Questions Raised About Security of Voting Technology The midterm elections are two weeks away, and questions remain about new voting machines. Guests examine the technology that will be available on Election Day. Can you count on your vote being counted?

Questions Raised About Security of Voting Technology

  • Download
  • <iframe src="https://www.npr.org/player/embed/6375956/6375957" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

NEAL CONAN, host:

This is TALK OF THE NATION. I'm Neal Conan in Washington.

Election Day is two weeks away and when voters get to the polls, nearly a third of them will face some sort of new technology. Lever machines and punch cards are their way out amid changes mandated after the debacle in Florida, in the 2000 presidential election. Some, though, worry that the new technology creates at least the potential for problems a lot worse than dangling chads.

Several candidates have told their constituents to vote by absentee ballot this year, rather than risk mistakes with new touch screen machines. Academics have exposed flaws in machine security. And primary elections over the past several months, most notably in Maryland, were plagued by a combination of glitches, mistakes, and poorly trained workers. The companies that make the new voting machines insist that they are secure and reliable.

Today, a look at the changes and what we can expect to go right and wrong. Later, we'll talk with filmmaker Noland Walker about a new documentary on Jim Jones, the People's Temple, and the horror of Jonestown; a project that for him cuts very close to home.

But first, changes in voting technology. We'd like to hear from voters and poll workers. Are you concerned or confident in the new technology for this election? If you used or manned a poll where new technology was used, what was your experience like? Our number here in Washington is 800-989-8255. That's 800-989-TALK. The e-mail address is talk@npr.org.

And we begin with Kimball Brace, he's president of Election Data Services, a non-partisan political consulting firm. And he's kind enough to join here in Studio 3A today.

Thanks for coming in.

Mr. KIMBALL BRACE (President, Election Data Services): Thank you.

CONAN: Your organization recently released a survey on new voting technologies, and there's going to be a lot of them. This is sweeping change.

Mr. BRACE: This is the largest change that we've seen, ever, in the history of this country. All at one time, over a third of the people will be voting on new pieces of equipment and new - totally new devices, from how they voted just two years ago.

CONAN: Mm-hmm. And when we say new devices, this is not all touch screen technology.

Mr. BRACE: That's correct. I mean, a lot of jurisdictions went to optical scan devices, and therefore got away from the electronic controversies that we've seen recently. And so, touch screen - or touch optical scan ballots were kind of a failsafe backup, as a way of giving the voters something that he could see; having a piece of paper, which is what people were looking for, and have something there for a recount possibility.

CONAN: And that's what a lot of people are concerned about, that there ought to be - you should excuse the expression - a paper trial. You could go back and recount the actual votes if you need to.

Mr. BRACE: Right. And that has been the main source of the controversy on the electronic systems this year. But keep in mind that electronic systems are not totally new. There's a new version of them that are out now. But electronic systems really came into being in the late 1980s and early 1990s. We had the old large full screen type of device down in Louisiana, another - a number of other jurisdictions. So we've had electronic systems, it's just the older models that we had before.

CONAN: Now, for those of us who are unfamiliar with technology in other counties or other states, or other towns in some places…

Mr. BRACE: Right.

CONAN: In this country, it depends on where you live as to who runs the election.

Mr. BRACE: That's correct. That's correct.

CONAN: But an optical scan device, how does that work? You make a mark on a piece of paper?

Mr. BRACE: An optical scan, I always describe it like the SAT test that you and I took when we were in college; and filling in the oval, filling in the box, that sort of thing.

CONAN: It's a conspiracy by the makers of Number 2 pencils.

Mr. BRACE: Absolutely. You got to have that Number 2 pencil. And that was one of the problems with the early optical scans, is that it required just that kind of pencil. Now they're a little bit more flexible in terms of other types of marking capability. But the optical scans are critical that people vote them properly. You need to fill in that oval, or fill in that box. Don't circle the name and think that that's going to count, because it's not. The machines are looking at just a small, little portion of that ballot, in order to see if you voted for Kim Brace or whoever.

CONAN: Mm-hmm. And now, the touch screen - if you're unfamiliar with that - that works like, well, an ATM machine a little bit.

