28 AM December 20, 2010, Eugen Leitl wrote:
On Mon, Dec 20, 2010 at 11:54:09AM +0300, underscore wrote:
If you have a well audit sealed electronic voting box scenario -- it
would be significantly more difficult to rig it using traditional
Traditional, yes. But these are not the droids we're looking for.
means -- at least until the ones attempting to fix an election have
caught up with it technologically.
People who design such systems can attack them. And do it so cleverly,
you won't realize until the entire vote is stolen. And you won't even
be able to prove it's been stolen.
~
There is considerable evidence that this happened in Ohio in the 2004
presidential election, and again in several states in 2006 and 2008.
There were anomalies between the exit polls and the final vote tallies,
too many votes reported from certain precincts and too few from others,
large discrepancies between Republican/Democratic votes for local
offices with the percentages for federal offices, and so forth.
In the US, this happened only in states with electronic voting machines
and not in states with an audit-able paper trail. There were also large
swings from Democratic to Republican in a very short time in some states
(Georgia, I believe, had a 15-percentage-point swing in the 10 years
following the introduction of EVMs. Some of this was probably real, but
a number of statisticians believe it to be exaggerated).
--
Heather Madrone (heat...@madrone.com) http://www.madrone.com
http://www.sunsplinter.blogspot.com
I'd love to change the world, but they won't give me access to the source code.
