California invites 'Black Box Voting' to hack Diebold voting machines
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.
BBV
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/mess...954/14296.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 in June 2005, 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 procedure contaminates 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.
