E-voting backers see success, but issues linger

By Grant Gross The Industry Standard  03 November 2004

Critics of electronic voting security conceded Wednesday that the number of glitches reported with e-voting machines likely didn't have an impact on the U.S. presidential race. As supporters of e-voting technology trumpeted the apparent success of electronic voting technology for an estimated 40 million U.S. voters in Tuesday's election, Will Doherty, executive director of e-voting watchdog Verified Voting Foundation, noted however that the 1,181 e-voting related incidents reported to the group's Voteprotect.org database may have affected other close races, if not the presidential race.

"So there weren't enough e-voting problems to shift the results of the presidential elections," Doherty said. "Does that mean it's OK to disenfranchise voters?"

In the presidential race, Republican President George Bush defeated Democratic challenger Senator John Kerry, with Bush winning the popular vote by about 3.5 million votes, according to unofficial numbers. More than 114 million votes were cast.

Even though predictions of massive e-voting problems seem not to have come true, e-voting critics won't stop raising concerns, Doherty said. His foundation will use the incidents reported to Voteprotect.org to push for better e-voting machine security, he said.

"(The database) will be helpful moving forward with negotiations with elections officials ... and lawmakers," Doherty said. "Verified Voting Foundation will focus on how we can make voting systems more secure and reliable."

The Electronic Frontier Foundation (EFF), another group that is critical of e-voting security, will move forward with plans to file requests with election officials to check the accuracy of e-voting machines, said EFF staff attorney Matt Zimmerman.

The relatively small number of reported problems Tuesday won't stop the EFF effort, Zimmerman said. "What we've been trying to say rather forcefully over the past couple of months is the things we're worried about are the things you can't see," Zimmerman said. "A mechanism needs to be put in place where we can see what's going on (inside the machines)."

Zimmerman and Doherty questioned whether voting officials are able to perform independent audits on e-voting machines, rather than simply printing out the same voting numbers as originally reported.

E-voting proponents, however, hailed Tuesday's vote as secure and accurate. The Information Technology Association of America (ITAA), a vendor trade group, estimated that 40 million U.S. voters used about 175,000 e-voting machines in the election, and said the machines performed with a "minimum of disruptions."

With the performance of e-voting machines this week, questions about the machine's security will "diminish," predicted Bob Cohen, senior vice president of ITAA. "I think the American people accept the benefits of digital technology," Cohen added. "They see that the technology will be put to use to ensure secure and accurate elections, and that's what happened yesterday."

ITAA President Harris Miller, in a statement, said a "small number of largely unsubstantiated reports of problems with machine implementation by self-appointed and often overtly political critics of reform" did not hurt the reputation of e-voting machines.

The reports of e-voting problems in the Voteprotect.org database are unverified, and in some cases, do not appear to be directly related to e-voting technologies. Among the 115 incidents of supposed e-voting problems in Ohio was a report of "rude" voting inspectors and a report of insufficient voting machines available for the people in line in Cuyahoga County.

But the database also contains dozens of reports of e-voting machines breaking down and people leaving without voting because of resulting long lines. People leaving polling places without voting may have had a significant effect on some races, Doherty said.

But the ITAA isn't aware of any significant delays directly caused by problems with e-voting technology, and the number of problems reported are small compared to the votes cast on e-voting machines, said the ITAA's Cohen. "When you consider the fact that you had 40 million voters casting an average of maybe 10 votes, what you're taking about is huge numbers of successful votes," he said.

Most problems related to voting Tuesday seemed to be more "process-based" than technology-related, added Mike Alvarez, co-director of the CalTech/MIT Voting Technology Project (VTP). Many of the long lines reported Tuesday were caused by issues other than e-voting technologies, including lengthy check-in procedures and record numbers of voters in some precincts, Alvarez said.

"One should be concerned with the entire security of the election process," added Jonathan Katz, a professor of political science at the California Institute of Technology and a member of VTP. "(E-voting machines) make some things easier. They are no panacea there are security issues but there are security issues with all technology.