Who Says Electronic Voting is Safe?
October 24th, 2008By Monica F. Helms
Since the first time they came out, electronic voting machines have been accused of providing inaccurate information and could easily be hacked into and change the results. On the one side, proponents and manufactures insisted that the machines are secure, while opponents insisted they could be remotely “adjusted” and would like paper backup for their voting results. Many states who use electronic voting machines print paper backup forms to verify the results.
In July of 2002, Georgia’s Secretary of State, Cathy Cox (D) signed a contract with Diebold to provide electronic voting machines for the entire State of Georgia. We became the first in the nation to go totally electronic voting. In August 2003, Walden O’Dell, then CEO of Diebold, announced he became the top fundraiser for President Bush and the Republican Party and said he would do whatever it takes to get Republicans elected. Pretty scary comment from the person whose company builds the very machines our votes get recorded on.
(Break)
Shortly after that, I had a chance to ask Cathy Cox directly that if the CEO of Diebold promised to get Republicans elected, then how can she assure voters that our votes were safe? Cox insisted the Diebold machines had no problems and our votes would be counted correctly. She wouldn’t even consider adding paper backup ballots in case of a dispute.
I then asked her two questions that she did not want to answer, or maybe could not answer. I asked, “Can Georgia technicians open the machines and look at the internal workings and can Georgia programmers program the machines?” I asked her these questions more than once in face-to-face conversations and in E-mails. She staunchly defended the integrity of the Diebold machines; machines that could only be worked on and programmed by Diebold technicians. It never gave me a good feeling that my vote would be counted correctly. In November of 2002, Republican Saxby Chambliss defeated the popular incumbent Democrat, Max Cleland for a Georgia Senate seat, even though the poles showed Chambliss behind. Coincidence? I think not.
Who says electronic voting is safe? Not me. I have worked for a major telecommunications company for 19 years now and I have seen advances in communications and electronics that make the original Star Trek series look silly in comparison. We have become a wireless world, making each person highly mobile and free. I have a new Sprint Instinct phone that has all the same features as the Apple iPhone, and some extras. The iPhone can do a few things the Instinct can’t and visa versa. You can read all of your E-mails, surf the web and even watch CNN live on the Instinct. If you can imagine it, computers and phones on the market today can do it, without wires. But, they can’t make coffee yet.
In this completely Wi-Fi world, why is it hard to believe that voting machines like the ones built by Diebold are immune from someone going into them remotely and making changes through wireless access? I sure can’t.
On September 8, 2008, Judge Michael Johnson heard the case filed in 2006 by Voters Organized for Trusted Election Results in Georgia (VoterGA) challenging the State of Georgia on the use of electronic voting machines in Superior Court of Fulton County. VoterGA lost the case. Georgia and Maryland are the only two states who still use the unverifiable Diebold AccuVote-TS R6 machine; however Maryland will be throwing them out after this election to go back to optical scanning machines. This mean that Georgia will still be venerable from remote access to our voting machines. The Republican controlled state government is probably smiling about that and thanking former Democratic Secretary of State Cathy Cox for not installing paper backup.
I have to go, now. I have a text coming in to my Instinct phone and I have to remotely send this article to a friend over a Wi-Fi network.
October 26th, 2008 at 3:28 am
For some interesting information on Premier Election Solutions (the Diebold subsidiary that make its voting machines), see:
http://en.wikipedia.org/wiki/Premier_Election_Solutions
October 26th, 2008 at 5:51 am
Also very interesting is the Wikipedia article on Electronic Voting:
http://en.wikipedia.org/wiki/Electronic_voting
One particular item of note from that entry:
“The Premier Election Solutions (formerly Diebold Election Systems) AccuVote-TSx voting system was studied by a group of Princeton University computer scientists in 2006. Their results showed that the AccuVote-TSx could be ‘installed with vote-stealing software in under a minute.’ The scientists also said that machines can transmit computer viruses from one to another ‘during normal pre- and post-election activity.’”
October 29th, 2008 at 1:15 pm
While the problems with the voting machines are well documented, there are many other ways attempts will be made to steal this election. I have heard directly from people in Florida about ways it was done there in the 2000 election. For example, they would fix streets surrounding polling places in predominately Democratic areas, and block the street around them. Voters are being challenged, and voter suppression is already a hot topic. I have already cast my vote by mail, and it was good did because I am visiting a sick friend in North Carolina, and may not have been able to vote in my home state of Colorado if I had not arranged previously to vote by mail. The American people must be vigilant on Election Day to be sure the election isn’t stolen – again.
November 2nd, 2008 at 9:15 pm
How to do it -
Have standard hardware, commercial off-the-shelf, so there’s no “trade secrets” involved. If not, have open-source hardware made to the same security standards as automatic teller machines.
Have independent checks and testing by academic institutions – and anyone else who wants to.
Have the source code “open source” so anyone can look at it, and the electoral officials verify both before and after that that was the system used.
Example.
Note that you too can download the source code. Anyone can. It’s all open for any concerned group to have a look at.
Security for the machines is exactly the same as the security for the parallel paper system, and anyone who wants to can use a paper ballot instead.
Disclaimer: I worked for Software Improvements when they made the ACT electoral system, though I wasn’t directly involved. I was too busy making satellites.
I did however work on the BILBY model compiler, which in future versions will directly take the specifications in Unified Modelling language and translate them into Ada-2005 code. This is the same language used in Avionics and Satellites, and is designed to be particularly clear to read and find flaws in. A system built using BILBY was used to collect votes over a Defence-Grade secure network from servicemen and women overseas, in Iraq, Afghanistan, and elsewhere.
How NOT to do it…..
Proprietary (and trade-secret) hardware running trade-secret software with only those chosen few granted access to test. Anything running over a non-secure network (ie less than full military encryption and both-end verification, unless physically secured over the entire cable run)
I note the Wikipedia article doesn’t mention either the Australian Electoral Commission’s experience with the systems, nor Software Improvements as a manufacturer.
I have to be careful about “bad mouthing” the competition, but the best way to show that the system works is to let anyone see the internals, and verify things for themselves.