Early on Election Day morning (at 1:16 AM) George Mason University’s e-mail system sent out a message to students under the Provost’s name stating that the Election Day had been moved to November 5. Here is the text of the message:

At 8:08 AM the Provost’s Office sent out a correction message titled “Urgent Voting Information":

As of Fall 2007 (latest available data on GMU’s website), total enrollment at GMU was 30,332. It is not clear how many students received this message.

Based on a very, very preliminary analysis by some of USACM’s members, it appears that the original e-mail was forged. E-mails can be forged if the SMTP (Simple Mail Transfer Protocol) mail server does not require adequate authentication. See CERT’s website on e-mail forging for more information.

Obviously this is a very serious issue and likely violates several federal and state laws. The Electronic Privacy Information Center just released a report on deceptive electronic campaign practices. This incident seems to fit squarely into the issues raised by that report.

Just to be clear, today, November 4, is Election Day for the United States. We encourage all eligible voters to get out and vote.

Saturday’s Washington Post detailed growing concerns and political battles over the implementation of new voter registration procedures as states use newly constituted statewide voter registration databases. One of the most difficult issues to resolve is the “no match” problem when state voter registration databases are compared against federal databases (such as the Social Security Number database) or other state databases. Matching problems occur when there are typos in entries, outdated data is used or there are just plain errors in the information. Depending on how tightly the state interprets no match problems, eligible voters might be dropped from voting rolls. While dropped voters are supposed to be notified, voters may not have enough of a window to file a protest if the automatic merging and purging occurs too close to an election.

This issue was raised in USACM’s 2006 report on statewide voter registration databases. The report made about 100 recommendations and spoke specifically to the no match issue:

According to the Washington Post article, 31 states are actively using their statewide databases for the first time in this election. This, combined with the anticipated heavy voter turnout, means the November election will be critical test of these systems.

In a recently released report, the Government Accountability Office (GAO) criticized the voting system testing lab accreditation program administered by the Election Assistance Commission (EAC) and supported by the National Institute of Standards and Technology (NIST). You can read the report online.

The voting system testing lab accreditation program is intended to certify the processes and standards of those laboratories that test voting systems. The main thrust of the GAO criticism is that the supporting materials - the standards and program manuals - have not adequately defined all the aspects of an effective testing regime. What standards and practices have been put in place are not sufficiently delineated to the point where they can be used repeatedly and consistently. Now the GAO acknowledges that the EAC has responded to the report by taking some necessary first steps, but more needs to be done.

Attention in the press and the public usually focuses on the more explicit face of electronic voting - the machines and the actual elections. But voting is a process, or rather, a series of processes, that start long before Election Day and involve certification and approval of every piece of equipment and process involved. Just as electronic voting equipment needs to be well designed, engineered and evaluated, so do the organizations and processes that approve this equipment. If you’re curious about how your state certifies its voting equipment, check with the state Secretary of State or state elections department.

Senator Harry Reid (D-NV) recently appointed USACM-EC member Barbara Simon to the Board of Advisors for the Election Assistance Commission. USACM had nominated her along with three other USACM members to serve on the board in the positions reserved for technical and scientific professionals. Our release on Simons appointment is below, and you can find our letter on the nominations here.

(more…)

Last week the Board of Advisors of the Election Assistance Commission (EAC) met over two days to discuss pending and new business. While this meeting (and most meetings of the EAC and its boards) covered many topics, the items having to do with the Voluntary Voting System Guidelines (VVSG) are worth passing along.

The staff at the National Institute of Standards and Technology was asked to investigate several technical questions related to the VVSG. They are:

Alternatives to Software Independence (SI)
Standards for Ballot-on-Demand systems
Impact of VVSG on vote-by-phone
Consequences of separately testing and certifying election system components
Impact of VVSG on early voting and vote centers
Alternatives to goal level requirements in the VVSG (requirements that state a goal but are not easily testable - if testable at all)

(more…)

Last week we started a series of posts highlighting our high-level comments on the the Election Assistance Commission’s Voluntary Voting System Guidelines (VVSG). Our first post focused on the important new concept – Software Independence. Today’s post focuses on a related new concept – the Innovation Class.

(more…)

Three weeks ago, USACM filed detailed comments with the Election Assistance Commission (EAC) on its draft of the Voluntary Voting System Guidelines. (Quick Review: These are proposed Federal standards for voting equipment, which are voluntary in nature but some states require manufacturers follow them.) Realizing that many people will not be able to read through our filing, we are excerpting portions of our high-level comments on some of key concepts in the EAC’s draft.

Arguably the most important change from previous standards is that all machines must be “software independent” to be in conformance with the standards. The VVSG defines this concept as follows:

This can be translated as the voting system has to have some system that allows for verification and/or auditing of an election result without relying on the voting machine’s software. USACM strongly supports this requirement and urges the EAC to retain it in the final document:

The concept of software independence may be one of the most controversial proposals in the new standards and is sure to come under fire. How the EAC decides to proceed with it may determine how voting systems will look for the foreseeable future.

We are doing a bit of catchup as we are behind on posting to the blog. On Monday USACM filed detailed comments with the Election Assistance Commission (EAC) on the draft 2007 Federal Voluntary Voting System Guidelines. Calling the standards a welcome step forward, USACM supported several important concepts in the draft – including software independence – and noted the improved accessibility and usability provisions.

So what are the Voluntary Voting System Guidelines? Some background might be useful to start.

The Help America Vote Act (HAVA) created the Technical Guidelines Development Committee (TGDC), staffed by NIST and chaired by its Director, to recommend Federal voting system standards to the EAC. These standards are used as the basis for certifying voting systems. While voluntary, many states have mandated vendors follow them.

Since HAVA was enacted the TGDC released the 2005 Voluntary Voting System Guidelines, which were an incremental update of the 2002 Federal Election Commission standards. The 2007 draft is a sweeping and fundamental change from the 2005 standards, incorporating several new important concepts such as software independence, innovation class, and open-ended vulnerability testing. Many of these concepts have been the subject of intense debate for the past year and were the focus of USACM’s comments.

The TGDC submitted the standards to the EAC last year, and the EAC has the power to modify, reject and/or adopt the standards. Monday was the end of the public comment period on the standards, and USACM commented on numerous parts of the draft. Below is a press release highlighting the major points and we encourage you to read the complete draft of our comments.

The EAC now has to process the reportedly 2000+ comments it received on the draft standards. In doing so, the EAC could adopt the TGDC draft as is, make changes, or reject the entire document. The first and the third scenarios seem unlikely, so we can expect the EAC to makes changes, but what they will be and in what time frame is anybody’s guess. Stay tuned to our blog for more coverage.

(more…)

HR 5036, a bill that would provide resources for jurisdictions to purchase paper-based voting systems for the 2008 elections, was voted down in the House on Tuesday. The bill was considered under suspension of the rules, which requires a two-thirds majority to pass. It fell short, 239-178. Bill opponents criticized the cost of the bill and claimed the program is redundant. USACM commented favorably on the legislation in a letter we sent to Rep. Holt earlier this year. With the 2008 elections quickly approaching, it would appear that the window of opportunity for this bill is closing fast. House leadership could reconsider the bill, but early indications are that this will not happen.

The Committee on State Voter Registration Databases of the National Research Council released an interim report last week. An ad hoc committee under the Computer Science and Telecommunications Board, and sponsored by the Election Assistance Commission, this committee is charged to identify and address issues and concerns associated with operating statewide voter registration databases.

The committee’s interim report is available online (registration required to read a free PDF file). The report is based on two public workshops and other work of the committee since mid-2007. The report is a compilation of recommended practices for populating and maintaining a statewide voter registration database. The report breaks its recommendations into two groups: those that can happen before the 2008 election, and those that would take more time. The recommended short term changes are:
(more…)

Last week the Committee on House Administration marked up HR 5036, the Emergency Assistance for Secure Elections Act of 2008. The bill, introduced by Representative Rush Holt (D-N.J.), would provide reimbursement for jurisdictions that would purchase paper-based voting systems for the 2008 elections. USACM commented on the legislation in a letter we sent to Rep. Holt in February. The legislation is consistent with our previous statements supporting transparent, accurate, and reliable election systems.

The bill was reported favorably out of committee by a voice vote, with amendments. The major changes included allowing retrofits and hand audits to qualify for reimbursement, and the amount of money available for reimbursement and for voting system research was changed to “such sums as necessary.”

It is unclear whether the House will consider the bill, given the short shrift given to Holt’s other voting legislation, HR 811. That bill was also favorably reported out of committee and then held from full consideration by the House due to issues in the Rules Committee.

Last night on the Chicago Fox affiliate, USACM Chair Eugene Spafford was interviewed as part of a story on the State of Illinois’ progress - or lack thereof - in complying with the Help America Vote Act (HAVA). You can watch the story online. Dr. Spafford appears near the end of the piece.

The Illinois Auditor General recently completed an investigation of the state’s progress in implementing HAVA requirements. Where developing and implementing a voter registration database is concerned, the state is far behind. The list fails to have a unique state identifier for each voter; a system to verify social security numbers; or a means to match names against incarcerated voters. For more details, watch the story, and read the full report.