Mr. BRACE: It is like an ATM machine. It - you've got names and that sort of thing in terms of what's on the computer screen. It allows you to touch it and it marks it either by a check mark next to the candidate's name, or highlights the candidate in a different color - that sort of thing. There are advantages to electronic systems. You can have multiple languages on the same device. You don't have to print up, you know, three different types of ballots with Spanish and Vietnamese, or whatever the case may be. They can all be on that electronic machine.

CONAN: Mm-hmm. There are drawbacks, as we mentioned: no paper trail.

Mr. BRACE: No paper trail from the - in a first cut kind of way. There's been a lot of improvements in terms of many of the vendors to include, now, a paper trail option onto it. There's also the capability with many of the electronic systems of being able to print out the ballots afterwards. Now, that is not something that a voter is verifying at that point in time, but it does provide some degree of backup capability to it.

CONAN: We want listeners involved in this conversation. If you've used these technologies, give us a call. Tell us what your experience was like. Particularly interested in hearing from poll workers too. 800-989-8255, 800-989-TALK, e-mail is talk@npr.org.

Roy joins us on the line. Roy calling from Columbia, Maryland.

ROY (Caller): Yes. I've used the new machines and have studied them. And, of course, have a new book out called The History and Politics of Voting Technology. But…

CONAN: Ah! This is a promotional call.

Roy: No. It is, but I want to tell you about privacy. The subject came up very recently in an article by Michael Hill on the difference between lever machines and the newer machines. Now if you know - if you've used lever machines, you know there's a curtain behind them and that protects the privacy of the voter. With these new machines, they are - in Maryland certainly - they are squunched(ph) together and nobody is paying attention to how the machines are placed. And one voter can look over the shoulder of another voter and see what they're doing, because there's no curtain and they're not paying attention to their requirements of privacy in state law.

CONAN: Is that a reasonable concern, Kimball Brace?

Mr. BRACE: It is to a certain degree. It really depends on the polling place, and whether or not there is enough room in that polling to take those booths and spread them out, so that you don't have them next to each other in a privacy type of concern. So it's really a function of how the machines were set up. But I think it's going to vary in every single one of our 180 to 185,000 precincts that we have in this country.

CONAN: Mm-hmm. And why was it that lever machines are being phased out?

Mr. BRACE: Well, lever machines are old gigantic gorillas, as they were called. They weigh an enormous amount. They're very expensive to move between the precincts, to truck them out of the warehouse and into the precinct, and they're old technology. The little odometer wheels on the back of them, like what we have on the car, if you're not careful with them the little pins on those could break off and so the odometer reel would go one to nine, one to nine, one to nine and never get to ten, and you wouldn't know that, unfortunately.

CONAN: Let's bring another voice into the conversation. The recent primary elections expose challenges that poll workers will face in two weeks when voters show up to use all of this new technology. Election officials been working hard to train workers to avoid a repeat of many of the problems we saw on primary days. Paul DeGregorio is chairman of the U.S. Election Assistance Commission and he's also with us here in Studio 3A. Thanks very much for being with us.

Mr. PAUL DEGREGORIO (U.S. Election Assistance Commission): Thank you, Neal.

CONAN: Now the Election Assistance Commission's been working hard to help prepare voters - excuse me - workers for the upcoming election. Two weeks away, are people going to be ready?

Mr. DEGREGORIO: Well, Neal, we have reports from all over the country that many jurisdictions are still short poll workers. They mainly rely on volunteers and it's about 1.5 million Americans that are needed to conduct an election - a nationwide election - in this country. And they're still short poll workers in Chicago, in Salt Lake City, in Maryland, in many jurisdictions, and these are good people that do a great job to make democracy work on Election Day.

Many jurisdictions have worked very hard to institute new training procedures. When you have new sophisticated equipment, the type that Kim just talked about, you need new sophisticated training techniques to make sure that people understand how to use this new technology. But it's not just about the technology. It's about the various rules - whether it's the Help America Vote Act requirements or state law - that these folks are required to know and to implement in a perfect way on Election Day. And in many cases we see that perfection is hard to achieve.

CONAN: And people are also accustomed to the way they have been doing things for many years. Change is hard.

