Iowa Voters

Wendy Barth, Green Party candidate for Governor, and her running mate Richard Johnson, both more knowledgeable about computers than most Iowa county auditors, have jointly rebutted the comments of Linn county auditor Linda Langenberg:

In a recent news article Langenberg referred to those who raise concerns about the accuracy of electronic voting machines as “terrorists.”

“As a career software engineer, I understand just how easily errors in programming code can be overlooked,” Barth responded, “and how difficult it can be to thoroughly test a piece of software this complex. In addition, it has been demonstrated by several testing groups that these machines can be compromised rather easily. Given this the concerns of the public are legitimate, and are deserving of consideration by the Auditor. I understand that Ms. Langenberg feels she deserves to be trusted, but the voting public deserve more reassurance than a simple ‘trust me.??

Barth’s running mate, Richard Johnson, added, “Having spent the last fifteen years of my career as a network engineer I can assure you that the risk of intrusion into these computers is real. There are documented instances where the computers controlling these voting machines have not only been “hacked,” to use the popular term, but where results have been changed without the operators being able to detect it. I cannot understand someone dismissing these threats so lightly.”

Recent testing by Princeton University
(http://itpolicy.princeton.edu/voting) has demonstrated how vulnerable these computers are, and how easily intrusion into these computers can be hidden. There have also been numerous reports of errors from these voting machines reported in Ohio, Florida, and Georgia. A number of states and counties are requiring that the machines have a verifiable paper trail that includes a receipt for the voter and a printed verification report.

At a meeting of supporters in Cedar Rapids earlier today, Barth announced, “Our campaign is calling for an independent audit of the code used to program these machines by qualified software engineers, as well as a real paper trail with receipt that can be verified by the voter as they cast their ballot. Every electronic cash register in the state prints receipts, so why not our voting machines? It is inexcusable that the gambling equipment in our casinos, which must pass regular independent inspection, is subjected to a higher level of scrutiny than our voting machines.”

Drake blogger Chris Woods has looked at the spending and fundraising by candidates for Secretary of State. His report is here.

Linn County auditor Linda Langenberg lashed out at voting machine critics in Tuesday’s Cedar Rapids Gazette, saying they engage in ” a form of terrorism.” She went on to add, “It’s making people lose confidence in the voting system without getting the other side from voting officials.”

Imagine that: We are unable to hear the government’s side of the argument.

Lucky for Langenberg, Gazette writer Dick Hogan reported ONLY the government’s side in his story, referring to criticism of the voting machines as “stories and rumors”. Rumors to him, maybe. He should read this blog. Or those linked on the sidebar of this blog.

And what was the Langenberg side of the story?

— That “electoral votes are never on the voting machines” and therefore it was foolish for the Diebold CEO to pledge to deliver Ohio’s votes to Bush in 2004.

— That her machines are never connected to the internet and therefore they are safe.

— That she doesn’t even use the machine modem to report results over the phone from polling place to courthouse.

Langenberg’s reassurances carry exactly as much weight as you would expect from someone who bothers to say that electoral votes are not on the voting machines.

Langenberg has not been keeping up with the news. Her Diebold ballot scanners were hacked in Florida last December, and her Diebold touchscreens were hacked by Princeton computer scientists this summer. On the 4th of July Blackboxvoting.org published complete directions on how to commandeer a Diebold touchscreen. None of these attacks required use of the internet.

Langenberg asserts, “The whole key is access (to the machines),” which she promises to strictly control.

No thanks, Ms. Langenberg. The whole key is public participation in vote counting. No one can see the machine count votes, so keeping it locked up is not reassuring to us critics.

And what is with this “terrorism” charge? Just yesterday on National Public Radio a member of the federal Election Assistance Commission said voting machine critics were “bomb throwers.” Such name-calling shows the defenders of paperless voting machines have little left to say for their products. So they resort to slander.

Which is better—-a well understood way of counting votes or a proprietary, secret way of counting? Seems like an easy choice to me. But our candidates for Chief Vote Counter are willing to accept secrecy, as we learn from their answers to part 5 of the Iowans for Voting Integrity survey:

5. Currently, voting machine software is a proprietary secret, and cannot be examined—even by elections officials. The only examination of software is by the Independent Testing Authorities, which are selected and paid for by the vendors and therefore may have a conflict of interest. Leading computer science experts and voting integrity groups are calling for public disclosure of voting machine software, which would still preserve patent and copyright protection. Where do you stand on this issue?

