A Soft Touch -- for Voter Fraud? (BBV - WP.com) 8/3/2003

[Bushfire]

Sunday, August 3, 2003; Page B06 Editorial page

MARYLAND RECENTLY committed more than $55 million to buy touch-screen voting machines. Now a study from researchers at Johns Hopkins University suggests that the machines may be vulnerable to fraud and other abuse. The warning deserves to be taken seriously.

State election officials are standing by their new system. It is intended to be efficient, accurate, consistent across jurisdictions and easy to use -- to guarantee, in other words, that Maryland does not become a Florida 2000. The Diebold Elections Systems machines already have been used in Montgomery, Prince George's, Allegany and Dorchester counties, as well as in Georgia, California and Kansas. State election officials say the system performed well in those places and during a rigorous battery of tests. They are aiming for almost all of Maryland to have the machines installed by March.

Researchers at Johns Hopkins's Information Security Institute say that the system is prone to errors and tampering. Because the machines use smart-card technology, they say, someone could program a counterfeit card to cast numerous ballots at a time. In addition, they raise concerns that voting officials could, accidentally or intentionally, program machines to count ballots cast for one candidate toward another's tally -- and voters would never know. Diebold officials say the researchers were studying a year-old, superseded version of their computer code. They also say that the kind of cheating that the researchers may have managed in a laboratory environment could not take place under real-life conditions.

more...

http://www.washingtonpost.com/wp-dyn/articles/A15485-2003Aug2.html?referrer=emailarticle

from the Diebold website they link a pdf article from Ohio minority owned newspaper "Call and Post" defending their company, and repeating the lie that the machines do not access the internet.

http://www.diebold.com/

http://www.diebold.com/checksandbalances.pdf

I can't wait to hear reports from their investor analyst meetings today & tomorrow...

[rogerashton]

1. There is a right way to use computer technology in voting.

It is:

1) use the machine to assist the voter to make her choice.

2) print out a paper ballot.

3) count the paper ballots.

[Dover]

2. ...and 4) For voting officials in Florida: If your district is primarily

Democratic, empty contents of your ballot box in the swamp and claim an alligator ate it...(those repug alligators are viscious, with big appetites).

[SharonAnn]

6. Yes, I think of it this way - The printed ballot is the actual ballot and

the touch screen is just used to create the ballots and to count the ballots.

Comparing to punch cards, it's actually a combination of the device where the voter punches the card and the device that "reads" the cards. And, in any case where there's a spot audit or a question, the actual PAPER ballots can be hand counted to verify the count.

[Bushfire]

3. Quote from the "Call and Post"

"So it was most unusual to find the industry leader being discredited in a report created by a couple of Johns Hopkins Univ. grad students. In what appeared to be a code-cracking homework assignment, the students prepared a report that directly contradicts information provided by Robyn M. Downs, elections administrator of Prince George's County, MD..."

last I checked these "grad students" are actually Avi Rubin, technical director of the John Hopkins Information Security Institute, Rice University computer-science professor Dan Wallach, and two John Hopkins grad students Tadayoshi Kohno & Adam Stubblefield. I would heed the security concerns of information security EXPERTS before Robyn Downs, Elections Administrator of Prince George's County MD. Nowhere does Diebold, or Call and Post backup their claim that this was just a homework assignment. The burden of proof regarding security still lies with Diebold, and other voting machine manufacturers.

http://reuters.com/newsArticle.jhtml?type=technologyNews&storyID=3155955

"It's unfortunate to find flaws in a system as potentially important as this one," Tadayoshi Kohno, a graduate student at the John Hopkins Information Security Institute, said in a telephone interview.

While researchers said they did not know for sure whether the software had been used in voting situations, they said comments and copyright notices in the code indicated that it was legitimate.

"I have no proof that this is what's running in their systems, but I would bet it's pretty close," said Avi Rubin, technical director of the Information Security Institute.

Rubin, Kohno, Johns Hopkins graduate student Adam Stubblefield, and Rice University computer-science professor Dan Wallach said they had uncovered several flaws in the system.

[BevHarris]

4. I spoke with the editor of that publication today.

She was quite concerned about the story, and urged me to call the writer and straighten him out. The writer is apparently a friend or relative of the person who is an election official that he quoted.

It struck me that the writer put the technical expertise of an elections director on the same level as the technical director of the John Hopkins Information Security Institute and a Rice University computer-science professor.

It also struck me how desperate Diebold must be to use this tiny publication (a Don King production, but I'm not knocking him, think he's a brilliant and entertaining guy, actually) -- but really, is this the best they can do? It's at the top of the Diebold web page!

Bev

[Bushfire]

5. Interesting comments on the Diebold Yahoo message board...

http://finance.messages.yahoo.com/bbs?.mm=FN&action=m&board=4687614&tid=dbd&sid=4687614&mid=6575

No Comment from Diebold????
by: dbolder69 (F/Canton) 07/25/03 05:54 pm Msg: 6575 of 6598

The newspaper article indicated that calls to "a Diebold spokesman" were not returned.
What are they hiding? Why are call from the media not returned?
A scandal concerning electronic voting machines is not going to help voters or shareholders. This whole thing could blow up and cost us a bundle.
Who is responsible?
What did Wally know and when did he now it?

[ramblin_dave]

7. The WP editorial is misleading

The WP editorial says:

No purely electronic system is fail-safe; most technological experts agree that an old-fashioned paper trail is the best defense against fraud and failure. Unfortunately, a paper record that a voter could remove from a voting site would open the door to vote buying or other coercive tactics.

The WP is misleading its readers into thinking the paper trail experts recommend is a fraud prone receipt that voters carry away.