Mr. DEGREGORIO: There's many jurisdictions around the country, Neal, that have been used to punch cards for 30 or 40 years, and now they're transitioning to electronic voting. And we saw in the primaries this year, some problems - some problems with poll workers not knowing how to start the equipment, not knowing how to get the results off the equipment. So, it's been a difficult transition. We've seen this before, though - in 2002, 2004 - when Miami made transition from punch card to touch screen machines, and I think we're going to see similar problems this year. But we at the EAC have given out a million dollars in grants over the last two years to recruit a new generation of poll workers, and that's college students.

We've given out these grants to many institutions - colleges and universities across the country - to get a new generation of poll workers out there to deal with the new equipment and to help make America's democracy work.

CONAN: Here's an e-mail we got from Jim in Hyattsville, Maryland: Worked as an election judge on September 12 during the primary at Age 61. I was the youngest of the ten judges at my precinct. We were not the most computer savvy group. While I'm rather confident there was no intentional fraud, at the end of the day our electronic poll books were out of sync by about 20 votes. Not enough to swing an election but still not enough to make me completely confident. Our precinct was assigned a technician who arrived late and missed the prior evening set-up meeting completely. This technician helped us but she was not that much better trained, and toward the end of the very, very long day she was more concerned about getting home to her baby than she was about methodically closing down each of the machines by following the many, many, many steps.

Mr. DEGREGORIO: Well, I'm sure that kind of e-mail can be received from many jurisdictions around the country. You know, this again is a human exercise and election officials are working very hard to find people. I know in Chicago they're hiring people there to be technicians for the day and they're paying them $500.00 a day to do that, and they're having trouble even recruiting people to fill that function.

It's just very important that the poll workers be trained as best they can and that we recruit - we do better recruiting in poll workers across the country to get a new generation. But there's a lot for them to know. And I always urge the legislators who are thinking about changing laws to make sure they understand how it's going to impact the poll worker who has to deal with these changes.

CONAN: All right. When we come back from a short break, many more of your calls. 800-989-8255, if you'd like to join us. 800-989-TALK. Our e-mail address is talk@npr.org. I'm Neal Conan. You're listening to TALK OF THE NATION from NPR News.

(Soundbite of music)

CONAN: This is TALK OF THE NATION. I'm Neal Conan in Washington. Two weeks before Election Day we're talking about changes in voting technology that many of us will face at the polls come November the 7th. Will high-tech machines solve the problems of earlier elections or just create new ones? We have answers to many of your voting questions from absentee ballots, from what ID to bring to the polls at npr.org.

Our guests here in the studio: Kimball Brace, president of Election Data Services, and Paul DeGregorio, chairman of the U.S. Election Assistance Commission. Of course, you're welcome to join us: 800-989-8255, 800-989-TALK. E-mail is talk@npr.org, and let's get some callers on the line. This is Joe. Joe's calling us from Tennessee.

JOE (Caller): Thank you. We've had a situation here - I'm a computer person and I'm very familiar with the GEMS Diebold central tabulators. Neal, these systems are so insecure it's an absolute joke. We've already got a lawsuit going in Nashville, or the Shelby County area, that was brought by several local candidates. They found illegal software, PC Anywhere software, to move the databases around on a thumb drive, a USB thumb drive.

I appreciate what Mr. DeGregorio is doing with the EAC. I'm sure he's familiar with the (unintelligible) his statement that he's come out with. There was no guidance with these electronic voting manufacturers. In many cases they are literally in bed with the local election officials and the election officials become in a position where they're defending the machine instead of listening to the people that are bringing up totally legitimate security concerns. I've got to tell you, I've got zero confidence because we would never know if there's any fraud perpetrated. These are computers. It's like changing totals on a spreadsheet.

CONAN: Kimball Brace, has Joe got legitimate concerns?

Mr. BRACE: Well, there's a certain degree of concern, but you have to take a look at procedures that election administrators use. There's a lot of different technologies…

CONAN: Joe, let him talk, all right?