Hanusa:

-The public disclosure of software, while still preserving patent and copyright protection, is not always possible, and may not be desirable. Making software used in electronic voting systems available to the public would make it easier for hackers to write harmful viruses and other malicious programs. Careful review and continued discussion of this issue is necessary. The sort of checking processes called for in HR550 may help here.

Mauro:

This is another reason why paper trail is imperative.

Both answers indicate acceptance of secret vote counting software. Both will try to compensate for it, but neither wants to end it.

Hanusa falls for the fallacy known as “security thru obscurity”, which is discredited by computer security professionals. Secret programming is more likely to contain mistakes than publicly available programming. And why would she want the SoS to know less about how votes are counted than some temp worker at the vendor’s programming sub-contractor?

Open source works. It works for Firefox browsers. It works for Linux. (Who hacks Linux?) I can no more alter an open source program than I can alter the morning newspaper after it is printed and made public. But I can sure see the errors! (Thanks to Justin Moore for the comparison)

More than one voting machine company has already told California that they will reveal their software IF it becomes state law that they do so. I don’t think they fear giving any help to hackers by making their wares public.

So this is the biggest disappointment in the questionaire.

Let’s wrap it up with the last two questions:

6. The DRE touchscreens were advanced by HAVA in the interest of making the voting process accessible to citizens with impaired physical or visual abilities. Given the existence of alternative equipment for accessible voting that is based on a paper ballot (such as Ballot Marking Devices or the Vote-PAD system) do you see any need to abandon paper-based voting to accommodate the disabled?
Hanusa:

No.

Mauro:

No I do not. Currently in Polk County people with disabilities us the automark machine, which still provides a paper trail. The abandonment of a paper trail does not need to occur to accommodate the disabled. The integrity of the system should be maintained for every voter and requiring a paper trail is the only method to do this.

7. Iowa recently announced a program for email voting for overseas military personnel. In this plan, voters sacrifice the secrecy of their ballot. In 2004, a similar Pentagon plan was roundly criticized by voting security experts[2]. What is your view of email voting?

Hanusa:

E-mail voting sacrifices the secrecy of the ballot(s), and also has the potential for misuse, manipulation, and other types of fraudulent activities. The current technology is not yet sufficient to supply the level of confidence and security that voters need for elections, but as technology changes, further review and consideration should be given to this matter.

Mauro:

The implementation of this process is an additional option for overseas military personnel who are burdened by location and military duty and may not have adequate access to normal absentee processes. As Secretary of State I will continue to encourage every Iowan overseas to use normal absentee methods to ensure ballot secrecy, however when service men and women located in combat zones lack the ability to request the normal process, these individuals deserve additional access to the fundamental right of voting.

There you have it, folks: The best coverage of the Iowa Secretary of State’s race you will find anywhere. You have two weeks to ponder your choice. Vote by November 7th.

Iowans for Voting Integrity have questioned the candidates for Secretary of State. The first two questions were in yesterday’s post.

Today we see both candidates being careful not to offend the private corporations that have usurped vote counting in our elections. While Hanusa takes this as a given and merely urges “oversight” of the process by our public officials, Mauro actually wants to increase the public’s authority over elections so that we can hold the corporatcrats to high standards. What a sad state of affairs we are in!

Question #3. With the advent of new electronic voting equipment, auditors and other elections officials have become increasingly reliant on the private sector to supply the machines, the training, the technical support, and often the ballot configuration files. Do you feel this is a satisfactory situation, and if not, what can be done to improve it?

Hanusa:

We have relied on contractors and vendors from private companies for many years. The source of the voting equipment is not so much the issue as is its use. County auditors should be assisted by the Secretary of State’s office to make sure procedures are proper and that there is careful oversight by the local election officials(emphasis added by JD) in the process. Trained and vigilant election workers are key.

Mauro:

Currently there are two vendors that provide equipment and ballots to County Auditors in the State of Iowa. This makes it very difficult for these auditors to maintain control over expectations because the competition is so limited.

During the current election cycle there were roughly 20 counties that did not receive their ballots on time due to delays from the vendors. I plan to require state level contract provisions that would grant the auditors more authority over their own election operations. (emphasis added by JD) These contract requirements would require companies to deliver ballots on time, meet statewide standards and laws, and ensure that there would be uniformity of performance across the state and that county auditors would not be at the mercy of a small number of vendors.