In November the EAC began a 120 day comment period on the next iteration of the Voluntary Voting System Guidelines, which are the Federal standards for all voting equipment. Today the EAC extended that deadline by another 60 days, pushing it to roughly May 5. USACM plans to file comments.

To review and comment on the standards follow this link.

As noted in a previous blog post, this morning the Task Force on the contested Florida 13th District election met briefly to hear the final report from the Government Accountability Office (GAO). The GAO was asked to review the election machines (and previously conducted tests of those machines) used in Sarasota County in order to determine their role in the high undervote rate for that particular election.

In testimony this morning (available at the GAO website), the GAO indicated in their testing that they could find no evidence that the voting machines contributed to the undervotes. Since the scope of their investigation was limited to the machines, they cannot - and did not - make any conclusions about what did contribute to the undervotes. The GAO report does note that a paper trail would have made their investigations a lot easier by providing an independent means of confirming the accuracy of the machine vote.

As a result of this testimony, the Task Force voted unanimously to recommend that the Committee on House Administration dismiss the contest of the 2006 election in Florida’s 13th District. The full committee will act soon on this recommendation. How the public responds to this, in Florida or elsewhere, is anyone’s guess.

Yesterday USACM sent a letter to Representative Rush Holt commenting on his newest piece of electronic voting legislation, HR 5036. The bill, the Emergency Assistance for Secure Elections Act of 2008, provides money to jurisdictions that rely on DRE voting machines to purchase paper trail systems in time for the general election. At the same time, USACM indicated a need for further electronic voting reform, such as the reforms discussed in Rep. Holt’s other e-voting legislation, HR 811, currently stalled in the House.

From the letter:
(more…)

Congressional Quarterly is reporting that the Committee on House Administration’s Task Force on the Florida 13th Congressional District contested election will meet this Friday at 10 a.m. The Government Accountability Office (GAO) will present its final report on the election, and the Task Force will make its recommendation to the full committee.

The contest of this election hinges on the comparably high percentage of undervotes (where there are fewer votes for a particular race on the ballot compared to other races on the same ballot) in only one of the counties covered by this district. The State of Florida conducted tests, and the GAO has been investigating the election and the subsequent tests. Several different factors could explain the undervote issue, but so far the publicly available information about how the voting machines have performed has been, at best, inconclusive.

The meeting is scheduled for 1310 Longworth Building, Friday February 8, at 10 a.m.

In the debates about the use of electronic voting machines there is a refrain that direct electronic recorder (DRE) voting machines are much easier to use and much more accurate in capturing voter intent than other voting systems. A new comprehensive usability study of five commercial e-voting machines (published by the Brookings Institution) finds that we still have a long way to go in improving machine usability and that ballot design remains a challenging issue. We haven’t read the full study (Available from Brookings Press for $19.95), but here are some highlights from the news release and a story on MIT’s Technology Review.

While the study found voter confidence and satisfaction with the systems was good, error rate increased as voter tasks became more complex:

The study has raised concerns within the computing community because it seemingly downplays security issues. The press release on the study clearly plays up that there is some tradeoff between usability and voter-verification. I suspect this is more likely to do with trying to generate a controversial enough angle for the press than the authors belief that completely software dependent voting systems are worry free. In general, we think this “tradeoff” view is flawed. There are clearly major security and reliability issues that justify having software-independent verification systems, just as there are clearly usability issues that need attention. Attention doesn’t need to be mutually exclusive.

The key to improving voting technology is careful engineering, research into voting technology and voter behavior, and strong standards and testing. We are a long way from meet these objectives. More attenion on all these issues should be the goal for election officials, vendors, technologists, and policymakers.

We are starting a series of weekly posts taking a high-level look at some of the hot tech policy issues in Congress for 2008. The first is an area that we’ve already covered extensively – electronic voting reform.

Last year three big stories dominated e-voting issues: 1) controversy over how e-voting machines preformed during the 2006 election in Florida’s 13th Congressional District, 2) Representative Holt’s legislative proposal to reform e-voting, and 3) newly proposed Federal standards for voting equipment. All three issues will spill over into this year.
(more…)

This past Sunday’s New York Times Magazine has a great piece on the dynamics of electronic voting issues in the United States. The article gets a number of things right and is well worth a read.

First, it discusses the relatively recent trend of local election officials asking much tougher questions about the reliability, security and accuracy of electronic voting equipment. Just after the Help America Vote Act of 2002 passed, local elections officials bought complex new voting technology, but had little to no IT support staff. Vendors became the IT support. When something went wrong, election officials were often in an awkward position of trying to answer questions from the public by turning to the vendors that sold the machines. While vendors still serve as de-facto IT staff, election officials are now asking much harder questions about e-voting. During the last half of 2007 California had outside experts conduct a top-to-bottom review of voting machines, Ohio revamped its election procedures in troubled Cuyahoga County, and Colorado decertified several e-voting systems after a review.

Second, it puts the appropriate focus on technology failing in unexpected ways versus hacking. While computer scientists are often concerned about both, hacking threats always seems to get the headlines. These can be explained away with more security procedures, but often it is difficult to diagnose bugs that cause failures. Here is a good example from the story:

Lastly, it is an extensively researched piece and is balanced. It doesn’t come off as stridently opposed to using computing-based solutions in elections, nor does it pretend that there are no serious problems with current e-voting systems. Too often we hear the extremes – advocates that don’t want any technology in elections or vendors that pretend anyone who raises questions about e-voting technology is a luddite.

The only quibble about the piece is that it implies that testing of voting equipment is more rigorous than it really is. While the testing has improved of late, current standards still do not call for the type of open-ended vulnerability testing that reveals many security and reliabilty flaws.

If you are new to e-voting issues, want a good overview of what has happened for the past few years, or a student looking to write a paper on e-voting, this article is a wonderful place to start.

Yesterday the Election Assistance Commission (EAC) released it’s online comment tool for commenting on the next iteration of the Voluntary Voting System Guidelines. Released in August by the EAC, this version is the first full rewrite of the standards in some time. The current version of the guidelines (sans commenting capability) is available online.

The comment tool is organized so that you can make a comment for each specific piece of the VVSG within the document. Public comments can also be reviewed, though there will be a few days delay in posting public comments.

This begins the first of two 120 day comment periods the EAC will hold prior to finalizing the guidelines. After the first comment period ends (the EAC website lists no deadline, but assuming it started on October 31 - the release date of the comment tool - it should end around February 28, 2008) the EAC will review the comments and incorporate them into the guidelines - along with feedback from their advisory boards. They will then start the second comment period.

It’s a large document, so you can’t start reviewing it for comment too early.

With the House of Representatives e-voting reform legislation stalled, we’ve turned our attention to the Senate. Senator Feinstein (D-CA) is leading the effort to reform e-voting systems. USACM sent the Senator a detailed letter noting that her legislation largely gets the big things correct, but there are some important details that should be addressed. From the letter:

As with most things, the devil is in the details, and that is where USACM’s comments are focused. There is a lot of nuance and detail in USACM’s letter that is too long for a blog post, so I encourage you to read the supporting comments.

Senator Feinstein’s legislation is largely similar to what Representative Rush Holt (D-NJ) proposes in the House, requiring voter-verified paper records that are secret, durable, and accessible. Her legislation also makes the paper record the official record for the election, and it reforms the testing and certification processes. Finally, it requires software review by outside experts, with some differences in the details that are significant. (You can find a side-by-side comparison of the disclosure sections here.)

One of the biggest departures from the House bill is with the deadlines for reform. The Senate legislation bans the purchase of machines that do not produce voter-verified paper records upon enactment of the bill and requires the usage of paper ballots by the 2010 election. Representative Holt, for the most part, stuck with a 2008 deadline. The Senator has indicated some willingness to make that deadline even more flexible.

We have tracked the progress of the disputed election in the 13th Congressional District of Florida for months now. In short, the race was decided by fewer than 400 votes, and there were several thousand undervotes (no vote recorded for that race compared to others on the same ballot) in Sarasota County, only one of the counties represented by that District. Tests conducted by the State of Florida found no evidence of machine malfunction, but associated court cases are working through the appeals process.

Yesterday afternoon the Government Accountability Office presented testimony at a hearing of the House Task Force responsible for investigating this election, which was formally contested to the House of Representatives. The title of the testimony hints at the problems involved - “Further Testing Could Provide Increased but Not Absolute Assurance That Voting Systems Did Not Cause Undervotes in Florida’s 13th Congressional District.” The GAO acknowledges that ballot design could play a role in the undervote, but is charged with examining the machines used in this election.
(more…)

Update 9/18: Apparently the title for this post is still apt. This legislation made it onto the calendar for House consideration earlier this week, but it looks like, once again, it has been pulled. We don’t expect to see action this week. The delay apparently still revolves around claims of whether the legislation unduly burdens state and local jurisdictions.