Mr. BRACE: There's a lot of things that the local election administrators do. The main thing - the logic and accuracy tests that are run before the election to make sure that they systems are counting properly, that are recording votes properly, so there's a lot of different procedures. Yes, I've seen - I've been out and looking at elections in more than 300 jurisdictions in the country in the last 30 years. You do see some degree of vendors providing a lot more assistance to local election administrators than you'd like to see, but local election administrators are usually strapped for cash for dealing with all these things. After all, election administrators tend to be in the bottom of the barrel when it looks towards improvements in the system. That's one of the great things about HAVA - is that the federal government did finally come through with money.

CONAN: HAVA, the Help America Vote Act.

Mr. BRACE: The Help America Vote Act.

CONAN: Okay, Joe.

JOE: First of all, the logic and accuracy testing is no good if you've got a virus that was put on the touch screen machine, as the Princeton studies just showed you can do. The virus knows when you're doing logic and accuracy as opposed to a manual election. So, what you're really pointing out, actually, Mr. DeGregorio, is the fact that…

CONAN: That's Mr. Brace, but go ahead.

JOE: I'm sorry - that the computers have actually created more problems and there is no guidance at all from Washington on this. The voting manufacturers are not required to open up their source code. We have no idea what's going on, but the issue with poll workers is just compounding the problem. We should not be using computers without paper to tabulate votes.

CONAN: And he's using a lot of words that I don't understand, Kimball Brace, but the last one I do understand, that a lot of people simply just don't trust this equipment - that it's too easy to tamper with.

Mr. BRACE: That is the claim that we're hearing all around the country, and a lot of people are talking about that. But you look and you talk to election administrators, they're the ones that are dealing with these equipment - day in, day out. They - many of them do have access to the source code. It's not some source code that is available for everyone out there, but it is being made available for the people that totally need it for their elections process.

CONAN: Here's an e-mail we got from Linda in Yucca Valley, California. I've been a poll worker in California for two years. We have to go to a training session every election. They give us a 60 plus page booklet with step-by-step instructions. We have roving supervisors that come by and help with any questions in terms of the electronic voting machines and how to use them. We've not had any problems. That must be an unusual comment there, Mr. DeGregorio.

Mr. DEGREGORIO: Well, maybe not that unusual. But I think that, you know, here you see somebody in California that's pleased with the way that they system's running and the gentleman in Tennessee who has concerns - and there's plenty of people in the country that have concerns. I mean, we certainly recognize that, you know, people need to get engaged in this subject with their local election officials. The Election Assistance Commission has sent out a management guide to every election jurisdiction in the country to make sure they have a check off list to make sure these systems are secure, and we encourage them to be transparent. We encourage them to have people like your caller from Tennessee in the room when they do the testing, and to ask these important questions.

CONAN: Let's talk with Warner. Warner calling us from Ohio.

WARNER (Caller): Yes, how are you?

CONAN: Very well, thanks.

WARNER: In Ohio we are very concerned about the electronic voting. And in fact, here in Summit County we are discussing an initiative petition drive to put on our ballot that in our county we have paper voting because of the concerns with these machines. Additionally, I think that paper voting is necessary to restore integrity to our process. People simply do not trust what's happening with these electronics.

CONAN: Kimball Brace?

Mr. BRACE: Well, there's one thing that also needs to be said about paper. Most of the studies have shown that a paper ballot is more apt to be voted incorrectly than an electronic system. And it's one concern that we've had for a long time that people - the American public has a marvelous way of fouling up their ballots in the most unlikely ways. Election administrators have seen this all across the country. So the key for all of this kind of voting is proper training, making sure that the voters know how to vote. And that's one key responsibility for voters, particularly those that are going to be voting absentee, because the thing that we've seen in studies after studies is that absentees and where ballots are sent to a central counting system have five times the error rates that precinct counts do.

CONAN: And Paul DeGregorio, there are some groups of voters for whom electronic voting poses actual advantages.

Mr. DEGREGORIO: There's no question about it, Neal. Not just your average voter who in touch screen voting can make the font larger or choose the language that they want to vote in, but certainly people who have a disability will be empowered this year, really for the first time, throughout the nation because the Help America Vote Act mandates this year that people with disabilities should have the ability to vote privately and independently for the very first time, and I witnessed this across the country and visiting ten primaries and I've seen people cry after they've voted for the very first time in private, and this is a new empowerment. This is something that's, you know, a really good benefit for the people of our nation.

