Skip to comments.Lax controls over e-voting testing labs [Amazing - no enforcement authority over 'independent' labs]
Posted on 05/30/2004 9:08:26 AM PDT by John Jorsett
California Secretary of State Kevin Shelley had a simple question: Had a new electronic voting machine been approved by an independent testing lab?
State law requires such approval before the device could be used by California voters. It guaranteed the machines counted votes accurately and would work reliably during an election. As the state's top election official, Shelley figured he could get a quick answer.
He figured wrong.
Wyle Laboratories of El Segundo refused to discuss the status of its testing of the AccuVote-TSx machine made by its client, Diebold Election Systems. The information was proprietary, Wyle said, and could be revealed only to Diebold.
And so the secretary of state was introduced to the looking-glass world of voting-machine regulation. Over the years, repeated references to ``federal testing'' by election officials have given the impression that the government oversees the certification of touch-screen voting systems. While there are guidelines for the machines, no federal agency has legal authority to enforce them.
(Excerpt) Read more at mercurynews.com ...
This relationship sounds similar to the Enron/Arthur Anderson one.
I've been beating up my state rep on the non-recount-ablility of touch screen systems, and emailing him articles like this, for the last 2 years. I think he's tired of hearing from me.
This issue is apparently on no one's radar screen here in GA.
I don't like it either.
In San Diego, our registrar assured us that these machines were great, no problems, blah blah blah. Then when disaster struck and 30% of them couldn't be made to work early on election day, she came on a talk show and was saying how everything was fine as her phone lines were melting down from angry voters and paniced poll workers. I can't make up my mind whether she drank the Diebold Koolaide or is just an incompetent ass. And our Grand Jury just came out with a report about how the CA Attorney General was wrong to force us back to paper ballots in November, the machines will be just fine (assuming their fixes are implemented by the above-named registrar; good luck with that), yada yada yada. I'm beginning to think Diebold has developed some sort of mind control beam or something.
The potential to rig elections with these machines is truly frightening. We should be marching on the election boards and elected officials, with torches and pitchforks, on this one.
I hav not been following this very closely maybe I need to send off some mail too..
The Diebold machines are stand alone registers. Most of the brouhaha on this subject assumes they are online and accessible to malicious hackers 24/7. It's simply a register that accumulates information - which can then be summarized, then connected to a network or offloaded (burned to CD, uploaded by some dial-up arrangement, etc.) at the end of the day.
I agree that the process of vote tabulation should be verifiable and transparent. But what is the standard that we have had for (at least the last 30 years that have been) voting? A bunch of old ladies who rotate between the garden club bake sale in an adjacent room at the firehouse and the registrars table - looking at green bar printouts and showing a level of scrutiny that is almost non-existent. Are you going to ask these little old ladies to fix paper jams in the printers and replace toner cartridges? California requires ballots to be printed in some staggering number of languages (something like 15) and has notoriously long referendum ballots - how long would it take for each voter to get a 37 page "receipt" for their vote and how much would it cost to provide that receipt?
I assume that the solution which will emerge is a bar coded receipt - on an ATM sized piece of paper. Voters would get one of these receipts with a sequence number printed on it, the number given to them by the registrar as they checked in (you are number 587 voting at this precinct today....). The bar code image would be unintelligible to the naked eye, but could be interpreted at a verification terminal - where the voter could privately verify that his/her selections were registered correctly. A random sample of the information collected on the verification terminal could then be matched with the final vote tabulation to see if any discrepancies were present before a summary was uploaded to the secretary of state's site/system. Not all voters will take the time to verify the receipt, but enough will to provide a realistic check on the count. Utilizing an ATM printer/paper format will lower the costs to whatever ATM machine transactions costs are - and servicing and maintenance of the machinery will not be significant problems.
I'm all for having voting systems with six sigma level of precision. Take an honest look at the system we have accepted until 2000 - with all the flaws in voter registration (and not just tabulation) - and then tell me that these systems do not represent an improvement over the old regime. If you want to float these theories about machine corruption come up with something resembling evidence.
Most democrats couldn't read your long post, much less the operating instructions that are presented to an electronic voter. I like computer voting for that reason but the hacking risk is too high. A hostile government or political party with $300 million to spend could pull some sort of magic. With the counting devices exposed out on the voting floor there's plenty of physical access.
I hope that was directed at the author of the article and not to me, since I'm not floating any theories about corruption.
Personally, I love touch-screen voting, but what I have observed is a complete lack of security on these things, at least here in San Diego. The poll workers were allowed to take the machines home and some had them for weeks before the election. The software was protected by just a peel-off sticker that could easily be removed and replaced without detection. When many of the machines failed to operate early on election day, poll workers allowed random people who wandered into the polls to try to get them going by mucking around in the devices. There were no printed receipts. I don't know of any actual tampering that was done, but the potential was there. As my late father, an accountant, once said, any time you don't have safeguards in place, someone will ultimately exploit the system, guaranteed. He was talking about accounting and audits, but the principle applies here as well. The machines need to be secured and the software certified, and there need to be printed receipts in the future so that, as you noted, correct operation can be verified and recounts done.