[mobile site, backup mobile]
[SoapBlox Help]
Menu & About Calitics

Make a New Account

Username:

Password:



Forget your username or password?

- About Calitics
- The Rules (Legal Stuff)
- Event Calendar
- Calitics' ActBlue Page
- Calitics RSS Feed
- Additional Advertisers


View All Calitics Tags Or Search with Google:
 
Web Calitics

Wire Services
Advertise Liberally Blue CA Ad Network

ACTION: Diebold Fails Another Security Test -- Demand Answers From S.O.S.

by: Todd Beeton

Thu Sep 14, 2006 at 00:52:11 AM PDT


( - promoted by SFBrianCL)

(cross-posted at The Courage Campaign)

On Wednesday, the Center For Information Technology Policy at Princeton University released a report that proves that the Diebold Accuvote-TS voting machines can be easily infected with an untraceable malicious virus that steals votes and gives a result different than the true vote tally:

This paper presents a fully independent security study of a Diebold AccuVote-TS voting machine, including its hardware and software. We obtained the machine from a private party. Analysis of the machine, in light of real election procedures, shows that it is vulnerable to extremely serious attacks. For example, an attacker who gets physical access to a machine or its removable memory card for as little as one minute could install malicious code; malicious code on a machine could steal votes undetectably, modifying all records, logs, and counters to be consistent with the fraudulent vote count it creates. An attacker could also create malicious code that spreads automatically and silently from machine to machine during normal election activities — a voting-machine virus.

Go here for the full paper.

UPDATE: The video that accompanies the study is below:

I contacted Chuck Hahn, Assistant Secretary of State Policy & Planning to get Secretary of State Bruce McPherson's response to this latest indictment of the very technology that he himself certified for use in California's elections. Remember that McPherson certified Diebold even after a study that his very own office conducted found "serious flaws" in the Diebold software. His certification was "conditional" apparently on their insistence that they would fix the vulnerabilities.

At the CA GOP summer convention, I asked McPherson directly what reassurances Diebold had given him that these flaws would be fixed. He simply said "I wouldn't have certified them if I didn't think they were secure." And when I asked if there's anything online or in the public record that can reassure voters of Diebold's good faith efforts, he told me that they held public hearings that anyone could attend. That's when Hahn stepped in, gave me his card and asked me to give him a call and he swept McPherson away.

So in the wake of this new revelation I did call him and e-mailed him to make sure they were aware of the report and to get the Secretary's response. I received a return e-mail from Hahn not long after that, which, while swift, was still quite dismissive UPDATE:he prefers not to be on the record so I've taken down the text of the e-mail. He's been helpful and responsive so far. Hopefully we'll get a swift response from McPherson soon. This is about keeping pressure on him to answer for his decision to re-certify these machines. As I say in the comments, it appears from his certification announcement that CA may not even be using the machine tested. But that's not the end of the story, nor is it the end of the explaining we must demand from McPherson.

Todd Beeton :: ACTION: Diebold Fails Another Security Test -- Demand Answers From S.O.S.
Tags: , , , (All Tags)
Print Friendly View Send As Email

video on their site is slow (0.00 / 0)
Try YouTube. Email the link to everyone you know, Democrats, Republicans, apolitical, whatever. People need to see this!

do you have a list of who uses what? (0.00 / 0)
I've been trying all day (in between work heh) to find a list of which counties in CA will use what for Nov. It would also be nice to know which counties if any have used the AccuvoteTS in the past. I tried even just finding out what San Diego county is using. The registrar of voters site didn't have any specifics other than a call for volunteer poll workers, with assignments including "touchscreen inspectors" whose responsibilities, among other things, includes driving the machines in their car to the polling place the night before (!!), and driving them back to the registrar after polls close. But no naming the machine specifically, or even the vendor. Then I found this "how to vote" video, that appears to have a voter-verified paper trail, which is GOOD. But still no specific make & model #'s, are all preccints using this??, etc.

OK, CA may not use this machine after all (0.00 / 0)
This study tested the Accuvote TS machine and as you can see HERE, McPherson certified TSX for use in California.

This may be what Diebold meant when they said in their response that the security software used in the test is "2 generations old."

I'll be back with McPherson's official response when it comes and will stay on this.


[ Parent ]
well its a mistake to think this is the only problem (0.00 / 0)
Any comparable product will have a comparable vulnerability. Any vendor, even the non-Diebold ones. If you have a single-pane glass window on your house, one potential way of breaking into your house is to break the window. It doesn't matter which builder built your house, or what brand the window is. The fundamental setup is the same, the vulnerability will be there. Same thing for voting.

