YIKES! And who is Dido Dupré?

"Is it possible that someone could hack voting machines and rig an election? Election officials insist that they ... train poll workers to recognize signs of machines that had been tampered with. ...[A]lso claim ... machines are carefully watched. Neither is entirely true." Clive Thompson in New York Times Magazine, January 6, 2008.

"Lost." Well, maybe.

"Destroyed." Hmmmm.

"Miscounted." Uh oh.

"Wrongly Attributed." Really?

"Hacked." Enough! Stop, already!

----

OPTICAL SCAN TO THE RESCUE [Ohio] Secretary of State Jennifer Brunner has made clear she does not trust touch-screen machines and wants all 88 Ohio counties to switch to optical-scan systems by the November election. From http://www.ohio.com/.

----

WHAT?!? ...Optical scanning is hardly a flawless system....

The machines also need to be carefully calibrated so they don't miscount ballots.... And the machines do, in fact, run software that can be hacked." Clive Thompson. Same NY Times Magazine January 2008 article.

The Times article reports error rates of .75% "at its highest." Other studies at MIT and Harvard suggest the same.

GOOD ENOUGH?

Some recent results: Taiwan 2004 President Chen Shui-bian beats challenger, Lien Chan, by 0.2%. 1952 Tennessee Dwight Eisenhower: 50.0% Adlai Stevenson: 49.7% (0.3%) 1968 Nationwide Richard Nixon: 43.4% Hubert Humphrey: 42.7% (0.7%) 2000 Nationwide George W. Bush: 47.9% Al Gore: 48.4% (0.5%)

---

WHAT?!?

"In November 2006, about 18,000 undervotes were reported in Sarasota County in the race for Florida’s 13th Congressional District."

"A test to determine whether a system can handle the expected volume of activity is commonly referred to as load testing. We found that ballots used for load testing during the certification testing were machine-generated using a testing program ..., i.e., users do not touch the screen to make a selection and cast a ballot."

"...[T]he generalization of the results ... for parallel testing cannot be supported because the sample drawn was not random and the sample size was too small."

From GAO [United States Government Accountability Office] Statement Before the Task Force on Florida-13, Committee on House Administration, House of Representatives. August 3, 2007.

---

Who is Dido Dupré?

---

WHAT?!? "A marginal mark is a mark within a voting target that does not conform to vendor specifications for a reliably detectable vote. The word "marginal" refers to the limit of what is detectable by an optical scanner, not the margin of the page. Marks that are outside of voting targets are called extraneous marks. A marginal mark is neither clearly countable as a vote nor clearly countable as a non-vote. It is an ambiguous vote, analogous to dimpled chad on a punchcard [emphasis added]."


"Although the criteria are not necessarily simple, vendors are required to specify what constitutes a reliably detectable mark versus a marginal mark .... If this cannot be accomplished, then the voting system is counting votes using a mystery algorithm. Such a system is not certifiable."



Short of banning the use of manually-marked paper ballots, which would create a crisis for absentee voting, the best we can do for this central count case is prohibit bias in the detection of marginal marks ... and advise that the detection of marginal marks be made as repeatable as possible." From Discussion Paper on Marginal Marks posted by the National Institute of Standards and Technology (http://www.nist.gov/). Discussion Draft. Context Description: Posted Dec. 1, 2006.