CONAN: Warner, despite all of that, I suspect you're not convinced.

WARNER: Absolutely not. I think that it's false to say that paper ballots create errors. Paper ballots is a design issue. If they are designed properly they are more secure and they are more accurate more than any other form of voting. And I think that bears itself out. We see that in other countries.

We have seen huge errors in the electronic equipment here in Ohio. And there are a lot of concerns about flipped votes. We've had testimony, taken public testimony in Cuyahoga county.

There are also concerns in Ohio in that we have a law that allows observation of the tally, and you cannot observe the electronic blips. It's that simple. You can observe paper ballots being counted. And we want to see that in this state.

CONAN: Warner, thanks very much for the call. We appreciate it.

WARNER: Thank you.

CONAN: And you can cite, Mr. Brace, all the studies you want. You can point to all the evidence that has been accumulated. People are incredibly suspicious of this technology because they can't see it, they can't touch it, and they fear it's being tampered with, both by mistake and by design.

Mr. BRACE: Yes, that is the major concern that a lot of people have. They need to be involved. They need to be involved with their local election administrator. To be down there, as Paul was saying, and watch the logic and accuracy test. To be there, first hand, to test out these devices. As the Election Assistance Commission has said, there is a lot of need for people to be involved in this process - more than just being rabble rousers and bomb-throwers, or whatever the case may be. But to be actually…

CONAN: Or more likely lawsuit-filers.

Mr. BRACE: Or lawsuit-filers, or whatever. But to be a poll worker. To go in there and work the operation on Election Day, to really see what goes on. There's a lot of things - I've been a poll worker - that people just don't realize - the magnitude of everything that takes place on Election Day.

CONAN: Paul DeGregorio, we wanted to thank you for coming in today, we appreciate your time. Paul DeGregorio is chairman of the U.S. Election Assistance Commission and he was kind enough to be here with us in Studio 3A.

Although the switch to more advanced voting technologies is unusual for its scope and speed, this is not the first time the nation has experienced growing pains with how it records the vote. Joining us to give some history on the changing technology of voting is Douglas Jones, associate professor of computer science at the University of Iowa, and the former member of the Iowa Board of Examiners for Voting Machines and Electronic Voting Systems, he's with us today from the studios of member station WSUI in Iowa City. Nice to have you with us today.

Professor DOUGLAS JONES (University of Iowa): Well, hello.

CONAN: There are a lot of fears, as we mentioned, about switching to touch screens this year, before. Tell us about the lead up to the switch to the lever machines back in the late 19th century.

Prof. JONES: Well, let me start by reading a media report and let you guess when it was written. This takes about a minute. The arrival of voting machines in America's polling place has not been greeted with universal enthusiasm, especially voting machines that leave no tangible record of the voters' intention.

Among the new devices, supporters are advocates for the disabled who are more likely to have their votes counted correctly when the machines are used. Because the new machines produce no paper printout, they keep ballots out of the hands of local and highly partisan election commissioners and local judges who are known to excel in a variety of recount games.

In response, the chairman of the Election Commission in the state's largest city voiced concerns about the technology, pointing out that the machines record votes directly in a way the voter is unable to verify. He insisted that if machines are used they should print a copy of the voter's ballot, so that it can be reviewed by the people, the candidates, and the courts when objections are made.

Democrats charge that some of the devices have been programmed to deliver two Republican votes for every one that was actually cast. A nationally known expert on election security meanwhile provided public demonstration of the ease with which voters could sabotage the machines to alter vote counts.

So when was that written?

CONAN: I'm guessing 1876.

Prof. JONES: It's from 1925.

CONAN: Okay.

Prof. JONES: And it's describing the controversy over the adoption of mechanical voting machines in New York City. But in fact this all started in the mid-1870s. Rutherford Hayes and Samuel Tilden…

CONAN: Which is why I mentioned it, yeah.

Prof. JONES: Yes. That election was a controversial as the election of 2000. The electoral vote went one way, the popular vote went the other way. In the same decade, Boss Tweeds' Tammany Hall Organization was exposed to had been incredibly fraudulent in their conduct of elections. And (unintelligible) the media…

CONAN: All over paper ballots?