Original Post 9/10: When I started writing this post last week, the House of Representatives had finally scheduled Representative Rush Holt’s (D-NJ) e-voting reform legislation – the Voter Confidence and Increased Accessibility Act (H.R. 811) for consideration. However, lingering concerns voiced by state and local officials and disability groups appear to have stalled the legislation’s consideration yet again. This latest delay is after the bill’s author made concessions – including more flexible deadlines, additional reviews of technology to determine accessibility and proposed additional federal funding - to address these concerns.
(more…)

Early this afternoon, the Technical Guidelines Development Committee (TGDC) formally approved its draft of the next edition of the Voluntary Voting System Guidelines. After some final editing and adjustments, they will forward this document to the Election Assistance Commission (EAC). The current draft is available online. Running at 578 pages and approximately 1170 requirements, this is the first top-to-bottom rewrite of federal voting system guidelines since 1990.
(more…)

On Friday, August 17, the Technical Guidelines Development Committee (TGDC) will meet by teleconference. This will be the Committee’s final meeting on the next generation of voting system guidelines. The meeting is scheduled to start at 11:30 a.m. Eastern time and run until approximately 5:30 p.m. A rough agenda is available one the same page as the webcast. The goal of this meeting is to approve the draft guidelines that the TGDC will submit to the Election Assistance Commission. You can read the current draft (dated August 7) online.
(more…)

Following the top-to-bottom review we posted about earlier, California Secretary of State Debra Bowen has changed the certification status of the voting systems subject to the review, as well as one system that was not submitted for the review. Bowen’s decision gives affected jurisdictions six months to adjust before the February 5, 2008 Presidential Primary.

You can read the formal decisions online. The affected systems are: (more…)

Update - August 1 - Senator Feinstein (D-CA), Chair of the Senate Rules Committee, has announced she will hold a hearing on this report sometime in September. This report *may* also come up in tomorrow’s scheduled hearing of the Elections Subcommittee of the Committee on House Administration, as the chair and ranking member are both from California.

Original Post - July 30

At 10 a.m. Pacific time today, The California Secretary of State’s office convened a public hearing on the results of its top-to-bottom review of three voting systems used in the state (while there are more than three different systems used in California, some are being phased out and one was not submitted in time for the review). It is still going on as I write this, and can be watched at http://calchannel.com (my connection is spotty at best, so the webcast may be overburdened by widespread interest).

This review was announced in May, and many of the relevant documents are available at the Secretary of State’s website. In short, the state contracted with the University of California to have separate teams review the voting systems, to include a “review of documents and studies associated with each voting system, a review of the computer source code each machine relies on, and a red team penetration attack to see if the system’s security can be compromised.” While all of the teams have submitted their reports, the Secretary of State is reviewing the document and source code review reports to see if they would disclose any security sensitive information. So all that can be reviewed today are the red team reports and the accessibility reports.
(more…)

With news that Representative Holt’s e-voting reform legislation has stalled (reported in New York Times) attention turns to the Senate this week as it holds a hearing on Senator Feinstein’s e-voting reform legislation. On Wednesday at 10:00 AM, the Senate Rules Committee will hold its first hearing on the Senator’s legislation S. 1487 – The Ballot Integrity Act.

The Senator’s bill is similar in many respects to Representative Holt’s legislation. However, it has some notable differences including a deadline of 2010 (instead of 2008) for replacing Direct Recording Electronic machines with ones that have durable voter-verified paper ballots, more comprehesive software disclosure requirements, and a more flexible auditing scheme for conducting post-election audits.

The hearing should be webcast. You can find the link here, and here is the witness list:

We have not heard when the Senate Rules Committee may take up Senator Feinstein’s legislation, but we don’t expect action before the August recess.

According to the Technical Guidelines Development Committee (TGDC) website, the plenary meeting originally scheduled for July 3 (and subject of an earlier post on this blog) will now take place on August 17. It will still take place by teleconference, from 11:30-5:30 Eastern time. Further details will be available by July 30 at the TGDC website. At this meeting the TGDC will finalize it draft of the next set of Voluntary Voting System Guidelines (VVSG).

There are a series of conference calls scheduled this month for the various TGDC subcommittees, you can find more information on those meetings online. Given the schedule, it is not surprising that the previously established July 31 deadline for submitting the TGDC draft to the Election Assistance Commission (EAC) has slipped. That said, given the rush to develop standards following the passage of the Help America Vote Act, it would be preferable to have the document go to the EAC late, but without outstanding technical questions.
(more…)

Update - As noted by one of our readers, the meeting has been postponed. Per the TGDC Website:

There are TGDC Subcommittee meetings scheduled through July, so this could run right up to the July 31 deadline for submitting the TGDC draft to the EAC.

Original Post
The Technical Guidelines Development Committee (TGDC) will hold its final meeting on the next version of the Voluntary Voting System Guidelines (VVSG) on July 3. Unlike its previous plenary sessions, this meeting will take place via teleconference. Anyone can call in to the meeting, and it will be webcast. There are many resources available on the meeting’s webpage.
(more…)

An article in today’s New York Times (registration required), describes the current status of military voting overseas. In brief, the systems in place to help service members stationed overseas to vote remain “slow, confusing and plagued with security and privacy problems.” Meanwhile, Americans serving abroad continue to be frustrated with the inefficiency of voting while out of the country.

The challenges of a reliable system for overseas voting have been highlighted in work on the SERVE project (Secure Electronic Registration and Voting Experiment), which involves some USACM members. You can read some of the work evaluating SERVE and related projects online.

On Monday and Tuesday, the Technical Guidelines Development Committee (TGDC) held their 9th plenary meeting at NIST headquarters in Gaithersburg, Maryland. Responsible for advising the Election Assistance Commission (EAC) on the next edition of the Voluntary Voting Systems Guidelines, the TGDC is working hard to make sure they can present their draft to the EAC by the end of July. The May 15th draft is available online for review and comment. While TGDC meetings typically don’t have time for public comment, Anyone may comment on the draft by emailing voting@nist.gov. The document is quite lengthy, so it’s not too early to review it and submit your comments.

As of the end of the meeting yesterday, the sections requiring additional work before final approval were summarized in a document available online.

Much of the discussion concerned clarifications of specific requirements, as well as for definitions that may have different meanings for the technical and election communities, or with the public. At this point most of the changes are small refinements and adjustments in requirements to insure a more consistent, uniform and accessible document. There are some areas, such as requirements for cut sheet voter verified paper records, that need refinement, but the bulk of the document is set.
(more…)

I apologize for overloading the blog today, but this has been a busy week so far.

Late yesterday afternoon the Committee on House Administration passed, on a partyline vote, Representative Rush Holt’s legislation – The Voter Confidence and Increased Accessibility Act of 2007. (Here is the amended version that passed out of committee. For background on the legislation see our posts (1,2.) This legislation may be referred to other House Committee for consideration, or may go straight to the full House for action. We should know more about this in a few days.

Below is our release on the Committee’s action:

(more…)

The Committee on House Administration has acted within the last week on both the Holt e-voting legislation, H.R. 811, and the contested election for the Florida 13th Congressional District.

First, we must note the passing of the Committee’s Chair, Representative Juanita Millender-McDonald, on April 22nd from cancer. She had just started a formal leave from Congress when she passed, and is the second Representative to pass this year (Rep. Norwood of Georgia also passed from cancer in February). Rep. Brady of Pennsylvania is the interim chair of the committee. Speaker Pelosi is expected to name a new chair within the next few weeks. Rep. Brady is currently running for mayor of Philadelphia, so he may not be chosen, depending on the results of his primary later this month. Rep. Artur Davis of Alabama has joined the committee.

The committee met on Thursday, May 3 to markup HR 811, Rep. Holt’s e-voting legislation. There are several amendments under consideration. One is a substitute offered by Rep. Lofgren that has some changes from the Holt bill - more flexibility for the states in how they select audit personnel, and an exemption of commercial off the shelf (COTS) technology from the requirements to place voting technologies in escrow. The Republican members have “9 or 10″ amendments, only one of which was available. This amendment, offered by Rep. Ehlers, was a slimmed down version of the bill, with some impractical computer security provisions. The markup was postponed until Tuesday, May 8, given the lack of time members had to review the pending amendments.
(more…)

This morning a Task Force of the Committee on House Administration held a closed meeting regarding the voting irregularities in Sarasota County’s election for the 13th District seat. The Task Force is headed by Representative Charles Gonzalez of Texas. The Task Force was formed in late March, as noted in the committee press release currently on the front page of the committee’s website.

The letter referenced in the press release is discussed further in this Wired.com article. In short, ES&S, the manufacturer of the iVotronic machines used in Sarasota County, notifed the county of a “flaw that sometimes caused machines to respond slowly to a voter’s touch ‘beyond the normal time a voter would expect to have their selection highlighted.’” Given the undervote issue in the Sarasota votes, this is a potential cause for some of those undervotes. If a voter did not touch the machine long enough, due to the flaw, then the vote would not have been registered.
(more…)

We are continuing (regrettably after some delay) to take a deeper look at Representative Holt’s proposal – The Voter Confidence and Increased Accessibility Act of 2007. In my last post, I discussed his central proposal, which is to mandate that all machines produce voter-verified paper ballots that preserve privacy and are durable. Today I’m going to review his proposals for reforming the testing and certification process and post-election audits.
(more…)

Readers of Hill Tech Happenings will have noticed the recent string of voting related hearings in Congress. It’s just one sign of the continued interest in voting reform expressed both in Congress and the Executive Branch. Once the flurry of hearings end (for the moment) tomorrow, we can give you a more in-depth analysis of the situation. But we thought it important to make a note of some things going on at the moment.

