Remember kids, I worked with these an election inspector in San Diego. I was not impressed. There was the fact that there were two executables on the PC card, there were multiple reboots, there was the guy who said his vote kept changing when he tried to cast it, and, of course, the complete mismatch in physical voter count v. ballots the machine said were issued.
And then there are other problems with anything that runs on a fricking Access DB.
###
Permission to reprint granted, with link: http://www.blackboxvoting.org
BLACK BOX VOTING INVITED TO HACK DIEBOLD VOTING MACHINES
Nov 23, 2005: The California Secretary of State has invited Black
Box Voting to hack away at some Diebold voting systems. The testing
is set for Nov. 30, 2005.
Diebold Election Systems has been trying to re-certify its “TSx” touch-
screen machines in California. Diebold has added stronger passwords
and encryption, but even the consultant hired by California to evaluate
the system reported that the voting system remains vulnerable to
alteration of vote results. (More on consultant report and vulnerabilities: http://www.bbvforums.org/forums/messages/1954/14296.html)
This week, officials at the California Secretary of State's office invited
Black Box Voting, a nonprofit, nonpartisan watchdog group for elections,
to try hacking into the Diebold system. A specific testing protocol was
provided by Diebold and the California Secretary of State’s office.
Though the opportunity was welcomed by Black Box Voting, negotiations
remain on the procedures. Black Box Voting contends that the proposed
testing violates California Election Code §19202, which governs the
request for voting machine testing formally submitted to the state of
California by Black Box Voting on June 16, 2005. Also, Black Box Voting
identified areas of bias in the proposed procedures, which would violate
normal scientific protocol and cause voters to lack confidence in the results.
At issue is Diebold’s insistence on being involved in setting up the testing
procedures, and Diebold’s provision of hand-picked machines, using new
voting systems not currently in use in California.
LET'S LOOK AT WHAT'S AT STAKE FOR DIEBOLD:
Black Box Voting had formally requested replication of the work by experts
Harri Hursti and Dr. Herbert Thompson. If Diebold does not survive the tests,
the firm may face a nationwide product recall, rivaled in notoriety only by the
exploding gas tank fiasco that afflicted the Ford Pinto. Diebold is dependent
on a particular outcome. Failing this test might cost them their elections business
altogether.
Diebold’s stake in the outcome is compounded by financial problems in the
Diebold ATM division, which produced a restatement of corporate profits and
caused a significant collapse in stock prices.
Though the formal request for replication of Black Box Voting security tests
was made over five months ago, Diebold delayed the test required by §19202
for more than five months. Diebold is now “permitting” the testing only under
conditions Diebold controls, using machines only Diebold provides.
THE PROPOSED PROCEDURES CONTAMINATE THE RESULTS
Black Box Voting has offered to resolve procedural defects in such a
way as to “enhance public confidence” as required by §104 (c) in the
California certification procedures. Instead of voting machines hand-
picked by the vendor which have never been used in elections, Black
Box Voting wants to test a randomly selected voting system used in
the last election -- the machines that elected the California governor
and the president.
Black Box Voting also proposes selecting machines from county
elections offices which have not shown a bias for Diebold, recommending
Alameda County for the evaluation of the touch-screens and Placer,
Modoc, Trinity or Santa Barbara County for evaluation of the optical
scan system. Within these counties, Black Box Voting proposed random
selection of the machine. The five counties suggested by Black Box
Voting are the Diebold customers who did not take part in a pro-Diebold
advertisement copied on back of the official flyer containing the agenda
for the certification hearing. Sec. State’s office’s Bruce McDannold
disavowed the flyer, saying his office had nothing to do with its distribution.
The scientific method attempts to minimize bias by removing the influence
of any party who profits from one outcome or another. In testing, those
who design the system are not supposed to be the same as those who
test the system.
Therefore, Black Box Voting is confident that the California Secretary
of State will comply with these reasonable adjustments.
# # # # #
No comments:
Post a Comment