Since elections officials are more reliant on these private sector entities to supply the machines, the training, technical support and in some cases the ballot configuration files, it is imperative that we hold these companies to high statewide accepted standards to ensure that the integrity of our election process in the State of Iowa remains intact.

Next up is the question of election auditing: that is, counting by hand/eye all the ballots in random precincts to see if the scanner or touchscreen tallied the ballots correctly. Hanusa oddly endorses this with a one word answer, despite not mentioning the need for any paper trail at all in question #2. Mauro (perhaps inadvertantly?) fails to answer the question directly, even though he endorses “anything” that helps ensure the integrity of elections

4. Proposed federal legislation (HR 550, the Voter Confidence and Increased Accessibility Act) would require random audits of two percent of precincts, to validate that votes are being accurately counted electronically. Would you support a similar measure at the state level?

Hanusa:

Yes.

Mauro:

I fully support H.R. 550 because it amends the Help America Vote Act of 2002 to require a voter-verified permanent paper trail. I support anything that helps to ensure the security and integrity of the process. Most importantly, I believe testing needs to occur prior the election, including sign offs and certifications from election personnel managing the election process. All testing certifications should be should be made available to the public for inspection.

But don’t get your hopes up, dear voter/reader. Tomorrow we get to question #5 and see that neither candidate endorses a perfectly good idea that would “ensure the security and integrity of the process.”

Secretary of State candidates Mary Ann Hanusa and Michael Mauro have answered seven questions put to them by Iowans for Voting Integrity. Having read through them just once, my first impressions are that both have resigned themselves to the use of vote counting software that is a corporate secret (!) and that Hanusa missed a chance to say she supports paper trails. Hanusa is cool with the corporate takeover of election administration whereas Mauro is more critical.

But don’t take my word for it. You can read the whole thing yourself. Here’s the first question that was posed to the candidates:

What do you see as the biggest challenge for the next Iowa Secretary of State, with respect to administering elections?

Here is Hanusa’s reply:

Balancing the public’s desire to use available technology to generate fast and accurate election results with the need to generate complete confidence in the entire election process. The voters need to be able to trust the technology, and part of this trust will be built on explaining why they can have confidence in the particular system they are using, in particular, what safeguards are in place to ensure accuracy and fairness.

Here is Mauro’s reply:

There are many challenges facing the next Secretary of State. The implications of policy decisions made with the Help America Vote Act (H.A.V.A), have changed the requirements of the office across the country. The biggest challenges lie with the election equipment and additional human capital to support the equipment. If elected Secretary of State, I think it is important to maintain the integrity of our elections by making sure every vote counts. I also want to make voting efficient, easy and accessible for all eligible Iowans.

I don’t like Hanusa’s talk about “trust”. Elections are a place for checks and balances, the glare of publicity, and public skepticism. No trusting types need apply. That’s why Iowa law requires every poll be staffed by members of two different political parties. But we now we ask those pollworkers to stand by in ignorance while we let the actual vote counting be done with software written secretly by anonymous distant corporations whom we are now asked to trust, but not to question.

Here’s the second question:

A recent Princeton University study[link] found that Diebold TS touch screen machines were vulnerable to dangerous security attacks. To quote, “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.” These machines are close cousins of the TSx machines being used in Iowa. What are your views on the use of these machines?

Hanusa:

The key to continuing to use existing machines is limiting the access (other than to voters) to a trained, technically savvy, and vigilant staff who can maintain and monitor the machines, and so prevent misuse.

Mauro:

This is a great question and it is imperative to understand the facts and then focus on the policy decisions that need to be implemented to ensure the integrity of our voting system. First, we must require a paper trail regardless of the equipment that is used. Without a paper trail any manipulations of the system will go unidentified.

Princeton computer professor, Edward W. Felten, wrote in his paper, “Security Analysis of the Diebold AccuVote-TS Voting Machine,” that there are real dangers of the election system being hacked by simply accessing a machine’s memory card. This type of manipulation of the system is a reality in this day in age and should be planned for in case such an atrocity occurs. Paper trails are the best way to catch any manipulation of the system. Electron Diebold Voting Machines cannot be used without paper trails due to the many security risks that exist according to Felten’s paper. Whether the Diebold Voting Machine or an optical scanner is used, a paper trail is necessary so the voter can confirm their choices. As with everything, there is always human error, but by requiring people to review their ballots, it ensures that any mistakes that occurred, any manipulations of the system, or any irregular occurrences will be identified and dealt with almost immediately.