Basically any system that doesn't have a VOTER-VERIFIED paper trail is totally unacceptable.

Other problems include machines being out of sight of the public AT ANY TIME. The whole "sleepover" thing (where machines are sent to polling places, including personal residences, day(s) in advance of the election) is horrible. As the video shows, you only need one minute. Imagine all night! Also, many machines have a demo/test mode. This is a horrible idea because it tells the machine this isn't a real election. So it is ridiculously easy for the hack software to appear normal for the test, and steal votes in non-test mode only. But even these problems are largely mitigated by having the voter-verified paper trail.


[ Parent ]
I would assume the paper records would be kept (0.00 / 0)
just like a normal paper ballot so people could just count them?

I think?

[ Parent ]
ya (0.00 / 0)
Voter-verified paper trail means that you use a touchscreen (or similar) to vote, but instead of just the software keeping count, invisible to everyone (therefore hackable and impossible to detect the hack), it prints out what you did on paper. The voter can view that paper through a window. They make sure that it matches what they want. If it does, they say accept and it drops into an attached locked ballot box. If not, it goes down a different chute into the trash (or shredded/overwritten/etc).

If there is any question about the total that the software kept, you open the box and recount the paper ballots. One nice thing is that this is better than voter-marked ballots, because machine marking is obviously more precise and uniform (no hanging chads, incomplete erasing of scantron, etc).

Basically it is like regular voting but the machine marks your ballot for you. Also the touchscreen can do nice things like easily show many different languages, protect against over-voting, etc. And again, the marking of the ballot is more precise/uniform than when you let people do it.


[ Parent ]
You answered my question completely (0.00 / 0)
It's what I thought.  It makes sense, and seems relatively easy to accomplish.  Why could anybody be against this other than to pervert democracy?

Ah those perverts...

I think?


[ Parent ]
glad to know countless thousands of dollars (0.00 / 0)
spent on my CS education wasn't a waste, haha.

[ Parent ]
not sure exactly what your question is (0.00 / 0)
But just to be extra-clear, the Diebold TS machine (like in the video) does NOT keep a paper record at all! The printer you see in the video is only used at the very end of the election--it prints out the final tally. In the mean time, nobody knows what the hell was happening inside, least of all the voter. Doesn't take Sherlock to figure out that's the worst idea ever!!

[ Parent ]
I was under the impression... (0.00 / 0)
that a paper trail is required in California. but perhaps not a "voter verified" one?

[ Parent ]
yeah that sounds very familiar (0.00 / 0)
... Ok I double checked the details and you're right. SoS Shelley announced in 2003 that he would require all machines used in CA to have a voter-verified paper-trail by July 2006. (link) That makes sense because I recently voted on a machine WITHOUT such a paper trail, but the policy wasn't in force until a month ago.

California will become the first state requiring all electronic voting machines produce a voter-verifiable paper receipt.
...
With a receipt, voters will be able to verify that their ballots have been properly cast. However, they will not be allowed to keep the receipts, which will be stored at voting precincts and used for a recount if any voting irregularities arise.

Beginning July 1, 2005, counties will not be able to purchase any machine that does not produce a paper trail. As of July 2006, all machines, no matter when they were purchased, must offer a voter-verifiable paper audit trail. This means machines currently in use by four counties in the state will have to be fitted with new printers to meet the requirement.

The article doesn't say anything about this being a law, it sounds more like it is just a policy announcement by Shelley. I don't know how binding it is now that we have McPhearson.


[ Parent ]
intent vs implementation (4.00 / 1)
One of the things that's been disturbing to me, is that McPherson talks about these security precautions that should be taken with the machines, but is anyone enforcing those precautions at the county level?

For example, the SOS talks about maintaining the integrity of the chain of custody, but then the counties do a terrible job of enforcing that by allowing "sleepovers" for days and weeks before the elections. The county RoV's justify it by saying "that's the way we've always done it" or they talk about how we have to trust them.

My mama told me to never trust anyone who says, "trust me".

In addition to the obvious concerns about the machines themselves, I think there's a breakdown between what the state says should be done and what the counties are willing to do. There's a big gap between intent and implementation.

Right after the June election I sent an email to the SOS office asking about this specifically, but about 4-6 weeks later I got a form letter in response that didn't address my questions about enforcement of election security at the county level at all.


Calitics in the Media
Archives & Bookings
The Calitics Radio Show
Calitics Premium Ads


Support Calitics:

Get discounted bestsellers at Barnes & Noble.com!

Advertisers


-->
California Friends
Shared Communities
Resources
California News
Progressive Organizations
The Big BlogRoll

Referrals
Technorati
Google Blogsearch

Daily Email Summary


Powered by: SoapBlox