Congressman:Touch-voting problematic

By Ryan Groshong

Electronic voting machines used in the 2004 elections can be inaccurate and hurt the integrity of the election process, a local congressman said on campus Monday.

Mike Honda, a Democrat who represents the 15th district that includes Santa Clara, hosted a panel of experts who addressed questions about its security and the possibility of fraud.

According to materials provided by Honda's staff, there were numerous reports of voting irregularities in counties that employ Direct Recording Electronic Touchscreen voting machines (DREs) in November's election.

The use of electronic voting is only expected to increase with the implementation of The Help America Vote Act of 2002 which requires that by 2006, "each polling place used in a federal election have at least one voting machine that is fully accessible for persons with disabilities."

"Like many of you I am concerned about the integrity of our election process and the way votes are counted in particular," Honda said in his opening remarks.

The first panelist, Jesse Durazo, from the Registrar of Voters in Santa Clara County, led Santa Clara County's implementation of electronic voting. He described its use as "successful."

"We see the ability that it presents for the disabled community and for the language requirements ease of access," Durazo said.

Honda said he would support legislation requiring a paper trail for electronic voting nationwide.

One of the first issues discussed was the security of electronic voting, as numerous panelists questioned the reliability of DREs.

"Computer technology is just not to the point where anybody can really know what's going on in a computer - that applies even to the people who designed the machine so certainly someone voting on a computerized system can't really know what's going on," said David Dill, a professor of computer science at Stanford.

David Jefferson, a scientist for Lawrence Livermore Laboratories, worries that voters may have a false sense of security when using DREs.

"When you touch the screen of a voting machine and you see the name light up it engenders confidence in you that your vote has been recorded but the fact is you actually don't know what happens," he said.

Jefferson was a member of the California Secretary of State's committee on Touch Screen Voting.

He went on to describe the approximately 100,000 lines of code between the screen and the internal memory of the machine, and how it is impossible to know how the computer is processing the vote.

But Alfie Charles, spokesman for Sequoia Voting Systems, a leading producer of DREs, believed security concerns were exaggerated.

"To suggest to people or to strike fear in people's hearts that the electronic systems deployed today are reckless uses of computers that can be hacked by any 10th grader with a laptop is I think a disservice to the voting public," he said.

Charles also countered that traditional methods of voting had their own risks as well, citing the recounts and "hanging chads" in Florida in 2000, and recent controversy in the Washington State Gubernatorial race where discrepancies existed between machine counts and hand recounts of votes.

"I think what we often lose sight of is the much greater level of security that's involved in the electronic voting system when it's done properly than it is to the paper ballot," Charles said.

One reform that each panelist supported was the development of a voter-verified paper trail, similar to a printed receipt to be used with electronic voting systems.

"You don't want to have a voting system that leaves ambiguity," Charles said. "I think that's what electronic voting systems with paper trails do: they take the ambiguity out and provide voters with that confidence."

California is taking national leadership on this measure. Beginning Jan. 1, 2006, the state will require a paper trail for all votes cast on electronic voting machines.

While panelists applauded this, they worried that many states would not adopt the same standards.

"You really do need national standards. It's not a tenable situation to have 50 different standards for election equipment," Jefferson said.

Jefferson stressed the necessity of increased communication between voting officials and engineers who choose to "stay in the lab."

Charles acknowledged the imperfections of electronic voting, but argued that they are partly indicative of the voting process as a whole, and that Sequoia was making necessary changes.

"If we could invent the perfect voting machine it would be nice if we could also invent the perfect process, the perfect election official, the perfect poll worker and the perfect voter," he said.

"But there are deficiencies - we're doing our best to add features so that we can meet all the market demands."

* Contact Ryan Groshong at (408) 554-4546 or rgroshong@scu.edu.