First, the third of a series of hearings the Committee on House Administration has been holding on election reform takes place Friday, March 23, at 9:30 a.m. in 1310 Longworth Office Building. Princeton Professor (and USACM-EC member) Ed Felten will be testifying at the hearing, along with several elections officials, voting reform advocates, two Members of Congress, and the Governor of Florida, among others. While the other two hearings (held on March 15 and 20) have focused on voting reform more generally, tomorrow’s hearing will focus more on Rep. Holt’s legislation, HR 811. Expect some discussion of the Sarasota voting trouble as well, with the Governor of Florida scheduled to testify. The previous hearings considered accessibility, source code review, and post-election audits. Berkeley Professor (and USACM member) David Wagner testified at the hearing on March 15. The Committee has been slow to post anything on these hearings, and the minority website has only a couple of press releases on the testimony.
(more…)

For the past few Congresses Representative Rush Holt (D-NJ) has taken the lead on legislation to reform electronic voting. Each year his efforts have garnered deep support from the Democratic party, but each year the legislation stalled with no Congressional action. With the Democrats now controlling Congress, Representative Holt reintroduced his legislation – the Voter Confidence and Increased Accessibility Act of 2007 (H.R 811) – last month with high hopes that he will be able to shepherd it into law. Over the next three posts, I’ll review the goals of his legislation and speculate about its prospects and potential obstacles. One thing is clear, if you are interested an e-voting issues, this legislation is the bill to watch.

(more…)

Last month the Chronicle of Higher Education ran a story (subscription required) about the unique relationship the State of Georgia has with Kennesaw State University. The State contracts with KSU to assist with all of Georgia’s e-voting machines, including inspection, ballot databases and training of poll workers. The article generated quite a bit of controversy within the computing community (Here is the start of a thread of debate within the community) about how it was written, the relationship between Georgia and KSU, and the topics it covered and missed. USACM felt the article missed several issues and responded with a letter to the editors. The Chronicle printed (subscription required) a condensed version of our letter on Monday. The full letter is below.

(more…)

The State of Florida has released its audit report for the contested election in Sarasota County. You can see the report from the state and the report from the SAIT lab at Florida State University online, along with the Secretary of State’s statement. Regular readers know we have posted on this subject before. And while the audit is complete, the legal case needs to be resolved, as well as the official contest process in the House. In other words, this isn’t over, but it’s probably the beginning of the end.

This was released only a few hours ago, so we have yet to give it a detailed reading. The audit team did not find that the voting machines contributed to the undervote. The audit team did recommend the following:
(more…)

Election reform was an active issue during the past ten days. Legislation was introduced, a hearing held on electronic voting, and the Election Assistance Commission decided to implement its full testing and certification program for voting systems in early March.

Two pieces of electronic voting legislation have been introduced in the House. HR 756, the E-Poll Book Improvement Act of 2007, was introduced on January 31 by Representative DeGette (D-Colorado). You can read the full text of the bill online. The legislation would require the Election Assistance Commission (EAC) to develop and adopt guidelines for electronic poll books in the same way the EAC is obligated to do so for voting systems. While Rep. DeGette has not been at the forefront of voting reform legislation, she represents Denver and many of its suburbs. You may remember the trouble the Denver area experienced with electronic poll books in the November 2006 election. Part of the problem appears to be due to a lack of sufficient testing, and this is a likely motivation for her legislation.
(more…)

While the first session of the 110th Congress gets underway today, here are a few notes on recent electronic voting activity.

First, the New York Times (registration required) is reporting about a recent disclosure that a testing lab has been temporarily barred from approving new voting machines. The company, Ciber Inc., is also having trouble with New York over plans for testing new voting machines for that state.

The major problems were noted last summer by the Election Assistance Commission, and not disclosed at that time. As transparency is a real concern and issue for many in the effort for electronic voting reform, this lack of disclosure is perhaps as troublesome as the issues with the lab. Ciber was found to not be following its quality-control procedures and could not document that it was conducting all required tests. While the company, predictably, claimed that the problems were not due to “incomplete, inaccurate, or flawed testing,” the lack of evidence parallels a major complaint about voting machines that lack an independent means of verifying votes. Things could go wrong, or go well, and we can’t tell.
(more…)

Happy New Year! We usher in 2007 with a continuing story from last year – security and e-voting machines.

Computerworld has a good interview with Eugene Spafford, USACM’s Chair and Director of Purdue’s Center for Education and Research in Information Assurance and Security (CERIAS) on the security of e-voting issues. The focus of the interview is on the recent federal action to create new “software independence” standards by the Technical Guidelines Development Committee, which is charged with drafting the federal technical standards for voting machines. We covered these actions in detail (1, 2, 3) and in our last newsletter.

Here is an excerpt from the interview:

Look for more e-voting issues throughout the year on this blog!

Update - December 21

The court hearing over access to the voting machine source code ended yesterday, with the judge in the case requesting written briefs to be submitted by Friday, instead of oral arguments. While one voting expert testified that it was highly unlikely for the undervote to be a result of voter choice rather than flawed voting machines, another advanced the argument that it was a result of flawed ballot design. The judge gave no indication of when he might rule. Meanwhile, the state’s election division continues its audit.

Jennings’ campaign did submit the paperwork contesting the election to the Committee on House Administration. In preceding cases (the most recent in 1997), the House typically seats the certified winner - in this case Buchanan - and investigates the election. While most investigations take a matter of weeks, the 1997 case lasted 15 months. Expect Congressional voting reform proponents (namely Representative Rush Holt) to use this investigation to push legislation requiring paper trails.

Original Post - December 19

It’s been a while since we posted about the Congressional race in Florida’s 13th District, where the undervotes in one county dwarfed the narrow margin of victory, prompting audits, parallel testing, and a pair of lawsuits. (more…)

Congress’ very short lame-duck session came to an end early Saturday morning wrapping up a largely unproductive 109th Congress in the technology policy space. In the waning hours, Congress did pass a few tech-related measures, but left almost all of the funding and competitiveness bills on the table – including funding for the President’s American Competitiveness Initiative (ACI).

Also, as the 109th Congress concluded, the Democrats were laying plans for the 110th Congress by naming their committee chairs. The Republicans also named their ranking members.
(more…)

The Election Assistance Commission (EAC) met yesterday in Washington, to handle regular business and hear testimony on the mid-term elections. The EAC was created as part of the Help America Vote Act (HAVA) to assist states and localities in the administration of elections and to administer the funds designated by HAVA for states to update their voting equipment. Here is the meeting agenda.
(more…)

Update - December 12
Materials from the meeting, including the webcast and text of the resolutions considered at the meeting, are now available on the NIST website.

Original post - December 5
As I suggested in yesterday’s post the Technical Guidelines Development Committee (TGDC) did revisit the software independent proposal during today’s session. It was the first item, and by unanimous consent a revised resolution was approved requiring that the next generation of voting machines approved under the next Voluntary Voting System Guidelines (VVSG). The revisions, as best as I can tell, were meant to address two main concerns from yesterday’s session.

First, while implementation of this next VVSG is approximately 3 years away, several TGDC members (primarily state elections officials) were concerned that currently existing voting systems would have to be scrapped and/or would be considered substandard. Language was inserted into the resolution (which hopefully will be available online later this week) which stipulated that the TGDC is not suggesting that current systems that are following EAC best practices be replaced at this time. Again, given the timeline for development and implementation of this VVSG, this was not likely to happen. At least one member was still confused about the innovation class resolution approved yesterday, as he attempted to insert similar language to the software independent systems resolution.
(more…)

As Cameron posted to the blog yesterday, the Technical Guidelines Development Committee is meeting today and tomorrow at the NIST Gaithersburg facility to discuss its advice to the Election Assistance Commission for the 2007 Voluntary Voting System Guidelines. The meeting is being webcast and will be archived for later viewing. Presentation slides should also be available online in the next few days. Contrary to the white paper we circulated late last week, and supported in a letter to the NIST director, the TGDC rejected a proposal for requiring voting systems to be software independent under the next Voluntary Voting System Guidelines.

As my post on e-voting late last week suggests, there is some confusion about the timeline the TGDC and EAC are working on with respect to voting guidelines. Some clarification is in order. (Those wishing to jump to the meeting activity should skip ahead.)
(more…)

Update: The TDGC rejected NIST’s and the security subcommittee’s recommendations for software independent systems on a 6-6 tie vote. We’ve got a story about the meeting posted here.

Update 2: The TDGC reversed course and adopted a compromise resolution that embraces the software indepence concept. David posted a story about it here.

Last Thursday we posted a story that the National Institute of Standards and Technology (NIST) released a paper recommending that federal standards allow certification only for “software independent” (i.e. ones that create a paper trail) e-voting systems. A key technical panel will consider and vote upon the recommendations this Monday or Tuesday. Calling these recommendations an important step forward for improving e-voting machine security, USACM issued a letter urging the panel to adopt the recommendations. These events are significant developments in the ongoing debate over e-voting and warrant a closer look.
(more…)

Two items of e-voting interest - aside from the ongoing audit in Sarasota County, Florida.

