Posted on 10/02/2003 9:52:12 PM PDT by ambrose
NBC 4
POSTED: 1:43 p.m. PDT October 2, 2003
UPDATED: 2:01 p.m. PDT October 2, 2003
But some computer scientists fear more trouble with electronic ballots. With almost one in 10 registered voters using touch-screen machines that don't automatically produce paper printouts, they say a legitimate recount would prove impossible.
County registrars and executives at the companies that sell and update the electronic voting machines say scientists' concerns are overblown and irresponsible.
None of the elections officials who supervise the 50,000 touch-screen machines serviced nationwide by Diebold Election Systems has reported glitches or hacks that have resulted in known miscounts or fraud, said Mark Radke, director of the voting industry division of North Canton, Ohio-based Diebold.
But David Dill, a computer science professor at Stanford University and a leading skeptic of touch-screen voting, is urging voters in the four counties using touch-screen terminals, Alameda, Riverside, Shasta and Plumas, to vote with absentee ballots, which use optical scan systems and provide paper ballots. He fears falsification or deletion of votes on touch-screen systems.
"You can't do a meaningful recount if the question is about the integrity of the voting machines themselves," Dill said.
According to a July study by Johns Hopkins and Rice universities, which analyzed Diebold source code that had been posted anonymously on the Internet, any clever hacker could break into Diebold's system, which is based on Microsoft Windows, and vote multiple times. Researchers found it was theoretically possible to insert "back doors" into software code that would allow hackers, or insiders, to change future voters' choices and predetermine the outcome.
In the case of the California recall, Dill said, changing code in a voting machine could make every "no" vote be recorded as a "yes," or vice versa. Every vote for Republican state Sen. Tom McClintock could be recorded as a vote for Democratic Lt. Gov. Cruz Bustamante, and without a paper record, the voter would never know.
Activists are demanding that ballot machine vendors include printers that produce paper receipts for every ballot, so citizens can confirm that paper results match their touch-screen choices. Receipts would go into a county lock-box for use in recounts.
If voters spot a problem, they could destroy the bad ballot and make sure they've voted properly before leaving the polling station, scientists said.
"It's horrifying and ridiculous that these machines don't have a voter-verifiable audit trail," said Rebecca Mercuri, a Harvard University research fellow who specializes in computer security and voting systems. "These four counties have taken the public's right to a re-count away."
Riverside County registrar of voters Mischelle Townsend she said she had "total confidence" in the electronic system used by the county's 650,000 voters. On election day before the polls open, the county tests all 4,250 touch-screens for logic and accuracy, confirming that a "yes" vote is recorded as a "yes," Townsend emphasized.
"The candidates have been very pleased with all our results," said Townsend, who has supervised 19 touch-screen elections and five recounts since November 2000. "The machines have always been adjudicated to be reliable and accurate. There's never been a single incident of what the scientists fear."
Electronic voting advocates acknowledge no system is perfect but said touch-screens are better than older technology.
A survey by CalTech and MIT found 6 percent of votes cast nationwide in the 2000 presidential election may not have been counted because of problems with antiquated systems. Punch-card machines, used by 44 percent of California voters, were the subject of a federal challenge to the election by the American Civil Liberties Union.
Each touch-screen machine has an "integral thermal printer" -- the same technology handheld devices use to spit out credit card receipts. But in their current configurations, the printers cannot produce receipts for voters to verify at the polls.
After polls close, elections officials make another accuracy check. They get printouts for 1 percent of voters in every precinct, and compare the digital record with the printouts.
In some states, margins of less than 1 percent automatically trigger recounts. In California, a recount happens only when a candidate or party demands one.
In a standard recount, officials would simply add up again the results recorded digitally on the machines' memory cards. If a more intense hand recount is ordered, as sometimes happens when the results are extremely close or differ from expectations -- officials would print the digital image of every ballot cast, and count them by hand.
"It's in essence a hard copy of the ballot," Radke said. Diebold provides equipment to Alameda and Plumas counties. Riverside and Shasta use equipment by Oakland-based Sequoia Voting Systems.
But simply recounting printouts wouldn't catch votes that were improperly recorded because of software bugs, malicious hackers or other abuses, activists said.
"The only thing this process would show is if your printers were working," said Kim Alexander, president of the nonprofit watchdog group California Voter Foundation. "If the machines fail to capture ballots accurately to begin with, the printed images produced after polls close will only produce those erroneous records, and the voter is no longer around to correct it."
The ACLU is watching closely for evidence of voter disenfranchisement, as is the California Democratic Party, which began soliciting $100,000 this week for a "No More Floridas!" campaign to scrutinize alleged violations.
The computer scientists will be watching as well, looking for statistical anomalies in touch-screen counties.
Threats of a recount worry election officials. Whoever requests the re-count must pay for it, but the lengthy process could overlap with some county elections in early November.
"The very thought of a recount, it's chilling," said Alameda County assistant registrar Elaine Ginnold. "We're all hoping there will be a huge margin because a recount would plug things up for quite a while."
Copyright 2003 by The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.
|
|
![]() |
FreeRepublic , LLC PO BOX 9771 FRESNO, CA 93794
|
It is in the breaking news sidebar! |
In a properly-designed system, there should be less than one undetected read error per million ballots (there may be some ballots which the system flags as unreadable; those should be handled by hand). Thus, if 100 ballots are selected at random, all 100 should match perfectly.
If discrepancies are noted between any of the physical ballots and the electronic copies, then all the paper ballots should be rescanned and compared against the original scan. All ballots which don't match should be checked to determine the cause of the mismatch. Once any problems are repaired, a sample of 200 ballots should be taken; all 200 should match perfectly. If not, continue problem analysis.
If the margin on a race is less than 3%, then using a sample of more than 100 ballots would probably be a good idea. But the odds of a 3% error rate (whether intentional or not) being undetected in a 100-ballot sample are less than 5%.
To my mind, designing a system to allow hand auditing would be much more practical than allowing only machine recounts using machines which may be just as defective or rigged as those on the original count.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.