|
Electronic voting paper trail idea attacked { December 21 2006 } Original Source Link: (May no longer be active)
http://news.bostonherald.com/politics/view.bg?articleid=173293http://news.bostonherald.com/politics/view.bg?articleid=173293
Paper jams hamper electronic voting
By Associated Press
Thursday, December 21, 2006 - Updated: 01:33 PM EST
GAITHERSBURG, Md. - The paper ballots and hanging chads that marred the 2000 presidential election have almost vanished from polling places, replaced by electronic-voting machines that are supposed to eliminate recount chaos.
But now election directors have a new worry: printer jams.
The new machines spool out a small paper receipt of each vote cast to verify the machine correctly recorded the vote and to provides a hard copy during a re-count.
Some states like Maryland have been using paperless systems using touch-screen ATM-like computers that record and tabulate votes. But that has produced its own problems and legislation is likely to be filed in Maryland next year to switch from touch-screen to optical-scannning devices, leaving a paper trail.
In Cuyahoga County, Ohio, a manual count of the paper record during the May primary didn’t match the voting results tallied by touch-screen machines.
Machines in some California, Missouri and Mississippi precincts jammed. In Guilford County, N.C., where the paper record would be used in a recount, an audit of a sample of machines showed 9 percent of printers that were supposed to record touch-screen votes either didn’t work properly or had paper problems.
"How many votes were lost as a result of that, with the printer chewing it up?" asked George Gilbert, elections director for the county that includes Greensboro, N.C. "If you don’t have a complete paper record, you can’t use it for a recount."
Paper trails have other shortcomings. Blind voters can’t read the paper to verify their votes were correctly recorded. And a paper printout from a touch-screen machine could be used to tell how a person voted, compromising privacy.
Even electronic voting critics who have long sought a paper record or other way to independently verify voting machines say the current paper trail systems are inadequate.
"This isn’t what we had in mind when we called for paper," said Johns Hopkins University computer scientist Avi Rubin, who has studied the security of voting machines. "I have yet to see a paper trail system I like."
After the problem-plagued 2000 election, most states fled voting systems that use paper ballots, spending millions of dollars on new electronic voting systems that were paper free. These include the touch-screen devices.
But critics claimed those computerized systems could be easily hacked or malfunction, altering election results without anyone knowing. They also fretted that some voting machine manufacturers don’t provide access to the computer coding of the machines, making it difficult for outsiders to check for glitches.
In response, many states have gone back to adding paper to voting. Printers that spool out a thin paper tape similar to an ATM receipt were added to touch-screen machines. Other states bought optical-scan machines where voters fill out ballots by hand that are then read by a computer.
Twenty-seven states now require the use of paper records, while another 18 don’t require them but use them either statewide or in local jurisdictions. More than half of all voters used machines with paper records during the 2006 elections. Five states - Maryland, Delaware, Georgia, Louisiana and South Carolina - use touch screens without paper trails.
In Congress, Sen. Dianne Feinstein, D-Calif., and Rep. Rush Holt, D-N.J., have said they plan bills next year to require paper trails.
Earlier this month, a report by staffers at the National Institute of Standards and Technology recently recommended that states not use voting machines that don’t have a paper trail or other means of independent verification.
An advisory panel to the U.S. Election Assistance Commission recently called for states to change to independently verifiable voting devices, but advised against using the thermal printers now widely used by touch screens.
While many printers performed well during this year’s elections, some had problems. Cuyahoga County, which includes Cleveland, is considering dropping its touch-screen machines after an audit of paper record samples from the May primary found 10 percent of paper ballots were either smeared, torn, crumpled or blank. In some precincts, the paper record count and the machine count were off by considerable amounts.
Diebold Election Systems, which makes touch-screen machines used by Cuyahoga and many other jurisdictions, said most of the errors were caused by election workers, not the printers. Company spokesman David Bear noted other states, such as Nevada and Utah, had few problems with the Diebold printers.
"The technology has proven itself in thousands of elections," Bear said.
New paperless systems that allow voter verification are under development. Some paper trail advocates say optical scan is the best option currently available, despite the fact it could be vulnerable to some of the old errors that voters made on paper ballots filled out by hand. About half of all voters used optical-scan machines this fall.
Maryland spent $70 million on Diebold touch screen machines and some election officials question why it should be abandoned in favor of scanners. "If we wanted to have paper, we could have stayed with the old machines," said Barbara Fisher, elections director for Anne Arundel County, which includes Annapolis, the state capital.
Groups representing the blind have largely embraced electronic voting because of audio features that allow them to cast ballots without assistance.
"We must have a way to independently vote in private," said John Pare, a spokesman for the National Federation of the Blind in Baltimore, arguing that if there is to be a paper trail it must be done in a way to maintain privacy.
© Copyright 2006 Associated Press. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.
|
|