Electionline.org has released its report on the 2006 election. It outlines the various difficulties experienced across the U.S. and notes that while there were no widespread problems, it would be hard to characterize the first widespread use of electronic voting machines as a success.

A draft report released by NIST and concerning the 2007 Voluntary Voting System Guidelines strongly encourages only using voting systems that are software independent (the accuracy of the election is not dependent on the accuracy of the software). The report does note that a DRE without a paper trail would not be considered software independent. To wit (page 12):
(more…)

Update - November 29
All the area papers have reported on yesterday’s parallel test in Sarasota County. There were some small discrepencies between the results and the scripts they had election workers run during the test. The video will be reviewed today. One item in the press coverage grabbed my attention. If I understand this Miami Herald piece correctly there were a total of 251 votes cast on the four machines that followed the scripts (a fifth one was used that followed no particular script). (Reviewing the Script and Review Screen Checklists for those four machines confirms the total number of votes.)

This is a very small number of votes (there were a total of 123,901 votes cast for Buchanan and Jennings in Sarasota County) on a very small number of machines. The two tests will use a total of 10 machines, while the county used 1,498 on November 8. That means .667% of the total number of machines will be tested - or .333% if you discount the 5 machines tested that weren’t actually used in the election. The audit process will continue beyond the two tests scheduled for this week, possibly extending well into December. Expect the process to be rigorously scrutinized by both campaigns and other parties. Hopefully that will prompt a rigorous review and testing process, but that remains to be seen.

Update - November 28
The first phase of parallel testing is happening as I write this. The State Division of Elections released its updated audit plan yesterday. The Bradenton Herald summarizes the procedure, as well as the participating observers. The new detail in this plan focuses on specific machines that will be involved. For the first part of the test, machines not involved in the election will be used, but they will follow scripts based on event logs from machines used in the election.
(more…)

Update - November 22 The major event since the last update is the filing of a second lawsuit on Tuesday contesting the election - this one by voters and poll workers, supported by several voting rights groups. I do want to note a couple of useful items put out by local press:

Interactive map of the undervotes by precinct - courtesy of the Herald Tribune. You can observe the percentage of undervotes (along with the vote breakdown for each candidate) per district. While the undervote percentage is generally higher in Sarasota County than the rest of the Congressional District, precincts that went for Jennings often have an even higher percentage of undervotes.

The Herald Tribune has a whole section on the recount, where you can see all of their coverage along with documents from the suit filed by Jennings.

The Orlando Sentinel has assessed the undervotes in Sarasota County, looking at how the undervoted ballots voted for other races. According to the Sentinel, these ballots trend very Democratic - even in races where the county favored the Republican candidate.

Update - November 20 As expected (per the Associated Press), the trailing candidate, Democrat Christine Jennings has officially contested the results of the election. The challenge, filed immediately after the State of Florida certified the results, calls for a new election based on statistical anomalies and eyewitness accounts of election difficulties. This challenge could have been filed whether or not the state ordered the audit of Sarasota County machines. In fact, the part of the audit that would test machines used in the election could not take place until the end of the challenge period (the 10 days following certification of results, or not before November 30). The AP reports that the first part of the audit - a test of the voting machines not used in the election - will take place on November 28.

Note: This story is still happening, so watch this space for updates.
(more…)

Update 11/16/06: One of the problems of doing a laundry-list type post like this is that you miss some issues, and some nuance when trying to summarize complex policy issues. Notably missing from this list are issues such as patent reform and reviving the Office of Technology Assessment (OTA). We don’t follow patent reform very closely, but our sense is that the the issue has been a fight between industries rather than over partisan ideology. The biggest breakthrough may be if new Chairmen in the House and Senate inject new energy into the debate. As for the debate over bringing back OTA, we posted some thoughts on the issue when Congress held a hearing on it earlier this summer. When the Republican’s eliminated this agency in 1995, the scientific community generally opposed it. We’d expect a renewed discussion about the issue to emerge, but, as our post points out, even a revived OTA isn’t likely to have a tremendous impact decision-making process on policy.

Also Tim Lee at the Technology Liberation Front pointed out that part of the post related to build out wasn’t clear and that was largely because I glossed over a key issue. I’ve attempted to clarify that section below.

Original Post 11/13/06: The outcome of last week’s election is now clear with the Democrats taking over the legislative branch. The dust, however, is far from settled. Democrats have the opportunity to make a profound and lasting impact on technology policy. What shape this will take is unclear, because much of it will be dictated by the new House and Senate Chairpersons who have not organized or set any agendas yet. Speculation will usually get you into trouble (or make you look foolish), but that won’t stop us from making some guesses about how technology policy issues will be shaped by the new Democratic Congress. These are in no particular order, but are areas that USACM has focused on in the past few years.
(more…)

Today is Election Day across the United States. The main story is who will control Congress when the polls close, but of equal interest (at least to us) will be the experience of voters’ using e-voting equipment, considering that approximately one-third of voters will be using new, electronic, voting equipment today. In an effort to capture a snapshot of some of the issues related to peoples’ experience with this new technology, I’ll try to link to stories that we come across and highlight some of the facts. Hopefully everything will go smoothly today, but with so much new technology and so many new procedures in place, I’m sure there will be some issues that will warrant a deeper look. As a reminder, you can access our e-voting resources here:

• ACM’s Policy Statement on E-voting, including links to Congressional testimony and press releases
• USACM’s Voter Registration Database Report, exporting issues related to new statewide electronic voter registration databases

(more…)

Election Day in the United States - November 7 - is approaching fast. Issues with electronic voting, whether it’s the voting machines, the voter registration databases, or other problems with the process, continue to crop up in the press. This has led to some concern on the part of members of the Election Assistance Commission (EAC), which met yesterday in Washington, D.C. While the EAC, and representatives of voting system manufacturers and testing labs testifying at the meeting, implied the concerns over having votes stolen or other problems with the systems are overblown, those concerns are real and aren’t likely to disappear any time soon. In fact, a representative from Diebold appeared on Good Morning America this morning, and during his segment with Diane Sawyer he seemed unable to convince her that the machines would be completely secure or reliable. Many organizations and individuals will be paying close attention to this election, for a number of reasons. As the relevant laws vary from state to state, the Center for Internet and Society at Stanford Law School has a good resource of election laws. Please consult this guide for election day rules, and your local election jurisdiction for rules about absentee voting, election observing, and provisional voting.
(more…)

While Congress is in recess, this posting typically disappears. However, there is one event of interest, especially for those who have followed electronic voting issues.

October 26
Meeting:
The Election Assistance Commission will hold a public meeting on their voting system testing and certification program.
10:00 a.m., 1225 New York Avenue, Suite 150, Washington, D.C.

The U.S. Election Assistance Commission (EAC), formed as part of the Help America Vote Act (HAVA), is taking responsibility for administering (among other things) the testing and certification (including re-certification and de-certification) of electronic voting systems. They have recently developed a draft procedural manual for testing and certification of voting systems. The EAC is seeking comments, and the comment period closes at 5:00 p.m. Eastern time on October 31, 2006. This is a process-oriented document; for technical standards consult the EAC’s Voluntary Voting System Guidelines (these standards - the 2005 edition - take effect in December 2007).

The focus of the testing and certification manual is on the information needed on the voting systems and their manufacturers. This information will allow the EAC to administer certification per HAVA requirements and the relevant voting standards. Contents of the manual include:
(more…)

Today the Committee on House Administration held a hearing titled “Electronic Voting Machines: Verification, Security, and Paper Trails.” Two USACM members testified. Barbara Simons and Ed Felten. Their testimony can be found here and here, respectively. Below is a press release on the hearing.

This Thursday at 10:00 the Committee on House Administration will hold a hearing on security, verification and paper trail issues related to e-voting machines. Two USACM-EC members have been invited to testify – Barbara Simons, past president of ACM, and Ed Felten, professor of Computer Science at Princeton University. The hearing will be webcast.
(more…)

Professor Ed Felten at Princeton University (also a member of USACM’s Executive Committee) and two associates (Ari Feldman and Alex Halderman) released a new study today confirming the security vulnerabilities found with a popular model of Diebold direct recording electronic (DRE) voting machines by many previous studies and exposing new, potentially more serious ones. Professor Felten says that this analysis is the first fully independent security review of both the machine’s hardware and software. The study confirms and demonstrates previous findings that someone with technical knowledge and physical access to a machine can insert malicious code that can switch votes, deny service, effectively hide from detection and overwrite security logs.

Among the new findings is that the machine is vulnerable to a virus that can spread its malicious code from machine to machine. Each machine has a memory card that is used both to record votes and update the system. Normally this card is behind a locked door on the machine, which, the paper argues, can be easily picked. A new card can be inserted carrying the malicious code and a viruses that installs itself on the machine. When a new memory card is inserted in that machine by a pollworker or a technical to either count votes or update the machine, the card will become infected. If that card is used in another machine, which is often the practice, the virus will spread.

This is significant because election officials have argued wide-spread fraud requires both technical knowledge and access to each machine. This new research shows that a virus can spread from machine to machine after access to only one machine.