Prof. JONES: Oh absolutely. But not with paper ballots the way we know them today. These were paper ballots printed by political parties and distributed outside the polling place to voters. But almost immediately in the wake of this controversy and even as the controversy was hitting the press, inventors began to come forward with all kinds of neat new ideas. Improved paper ballots, mechanical voting machines.

And the inventors were clearly motivated by political reform. Writing things like: it's well understood by those who have given the subject study that the system of open balloting, now in use in most of the United States, is open to serious objection through the liability that the will of the people may be defeated by fraud and bribery.

So serious has this evil become and so great the danger of corruption of the ballot as to lead to much discussion of the subject and a universal demand for the passage of laws to ensure the purity of the ballot.

Now the interesting thing is that's from a patent issued in 1889. You don't often get political reformist rhetoric in patent applications. They're usually as dry as bones.

CONAN: We're talking about new voting technology, and you're listening to TALK OF THE NATION from NPR News.

Now it turns out from time to time that fears about rigging lever machines may have been justified.

Prof. JONES: Oh yes. I think that the reformists pushing for mechanical voting in the 19th century, they were right that paper ballot elections, as they had been traditionally conducted, were seriously flawed and were subject to massive rigging in some countries. There are well-documented reports of 30 or 40 percent of the ballots being counted being fraudulent in some counties.

And two threads of reform emerged simultaneously. One thread pushed for mechanical voting because you can trust machines when you can't trust people. The other thread pushed for improved procedures and policies that allowed you to conduct a trustworthy election on paper.

In a way both threads of reform won. Because when, for example, New York finally went from paper to mechanical machines, the major election upheavals that people expected were they hoped that the Democratic machine would be thrown out and the Republicans would be swept into office by the first honest election ever.

That didn't happen. Things were pretty much the same, which is to say that the advocates of reformed paper balloting had done a pretty good job of reforming the accounting process on paper, and the machines weren't that different.

CONAN: It also suggested no matter what the technology, nothing is foolproof.

Prof. JONES: That's right. It turns out that the machines, while they were resisted incredibly by the political machines in the various states that had strong political machines, the mechanical voting machines, eventually people figured out how to rig them. Typically rigging the straight party lever to cut out candidates who the local organization didn't like.

In a sense they replaced the risk of manipulation at the precinct by corrupt election workers at the very local level with the risk of manipulation by people in the county building. Typically the technician who was charged with programming the machine for the election and setting things up.

CONAN: Let's see if we can squeeze one more phone call in. I'm afraid we're running out of time…

Prof. JONES: Yes.

CONAN: So Mark. Mark in Charleston, South Carolina. If you could make it quick.

MARK (Caller): Yes. Are the machines individually pieces or are they interfaced with one another?

CONAN: The voting machines today?

MARK: Yeah. Are they separate or they interface with one another when they can tally up the vote?

CONAN: Kimball Brace.

Mr. BRACE: They're usually separate. On election night, in particular like electronic systems, they may be daisy-chained together in the precinct but they're not daisy-chained back to he central office, that sort of thing. The cards from the machines are taken out and either physically taken back to the county courthouse or transmitted electronically but in a separate device - that way.

CONAN: Does that answer your question, Mark?

MARK: Yeah. Thank you.

CONAN: All right. Thanks very much. And I guess the concern there is, again, rigging but this time by hacking. And as we're hearing form Douglas Jones is not a new concern.

Mr. BRACE: It's not a new concern but keep in mind that what Doug was talking about took many, many years for things to change. We're seeing so much change going on immediately right now. I mean the Help America Vote Act basically empowered everything that happened within a short six-year time period.

CONAN: Going to be a busy couple of weeks for you.

Mr. BRACE: Absolutely.

CONAN: Thanks very much for being with us. Kimball Brace, president of the Electronic Data Services. He was with us here in Studio 3A. Douglas Jones, thank you for your time as well. Douglas Jones, associate professor of computer science at the University of Iowa. With us from member station WSUI in Iowa City. When we come back, the life and death of Jonestown.

It's the TALK OF THE NATION from NPR News.

Copyright © 2006 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.