Technology gets passing grade at this election

PASADENA -- Problems at polling sites across the nation Tuesday were less technological in nature than many had expected, based on preliminary evidence from the election, voting technology experts said Wednesday.
"What we saw yesterday was that voting technologies across the board seemed to have worked very well,' said Michael Alvarez, a Caltech professor of political science. "We did not see the technological glitches that I think many people anticipated before the election.'

Alvarez and the other experts are members of the Caltech/MIT Voting Technology Project, a research group of computer scientists and political scientists brought together by the presidents of the two universities in December 2000 in an attempt to prevent the problems that plagued that year's presidential elections.

Alvarez, who observed the voting process from precincts in Riverside and Orange Counties where voters used touch-screen voting machines, said the major problems Tuesday seemed to have been more mundane procedural difficulties.

From registration errors and confused poll workers to provisional ballots and long lines, "a lot of the bottleneck occurs when voters simply try to check in,' Alvarez said.

Los Angeles' Registrar-Recorder/County Clerk's Office echoed those sentiments. Spokeswoman Grace Chavez said 59,594 voters in the county used touch-screen machines to cast their ballots at the 17 sites open between Oct. 20 and Oct. 29. And to her knowledge, she said, there were no major problems with those machines or the county's new InkaVote system that replaced the punch card system and its infamous chads.

The county's major problem during this election, said Kristin Heffron, chief deputy registrar-recorder, was a result of the new cut-off date for new or updated registrations. This year, for the first time, that deadline fell 15 days rather than 29 days before the election.

"We ended up in the last 15 days with 337,000 voter registrations to process,' Heffron said.

That number, combined with the more than 700,000 absentee ballot requests, stretched the resources and capacity of the staff too thinly in those final weeks leading up to the election.

As a result, officials said, many new registrants showed up to vote only to find their names missing from the rosters. Those people were given provisional ballots and Heffron estimated they were about 100,000 to 200,000 in total.

Jonathan Katz, another Caltech professor and member of the voting project, said there are numerous weak links in the voting process and that focusing solely on security concerns posed by any type of machine can be paralyzing.

"No voting technology is a panacea,' he said. There will always be potential security problems but there are many other flaws in the system that should be looked at with equal concern, he said.

The project's co-director from MIT, Ted Selker, said current voting technologies have fewer problems than those from 10 years ago and will continue to improve. Largely, he said, "people are the issue,' not the machines. The voting technology experts are awaiting final results and voting data. The project's first major report, which found that four million to six million votes were lost in the 2000 election, came out in July 2001. "Hopefully we'll get something out before that this time,' Alvarez said.

-- Kimm Groshong can be reached at (626) 578-6300, Ext. 4451, or by e-mail at kimm.groshong@sgvn.com .