So what is the take away from all this? Professor Felten argues:

ACM and USACM have consistently pointed out (1,2) that we need better engineered and tested system. Reports of security vulnerabilities are on systems that have already been certified as meeting federal voting system standards. Clearly we need to strengthen both the standards and testing process. Beyond that, a voter-verified paper audit trails will provide a safeguard against the security and reliability issues that are inherent in any computer system.

Update - September 12

While audit trails are important, proper planning works wonders for successful voting. Today (9/12) is a primary election day in my local jurisdiction - Montgomery County, Maryland. Arriving shortly after the polls opened this morning, I found that the smart cards necessary to use our DRE machines were not there. In fact, after calling the County Board of Elections, I found that they had not arrived at any polling place in the county. Poll workers had no instructions for what to do, and only let voters use provisional ballots after hearing from the Board. I’ll go back to vote this evening, but how many people would be so inclined in similar situations?

Why election officials couldn’t have given this guidance before Election Day escapes me. Technologies fail, whether or not computers are involved. Planning for problems can help mitigate them when they happen. Check with your local election board and see what its plan is for trouble with the equipment, in particular what poll workers are empowered to do in such a situation.

Original Post - September 11

One of the critical points in the USACM statement on electronic voting is the need for an audit trail, a mechanism (usually, but not always, paper) independent of the voting technology that allows the voter to verify their vote. A common complaint with many voting technologies is the absence of an audit trail or difficulties in accessing or producing that audit trail.

The National Institute of Standards and Technology (NIST) is taking over the responsibility for approving the laboratories that will certify voting technologies. In a report written for NIST, (PDF) Roy Saltman, author of “The History and Politics of Voting Technology: In Quest of Integrity and Public Confidence,” argues for the importance of auditing direct recording electronic (DRE) voting machines - a principle USACM finds critical to secure and reliable elections. He discusses the use of paper trails, as well as several different independent auditing technologies.

(more…)

Congress ended their summer recess this week with just three short weeks left before heading home again to campaign for the November elections. On the technology policy front, things will likely be busy as several initiatives will wind their way through Congress, but we don’t expect many, if any, on the President’s desk before October. This means most of the endgame will play out in a November/December lame-duck session of Congress. Let’s take a closer look at what is on Congress’ plate:
(more…)

July has seen a lot of attention focused on e-voting issues. First, the Brennan Center releases a major report on threats to e-voting systems. Then Congress holds a much-needed hearing on e-voting (USACM offered testimony). And this week the National Academies released an interesting new report discussing emerging problems with e-voting systems and making recommendations for the upcoming election.

According to the Election Assistance Commission, an estimated one-third of voters will be using different equipment in 2006 than 2004. The academies’ report stated that some jurisdictions, possibly many, are not prepared to use this new equipment for the November election. Several factors are contributing to this (although not uniformly across jurisdictions):

• State and localities either not meeting or rushing to meet deadlines for new equipment mandated by the Help America Vote Act
• New state requirements on e-voting systems, such as Voter Verified Paper Trails
• On going security threats, such as those outlined in the Hursti report
• Poor vendor performance
• Training poll workers on how the new equipment works
• Educating first-time voters about using the machines

(more…)

Facing a steady stream of reports about e-voting machine failures and security vulnerabilities, yesterday Congress tackled the question of whether new federal standards will improve this technology. The short answer from the witness and the Members of Congress seemed to be that the standards, while a good first step, were far short of what we need to ensure accurate, secure, reliable and usable e-voting machines. We will have to wait until the next update of the standards (likely effective in 2010) to see anything more than marginal improvements.

USACM mirrored these concerns in a letter it released to Congress in advance of the hearing; noting that even with improved standards, there are still gaps in the current testing and certification system. USACM made five recommendations to close these gaps:

• Create a formal feedback process that will ensure that lessons learned from independent testing and Election Day incidents are translated into best practices and future standards.
• Make the testing process more transparent by making the testing scope, methodologies and results available to the public.
• Ensure that the guidance for usability and security standards provides performance-based requirements and is clear so as to minimize the variance of human interface designs from jurisdiction to jurisdiction.
• Create a mechanism for interim updates to the standards to reflect emerging threats, such as newly discovered security defects or attacks.
• Require voter verified paper trails and audits to mitigate the risk associated with software and hardware flaws.

USACM also issued a press release on the hearing, which can be found here.

(more…)

In late June the Brennan Center for Justice, located at the NYU School of Law, released a report on voting system security. The report, The Machinery of Democracy: Protecting Elections in an Electronic World, was compiled over the course of a year by a task force of government, academic and private sector experts and is a systematic assessment of vulnerabilities in each major kind of voting system. The task force found significant security and reliability vulnerabilities in each voting system, though proper countermeasures could reduce these vulnerabilities. However, few jurisdictions have taken such steps.

Reviewed extensively by the National Institute of Standards and Technology, the report is available in full as a PDF file. You can also read the press release and the executive summary online.

This complements the work USACM has done on e-voting, including our study on voter registration databases.

With so much happening in the technology and technology policy fields, it is a rare day that computer experts focus on one particular issue. So it seems pretty significant that several blogs and news reports with quotes from well-respected computer scientists are focusing on a new report by Harri Hursti for blackboxvoting.org outlining several new serious security flaws. While many of the details of the report have been redacted, apparently the most significant flaw is a well-known exploit that can be used to completely compromise the machine including vote counting, using widely available tools. The ease with which this security hole can be used and the relatively trivial knowledge it requires has shocked computer scientists.

I’ve just started reading the short report, but let me put up pointers to the stories and blog posts which outline the issue very well:

• Inside Bay Area.com report
• New York Times’report
• Post by Ed Felten and Avi Rubin on Freedom-to-tinker
• Post by Bruce Schneier on Schneier on Security Blog

ACM issued a statement in 2004 calling for voting systems to have a physical (e.g., paper) record to verify that individual’s vote has been accurately cast. That statement also called for all voting systems to “embody careful engineering, strong safeguards, and rigorous testing in both their design and operation.”

Friday’s Computerworld has an interesting article about problems with the new California statewide voter registration database. According to the report, in the past three months about 43 percent of voter registration forms in Los Angeles County were rejected. Alademda County’s rejection rate was about 10 percent. Although the article is a little vague, the likely cause is data entry errors that cause conflicts when the voter database is compared against the motor vehicle database.

This cross-checking is a new requirement under the Help America Vote Act (HAVA). HAVA requires that states authenticate each potential voter by cross-checking with other state databases. If a potential voter does not have a state driver’s license, then the last four digits of the voter’s Social Security number must be used for authentication.

We recently issued a report USACM commissioned looking at statewide voter registration databases, and the cross-checking of databases was one of the key issues the study committee identified (page 23). Because data entry errors are a significant problem and databases (voter or motor vehicle) can be inaccurate, the study committee recommended that:

• When other databases, such as driver registration databases, are used to check for eligibility, those databases should be used for screening and not to automatically enroll or de-enroll voters.

• An automated check can be used to flag some voters for further scrutiny, but the final determination of eligibility should be performed only by an appropriate election official.

Apparently under the California system, any errors are kicked out for manual checking. While the Computerworld article notes criticism of the rigid cross-checking protocol, what is encouraging is that they are manually checking these errors instead of automatically dumping these potential voters. However, the article also points out the problem the timing when dealing with these errors:

Closing voter registration well before an election, or adding additional staff to deal with the manual checking may be the only ways to ensure that voters aren’t disenfranchised.

The Washington Post reports that Maryland Governor Ehrlich (R-MD) wrote a letter to the Chairman of the State Board of Elections calling for the board to replace touch screen voting machines with optical scan machines. According to the Post, his letter states:

The Governor is endorsing legislation by Maryland House Member Sheila Hixon (D-Montgomery) that would require the state to lease optical scan systems for the next election. Her bill was approved by committee last week, but according to the Post its prospects in the Senate aren’t clear.

ACM issued a statement in 2004 calling for voting systems to have a physical (e.g., paper) record to verify that individual’s vote has been accurately cast. That statement also called for all voting systems to “embody careful engineering, strong safeguards, and rigorous testing in both their design and operation.”

Update 2/17/06: Declan McCullagh wrote a nice story about the study for CNET News.com.

Original Post 2/16/06: Citing the danger of voter fraud and disenfranchisement from poorly implemented databases, a committee of experts commissioned by USACM released a report today making almost 100 recommendations to state and local officials charged with creating and managing statewide voter registration databases (VRDs). The report is a comprehensive “soup-to-nuts” look at challenges elections officials face in making sure databases are accurate, private, usable, secure, and reliable.

The report responds to the mandate passed by Congress in the Help America Vote Act that all states create and manage statewide VRDs. The committee wanted to give objective technical advice to elections officials on the many challenges they will face in meeting this mandate. The report has been about a year in the making, and we’ve covered some of the work of this group on the weblog (1, 2) including comments they filed with the Election Assistance Commission about this issue.

The press release on the report is below (and a .pdf is here), and here is a link to the page hosting the report.

The policy office personally wants to thank the study committee members for their hard work on this report.  Each took time from his or her regular life to scope out the issues and then write a very detailed and comprehensive report.

