News

Garber Announces Advisory Committee for Harvard Law School Dean Search

News

First Harvard Prize Book in Kosovo Established by Harvard Alumni

News

Ryan Murdock ’25 Remembered as Dedicated Advocate and Caring Friend

News

Harvard Faculty Appeal Temporary Suspensions From Widener Library

News

Man Who Managed Clients for High-End Cambridge Brothel Network Pleads Guilty

Electronic Election Economics

By Susan E. Mcgregor

In Cleveland, OH, the elections board is running television advertisements on how to properly punch an absentee ballot, and in Florida lawsuits are already underway to challenge the use of touch screen voting machines in 15 counties. While the final weeks before the election may find much of the country arguing over who they will vote for, in many places, the real question is how they will vote.

At the center of this debate is the issue of Direct Recording Electronic (DRE) type voting machines, much like those currently in use in Florida. Such systems typically do not produce simultaneous paper ballots, meaning that voters do not have the opportunity to physically confirm the content of their vote. And while the 2000 presidential election demonstrated the importance of voter-confirmation, the proposed solution for providing that feedback is more dangerous still: the widespread use of DREs that do not create a physical, human-readable ballot at the time of voting. And as long as the concurrent production of a human readable paper-trail, and/or open source code are not required in voting machines, elections-systems companies could literally own America’s elections.

Widespread DRE use is already being realized. In response to the debacle that was Florida in the 2000 presidential election, Congress passed the Help America Vote Act (HAVA) in 2002. HAVA allocated $3.86 billion in federal funds for the replacement of lever and punch card voting systems across the country. To comply with HAVA, a state may either continue to use punch card systems by “(i) establishing a voter education program specific to that voting system that notifies each voter of the effect of casting multiple votes for an office; and (ii) providing the voter with instructions on how to correct the ballot before it is cast and counted” (sec. 301), or the state may simply purchase optical scanning or DRE voting machines.

Not surprisingly, numerous counties have turned to DREs, most likely because their interfaces are more user-friendly than the bubble sheets (think SATs) read by optical scanners. In a notable example, in its 2002 gubernatorial election, Georgia became the first state “to implement a statewide uniform system of electronic voting.” Coincidentally, that election also represented the first time in 130 years that the state of Georgia elected a Republican governor.

The machines used in Georgia in 2002 were produced by Diebold, one of the largest voting machine manufacturers in the country. Shortly before the 2002 election, Diebold administered “patches” to 22,000 voting machines across the state of Georgia. These patches were never independently reviewed by any authority; even Georgia’s independent certifier, Dr. Brit Williams, merely accepted an assurance “by the vendor that the patch did not impact any of the things that we had previously tested on the machine.” Williams also admitted that he had never reviewed the source code on the machines himself, and the national Independent Testing Authorities (ITAs) apparently have, “neither the staff nor the time to explain the process to the public, the news media or jurisdictions.” So it’s unclear whether the source code for these systems has ever been reviewed, line by line, by any independent election authority. In terms of the voting systems’ integrity, in other words, Georgia took Diebold’s word for it.

And who is Diebold, anyway? Perhaps most notably, a company that has donated over $400,000 to the Republican Party in the last four years, and whose CEO has been quoted as saying that he would help “deliver Ohio’s electoral votes” to President Bush. Of course, it’s impossible to prove that Diebold effectively rigged Georgia’s election, but that’s exactly the point. As long as their code is proprietary—and it is—no one can. Ask a lawyer to prove a case but deny him any evidence.

In May of 2003, twin bills were introduced in the House and Senate (H.R. 2239 and S. 1980, respectively), that addressed both the need for a human-readable paper trail, and open-source code in electronic voting machines:

“(i) No voting system shall at any time contain or use undisclosed software. Any voting system containing or using software shall disclose the source code of that software to the Commission, and the Commission shall make that source code available for inspection upon request to any citizen.”

At the end of the Congressional session this summer, however, the bill had yet to receive a hearing. This was probably because on July 7, 2004, the House Administration Committee, to which the bill was referred, heard testimonial from elections officials who stated that they “strongly advocate the advantages of current generation electronic voting technology.” Who were these officials? Kathy Rogers, Director of the Georgia Elections Division, and that diehard Diebold believer, Dr. Brit Williams.

Meanwhile, in Florida, 15 counties are hoping that their new voting machines don’t melt down in the Florida heat, and that the votes they cast will be the ones that are counted. A simpler way of doing this, of course, would have been to buy optical-scanning machines - which comply with HAVA standards, reduce the number of machines required, and already produce the contemporaneous, human-readable physical ballot necessary for true voting security. For those who prefer things truly paper-free: require open source code in any voting machine, and print out a final receipt at the end of the day. This would require no adjustments to existing machines whatsoever. It would simply require that a company like Diebold publish its code.

Of course, it’s not hard to see why Diebold and other voting systems manufacturers would oppose an open-source requirement for election-systems—without proprietary code, there’s no money in it. That means that elections would become simply—a public service—instead of a profitable business. But perhaps that’s as it should be. Perhaps elections shouldn’t be for sale, especially at the public’s expense.

Susan E. McGregor '05, an editorial comper, is a special concentrator in "Interactive Information Design" living in Quincy House.

Want to keep up with breaking news? Subscribe to our email newsletter.

Tags