So there it was: a chance to endorse paper trails. Mauro seized it, but Hanusa thinks everything will be fine if she can hire only “trained, technically savvy, and vigilant staff.” I wonder if that includes the pollworkers. “Technically savvy” does not describe the folks who last worked my local polling place.

Besides, the point of the Princeton hack is that it’s virtually untraceable. Hanusa needs to study up!

Five more questions to go. Come back soon to read the rest.

Voters were turned away from polls in Maryland during this summer’s primary election when touchscreen voting machines were inoperable. The same thing happened in California in 2004. The possibility that it will happen in November is so ominous that some slow-moving Congress critters introduced a bill just days before the recent adjournment to address the situation–like they thought there was time to make changes before November!!

The bill would have required every precinct to have emergency paper ballots so that no voter would be turned away by machine failure.

Welcome to Iowa.

We already have such a law, as recently pointed out by Doug Jones. Here it is, from the administrative code,
721–22.431(52)

Temporary use of printed ballots in voting machine precincts. The county commissioner of elections shall furnish a supply of printed ballots to each precinct where voting machines, including direct recording electronic machines, are to be used for any election.

22.431(1) Conditions under which paper ballots shall be used. In any precinct in which voting machines are designated as the only method of voting for any election, a paper ballot shall be furnished to any person offering to vote, in addition to those provisions set out in Iowa Code sections 49.81 and 49.90, if:

a. A power failure prevents use of the voting machines.

b. A malfunction occurs which prevents the use of one or more voting machines.

c. It is found that any voting machine has been prepared with all or part of a ballot strip meant for another precinct.

d. It is found that the ballot strips for any voting machine have been misprinted.

e. Any other condition exists due to a fault of the voting machine or machines which prevents the person or persons offering to vote at that precinct from casting their votes.

So the 18 Iowa counties that use only touchscreens still must have enough paper on hand to prevent the embarrassing messes we saw in the Maryland and California cases. Actually, it should be easy enough to use the leftover absentee ballots that every county had to prepare anyway.

My auditor is always repeating that Iowa has good election laws. Here’s a feather in her cap. But I wonder if she had paper ballot backups on hand for the September school board race when she used only the touchscreen machines and forsook her paper ballot scanners .

UPDATE on 10/10/06: Press account of the event is here.
************************************
The Ottumwa League of Women Voters and Indian Hills Community College are co sponsoring a program on “The Trials and Tribulations of Electronic Voting” on Monday evening October 9th at 7PM in the Advance Tech Building Room #134 on the IHCC campus.

The speaker will be Douglas W. Jones PhD., Associate Professor in the Department of Computer Science at the University of Iowa .

Dr. Jones has testified regarding voting system technology in several states since 2000 as well as the Federal Election Commission. His statements on problems with voting technology have been quoted by the New York Times, the Ft. Lauderdale Sun Sentinel, and Scientific American . He has been a guest of NPR Science Friday.

He participated in the 2006 revision of New York ’s voting system standards. He served on the Election Observation Mission for the 2005 presidential election in Kazakhstan .

The public is invited to attend this lecture and participate in the dialog following Dr. Jones presentation.

—LWV press release

Senator Kerry was in Iowa today campaigning for President for Michael Mauro. I heard Kerry on the radio this afternoon warning against paperless voting. He was citing the Princeton hack of Diebold’s paperless wonder.

I raised the paperless voting issue with Kerry myself in 2003 at a reception in Spencer one night after he had been to the Clay county fair. His answer was that I shouldn’t worry about election mismanagement because he planned to have attorneys at polls everywhere if he became the candidate in 2004.

Well, he did become the candidate and he did have a fleet of attorneys working for him. But his answer was ignorant–as if any attorney could tell whether the voting machine was working honestly or not! Typical lawyer arrogance, I thought.

Of course, at the time no Ivy League university had weighed in regarding voting machines. The cries of “Wolf” were coming only from the boondocks: U of Iowa, Johns Hopkins, Stanford, Rice—you know, not places Kerry ever heard of.

The Iowa City Press Citizen has a profile of professor Doug Jones, nationally known voting machine expert who helped win a court case in Colorado in late August.

An excerpt:

“We have to design technology that’s extraordinarily easy to use. Our current voting regulations don’t look at how complicated it is,” he said.

Aside from technology, Jones said the American election system is the most complicated in the world.

Thanks to reporter Hieu Pham in Iowa City.