(more…)

North Carolina’s State Board of Elections created quite a stir recently when, surprisingly, it decided to certify Diebold e-voting equipment for use in the state despite a still-running debate about the state’s new source code “escrow” rules and the company’s compliance with them. The requirements call on companies to deposit the source code for their e-voting equipment with a state-approved third party and for the State Board of Elections to evaluate that code. A ZDNet article describes the situation in more detail:
(more…)

Yesterday, USACM sent a letter to Virginia Delegate Tim Hugo, chair of the Joint Subcommittee Studying the Certification, Performance, and Deployment of Voting Equipment, regarding the issue of voter-verified paper records and electronic voting machines. The subcommittee is scheduled to hold a hearing on Monday, November 21, 2005, to address the issue (an agenda for that hearing is available here).

The letter seeks to draw the lawmaker’s attention to the e-voting position that ACM adopted last year. Readers may recall that the committee sent a similar letter to lawmakers in Hawaii earlier this year.

The full letter is available below or here as a PDF file.
(more…)

Update (Sept. 26) – Carter and Baker have issued a response to some of the criticism their report has received – their comments appeared in the NY Times on Sept. 23 and are available here.

As pointed out in today’s Washington Post, the Federal Commission on Election Reform (led by former president Jimmy Carter and former secretary of state James A. Baker III) issued its report today which recommends

Here in ACM’s Policy Office, we see the Carter-Baker report as a decidedly double-edged sword:
(more…)

EDRi’s latest newsletter informs us about a recent article in the Irish Times [subscription required] describing the Irish government’s plans to subject their e-voting machines to additional security and risk-related scrutiny:

EDRi’s newsletter explains how last year the government “decided at the last minute to cancel the usage [of e-voting machines], after the Independent Commission on Electronic Voting concluded in an interim report” that it could not “satisfy itself as to the accuracy and secrecy of the system.” The commission’s full report is available here.

Meanwhile, as regular readers will know, ACM has a study underway currently looking into implementing the statewide voter-registration databases mandated by the Help America Vote Act. Study members are making good progress and expect to release a report this Fall.

The NY Times is running an editorial today urging House members to support Rep. Rush Holt’s electronic voting bill (H.R. 550):

More information about EFF’s efforts on behalf of H.R. 550 can be found on Deep Links, their weblog.

Readers may recall that ACM issued a statement in 2004 calling for voter-verifiable physical records in electronic voting systems and improved reliability, security, and verifiability of public elections.

Previously, we reported that ACM created a committee to study issues related to statewide voter registration databases and that the Election Assistance Commission (EAC) recently released its draft guidance on this topic, seeking additional comments. ACM’s committee took the opportunity to provide expert comments to the EAC.

The Help America Vote Act requires that all states have statewide voter registration databases in place by January 1, 2006. Seeking to give the states guidance on this law, the EAC released draft guidelines on how to implement statewide registration databases. ACM’s committee felt that the guidance was a good start, but that it covers only a few issues and lacks much of the technical detail that states really need:

ACM recently formed a committee of experts (names and affiliations below) to provide states with guidance on implementing statewide voter-registration databases. Today the Committee outlined its efforts before the Election Assistance Commission (EAC), which is seeking input on its proposed guidance to the states regarding these databases.
(more…)

Current Election Assistance Commission (EAC) member and former EAC chairman DeForest B. Soaries Jr. recently announced his resignation from the EAC, citing, among other things, dissatisfaction with the level of support the EAC has received from the federal government:

(more…)

Washington D.C. hosts two major voting-policy events this week. First, the Carter/Baker Commission on Federal Election Reform kicked-off its first hearing yesterday, part of a six month effort to study the 2004 elections and make recommendations to policymakers. The Commission is led by former President Jimmy Carter and former Secretary of State James Baker and is similar to a commission President Carter and President Gerald Ford formed in 2000.

The hearing focused on a number of subjects, with one panel specifically on voting technology and election administration. USACM member David Dill testified on this panel. He argued that transparency is the most important factor for ensuring our elections are fair and widely accepted by the public, and that the growing use of paperless electronic voting systems is undermining this goal. He specifically advocated ACM’s position noting it was one of those rare cases where a vast majority of computer scientists could find common ground. (ACM’s polled its members and found 95 percent supported its statement on voting). (more…)

USACM today sent a letter to state policymakers in Hawaii to call their attention to the e-voting statement that ACM adopted last year.

The letter seeks to offer USACM’s technical and policy input as electronic voting legislation works its way through the Hawaii legislature (something that’s currently going on in many state legislatures across the country).

The letter appears below and is also available as a PDF file.
(more…)

“MARYLAND VOTERS will never know for sure whether their election choices last year were recorded correctly – and the same uncertainty could haunt them next year if lawmakers again fail to address a serious defect in the touch-screen voting machines used throughout the state. When functioning properly – and the state elections administrator, Linda H. Lamone, insists that nearly all the machines did work last time – they are said to be as accurate as they are efficient. But without a paper trail showing each vote cast, who’s to know? [emphasis added] And what about the machines that did freeze or had mechanical problems? Voters should not have to take it on faith; yet as it stands, there is no way to conduct a solid recount or audit […]”

SOURCE: Washington Post

Note: ACM issued a statement in 2004 calling for, among other things, improved reliability, security, and verifiability of public elections.

“After a second consecutive presidential election marred by significant flaws in the mechanics of voting, it’s time for Congress to take a hard look at fixing the system. Two Senate bills aim to do that. A Republican-sponsored bill is narrowly tailored around making electronic voting more reliable. A more ambitious bill, sponsored by the Democrats, would take on a broad array of problems, from long lines at the polls to odious maneuvers aimed at keeping people from voting. Both bills would greatly improve the functioning of American democracy.

The Republican bill, introduced by Senator John Ensign of Nevada, would focus on the most critical weakness in the system by requiring that electronic voting machines produce voter-verifiable paper records of the votes cast. The paper records would take precedence when there were inconsistencies. (more…)

” Voting machines must include a verifiable paper trail and audit capability in time for the 2006 elections, according to a bill introduced this week in Congress.

[…] Called the Voting Integrity and Verification Act, the bill says states must allow the “voter to review an individual paper version of the voter’s ballot before the voter’s ballot is cast and counted.” The paper ballot, typically viewed under glass, would become a “permanent paper record” that must be preserved in case of a recount.

[…] Computer scientists such as those represented by the Association for Computing Machinery have flagged potential security problems and have called for electronic voting machines to produce a “physical record.” The Information Technology Association of America has opposed mandatory paper trails […]”

SOURCE: CNET News.com

Note: View ACM’s statement on e-voting from 2004 here.

“The November election may feel like ancient history, but it is still going on in North Carolina. The state has been unable to swear in an agriculture commissioner because a single malfunctioning electronic voting machine lost more ballots than the number of votes that separate the two candidates. The State Board of Elections, the candidates and the public are sharply divided on how to proceed. The mess North Carolina finds itself in is a cautionary tale about the perils of relying on electronic voting that does not produce a paper record.

When the returns came in for the agriculture commissioner race, two things were clear: the Republican, Steve Troxler, and the Democrat, Britt Cobb, were just 2,287 votes apart, and a voting machine in Carteret County had lost 4,438 votes. The machine had mistakenly been set to keep roughly 3,000 votes in its memory, which was not enough. And in a spectacularly poor design decision, it was programmed to let people keep “voting” even when their votes were not being saved.

[…] North Carolina’s plight underscores a basic point about elections: because there are often problems, there must be a mechanism for a recount. If the Carteret County voting machine had produced a voter-verified paper record each time a vote was cast, these paper records could have been be counted and the matter would be resolved. But electronic voting machines that do not produce paper records make recounts impossible […]”

SOURCE: NY Times

Note: ACM issued a statement in 2004 calling for, among other things, improved reliability, security, and verifiability of public elections.

Responding to an earlier letter to the editor in the NY Times, Cem Kaner (professor of software engineering at the Florida Institute of Technology, IEEE e-voting working group member, and USACM member) has written a letter focusing on e-voting auditability and the standards-making process:

ACM issued a statement in 2004 calling for, among other things, improved reliability, security, and verifiability of public elections.

“In a ceremony led by Members of Congress, the U.S. Election Assistance Commission (EAC) today installed the Hon. Gracia Hillman as chair of the agency.

[…] Hillman, a former executive director of the League of Women Voters of the U.S., will hold the EAC leadership position for one year. She is a Democratic appointee who served as EAC vice chair during 2004, its first year in existence.

[…] Federal law requires EAC to carry out research studies and produce voluntary guidance on a range of topics in 2005. EAC is working with the National Institute of Standards and Technology (NIST) to develop updated voluntary voting systems standards and recommendations for making e-voting equipment more secure […]”

SOURCE: EAC

“A study by Berkeley grad students and a professor showing anomalies with electronic-voting machines in Florida has been debunked by numerous academics who say the students used a faulty equation to reach their results and should never have released the study before getting it peer-reviewed.

The study, released three weeks ago [covered here] by seven graduate students from the University of California, Berkeley’s Quantitative Methods Research Team and sociology professor Michael Hout, presented analysis showing a discrepancy in the number of votes Bush received in counties that used touch-screen voting machines versus counties that used other types of voting equipment.

But Bruce McCullogh, a decisions science professor at Drexel University in Philadelphia, and Binghamton University economics professor Florenz Plassmann released an analysis (.pdf) of the Berkeley report criticizing the results […]”

SOURCE: Wired News

“Electronic voting machines in Florida may have awarded George W. Bush up to 260,000 more votes than he should have received, according to statistical analysis conducted by University of California, Berkeley graduate students and a professor, who released a study on Thursday.

[…] Their aim in releasing the report, the researchers said, was not to attack the results of the 2004 election in Florida, where Bush won by 350,000 votes, but to prompt election officials and the public to examine the e-voting systems and address the fact that there is no way to conduct a meaningful recount on the paperless machines […]”

SOURCE: Wired News

Note: See ACM’s recent statement on e-voting systems.

A panel of experts convened by the AAAS on election technology and administration warned in a report recently that “the American system of voting is broadly vulnerable to error and abuse, and called for a crash-course of study and reform to make results more reliable and to promote better access by voters, especially those who have historically encountered serious impediments to exercising their right to vote […]”

SOURCE: AAAS

“MIAMI - Eight days before Election Day, a federal judge ruled Monday that Florida’s touch-screen voting machines do not have to produce a paper record for use in case a recount becomes necessary.

U.S. District Judge James Cohn said that the machines “provide sufficient safeguards” by warning voters if they have not cast a vote in a particular race and by allowing them to give their ballots a final review.

Rep. Robert Wexler [D-Fla.] had sought either a paper record or an order switching voters in 15 counties from touch-screens to optically scanned paper ballots by 2006 […]”

SOURCE: AP via Yahoo! News

” Electronic voting systems that were touted as the solution to the paper ballots and hanging chads of the 2000 presidential election have become a new source of controversy as experts debate the reliability of software that operates the new systems, whether local election officials have the technical competence to run them and how there can be a recount on machines that keep no paper record of votes cast on them.

In the days leading up to the Nov. 2 election, critics of the new systems are voicing their concerns about the integrity of the ATM-like machines that will be used by one-third of the nation’s voters – more than double the number that used them four years ago […]”

SOURCE: Washington Post [free reg. req.]

From Slashdot:

“The Council of Europe Committee of Ministers [has agreed to] the first international legal text on e-voting in elections and referendums.

The recommendation sets out a blueprint for governments planning to use new technologies for future elections and referendums.

[…] The legal and technical guidelines of the Council of Europe indicate how to build, run and supervise e-voting systems to ensure that results are as reliable as those delivered by traditional paper-based methods. The Council’s recommendation emphasises the need for new voting methods to meet the principles of universal and equal suffrage, free and secret ballots and for the systems to be secure, transparent and accountable. It covers issues such as electoral lists, information to voters and vote counting […]”

SOURCE: Council of Europe

“[…] While many voter rights’ advocates are fighting to decertify electronic voting machines, arguing they’re not reliable, one bloc remains steadfast behind the new equipment – disabled voters who say the machines give them long-denied privacy.

“The need for greater access by millions of people should not be overshadowed by this concern about security to the point that some people throw up their hands and say, ‘Let’s go back the punch card,’” said James Gashel, an executive at the Baltimore-based National Federation of the Blind, or NFB.

The Help America Vote Act, passed by Congress in 2002, requires polling places to provide the same privacy and independence to all voters by 2006 […]”

SOURCE: Wired News

“Gov. Arnold Schwarzenegger signed legislation Monday that will bar the use of electronic voting machines that don’t produce paper trails to verify votes.

The requirement, which takes effect in 2006, is a response to concerns that the machines could be tampered with or produce incorrect results […]”

SOURCE: AP (via SiliconValley.com)

Cites Risks of Computer-based Systems

Seeking to improve the security, accessibility, and public confidence in the voting process, ACM’s elected leadership has approved a public statement on the deployment and use of computer-based electronic voting (e-voting) systems for public elections. ACM’s position is that while computer-based e-voting systems have the potential to improve the electoral process, such systems must embody careful engineering, strong safeguards, and rigorous testing in both their design and operation. [See the official release here.]
(more…)

“[…] A partial answer to the other question is that some basic international requirements for a fair election are missing in Florida.

The most significant of these requirements are:

[…]

• Uniformity in voting procedures, so that all citizens, regardless of their social or financial status, have equal assurance that their votes are cast in the same way and will be tabulated with equal accuracy. Modern technology is already in use that makes electronic voting possible, with accurate and almost immediate tabulation and with paper ballot printouts so all voters can have confidence in the integrity of the process. There is no reason these proven techniques, used overseas and in some U.S. states, could not be used in Florida […]”

SOURCE: Washington Post [free reg. req.]

“The Election Assistance Commission Technical Guidelines Development Committee asked EPIC to offer testimony (pdf) on the impact that new voting technology and polling place practices has on the privacy rights of voters. The hearing was an opportunity for the committee charged with making recommendations on voluntary standards for election systems and voting technology […]”

SOURCE: EPIC

“Roughly a third of the votes cast in the November presidential election will be made on controversial paperless electronic voting machines, but as any political analyst can tell you, the only votes that will matter a great deal will be cast in a handful of swing states.

And just as the Kerry and Bush campaigns are spending most of their efforts in those states where neither holds a heavy margin in the polls, voting advocacy groups concerned with the integrity of voting technology are devoting their resources toward the states which matter most […]”

SOURCE: Wired News

“Just over six weeks before the nation holds the first general election in which touch-screen voting will play a major role, specialists agree that whatever the remaining questions about the technology’s readiness, it is now too late to make any significant changes.

Whether or not the machines are ready for the election - or the electorate ready for the machines - there is no turning back. In what may turn out to be one of the most scrutinized general elections in the country’s history, nearly one-third of the more than 150 million registered voters in the United States will be asked to cast their ballots on machines whose accuracy and security against fraud have yet to be tested on such a grand scale […]”

SOURCE: NY Times [free reg. required]

“Alarmed by software glitches, security threats and computer crashes with ATM-like voting machines, officials from Washington, D.C., to California are considering an alternative from an unlikely place: Nevada.

Silver State voters cast electronic ballots Tuesday on a $9.3 million voting system with more than 2,600 computers and printers in every county. The primary was free of serious problems that have embarrassed registrars in Florida, California, Maryland and other states with touchscreen machines.

[…] Credit the training in Nevada, and credit the printers – which give computer scientists and voter-rights advocates assurances that elections can be fully audited. As many as 50 million Americans elsewhere will use paperless touchscreens this November, and critics say hacking, malfunctioning and other problems in only a few counties could have huge implications in a tight presidential contest […]”

SOURCE: AP via NYT [free registration required]

“Maryland’s electronic voting system remains “terribly vulnerable” to fraud despite steps taken by the state to correct security flaws, a consultant who conducted a study of the system last January testified Wednesday.

Michael Wertheimer of RABA Technologies said his review of the latest report by the State Board of Elections on what it is doing to protect the integrity of the November election “leads me to the opinion this system still would receive a failing grade.”

Wertheimer was a key witness on the opening day of a hearing in Anne Arundel County Circuit Court in a suit filed by TrueVoteMD seeking to force the state to take further steps to assure that there will be an accurate counting of votes in the presidential election […]”

SOURCE: AP via WTOP News

“Judge to Hear Activists’ Demand For Paper Trail on Electronic Ballots

When critics of electronic voting machines warn of Maryland becoming another Florida, with the potential for hundreds if not thousands of lost votes this fall, State Elections Administrator Linda H. Lamone shudders.

[…] But today, Lamone finds the touch-screen system she has championed under attack as an Anne Arundel County judge begins a three-day hearing to determine, among other things, whether elections officials should be forced to provide a paper trail that can verify the results […]”

SOURCE: Washington Post [free registration required]

“The new national elections chairman this week praised computer scientists for calling attention to security problems with e-voting machines and for helping develop new standards for building machines that will be more secure in the future.

“The country owes you a debt of thanks to have taken this challenge of voting systems seriously,” DeForest B. Soaries Jr., chairman of the newly formed federal Election Assistance Commission, said to members of the Institute of Electrical and Electronics Engineers, or IEEE, who are in the process of drafting new standards for electronic voting systems. […]”

SOURCE: Wired News

“Nevada election officials are confident that an electronic voting system being used for the state’s Sept. 7 primary will perform well, despite a problem that showed up in a demonstration of the technology this month in California. Nevada also plans to use the system for the general election in November […]”

SOURCE: ComputerWorld

“[…] A federal judge today upheld California Secretary of State Kevin Shelley’s April 30 directive that decertified touch-screen voting machines and withheld future certification until vendors of those systems could meet specific security requirements, including voter-verifiable paper audit trails (VVPAT).”

SOURCE: ComputerWorld

In a letter to the newly created U.S. Election Assistance Commission, USACM urged that independent technical experts be involved in the development and assessment of electronic voting standards and technologies. USACM offered to provide its technical assistance to the Commission.

USACM has joined a diverse coalition of voting rights groups and reform advocates in sending a letter to Congress urging full funding for research and standards initiatives intended to enhance the security of electronic voting systems. USACM also issued a press release.