March 15, 2006
Webb County officials complain about voting machine problems

You may recall that Primary Day results in Webb County were delayed due to voting machine problems. Yesterday, the Laredo County Commissioner's Court held a hearing on this, and it got a bit heated.

When Chris Moody, the Texas director for Election Systems and Software, came to the podium to offer an explanation, he simply stated that they had not expected to use flash cards to record the votes. The plan was to use the personal electronic ballots (PEBs) that they trained county officials to use.

A flash card is installed in each electronic machine, and records the votes cast. A PEB is a device that is used to read all the machines from a particular precinct. Due to a programming error, the PEBs could not be used and tabulators had to read each individual flash card, significantly delaying the vote tally.

After Moody's brief explanation Monday, dead silence ensued. Then Carlos Villarreal, county executive administrator, stormed to the podium and angrily told the court, "I think you need more explanation."

[Commissioner Jerry] Garza responded that he remembered officials with the company coming to Webb County every two weeks to lobby the county to buy their equipment.

"Then we cut you an $800,000 check, and that's it," Garza said. "That's not acceptable.

"It's absurd. You promised us the world, and we got nothing."

Oscar Villarreal, elections administrator, later said county officials were trained how to use the machine three weeks ago, and were not trained to use flash cards because they were only to be used in emergencies.

There was a representative on site Election Day from Election Systems and Software, but Moody said that employee was not trained to use flash cards, either.

When asked what the company would do about the runoff, Moody said he, personally, would be in Laredo to assist with any technical problems.

Part of the reason for the heat was that a couple of the members of the Court had just gotten unelected on Tuesday, so some feelings were still a little raw. If it makes anyone feel better, the legally mandated random audit they did of one race, the Supreme Court Place 2 election, gave the same result as Election Day.

Now any time one blogs about electronic voting machine problems, one will get an earful of suspicion and doubt about them with varying ranges of intensity. My biggest concern is what I perceive to be a lack of redundancy in the system. It's nice to know that if the PEBs fail you could still count the votes from the individual flash cards, but what happens if a flash card fails? I support equipping all electronic voting machines with printers like the ones on ATMs so that each voter also creates a paper record of his or her vote. Whether that serves as a backup/sanity check for the system or they're counted as the "real" votes doesn't matter to me so much as the idea that there's a separate system in place to ensure integrity in the event of a problem with the machinery. Admittedly, this would not have sped things up in Laredo, but at least it would have put to rest any questions about the actual vote totals.

It should be noted, as this earlier story points out, that paper ballots are not immune to problems, either.

The final vote for 49th District Court judge was not released until Wednesday night because of a problem in Zapata County.

Doroteo Garza, Zapata County Democratic Party chairman, said that while the county's delay was the result of new equipment, it was not the fault of the new electronic voting machines.

It was actually a machine used to scan paper ballots at each of the polling sites that caused the delay. Although machines were used at each of the sites, it was only the one at Precinct 4 that gave officials problems, Garza said.

In Zapata County, when the voter uses a paper ballot it is put into a machine that scans it and puts it in a sealed box. But the machine in Precinct 4 began to act up when a voter's ballot was dropped into the box. The machine began to kick the ballot submitted previously out of the box in order to add the new one.

Garza said those ballots that were spit back out were taken and put in another slot for "spoiled ballots."

Because of the defect, the ballots had to be manually counted, but the numbers didn't add up and Zapata officials had to call technicians in from Heart Graphics, which provided the machines.

"We brought in the experts because they wanted to make sure everyone got a fair deal," Garza said. "That's why it took so long."

Speaking as an IT professional, redundancy is your friend. Equipment sometimes fails. That's just the way it is. You have to be prepared to overcome it when it does happen. Really, I think the biggest problem here was that there was no one from the vendor onsite during the primary, when all this new machinery was being used for the first time. It's nice that they'll be there for the runoffs, but they blew it by not being there last Tuesday. That should have been a requirement in their contract.

Finally, while the focus was on Laredo because of the CD28 primary, Webb County was not the only place to encounter glitches.

The error caused Tarrant County to report as many as 100,000 votes in both primaries that never were cast, dropping the local turnout from a possible record high of about 158,103 voters to about 58,000.

Because the errors added votes equally for each candidate, the glitch did not change the outcome of Tarrant County races but narrowed the margin of victory in some statewide races. In the close Republican primary race for Texas Supreme Court, for example, incumbent Don Willett edged past former Justice Steve Smith by only about 1 percentage point with the corrected vote tallies.

The math geek in me feels compelled to point out that if the same number of votes went to each candidate, the margin of victory would have remained the same if by "margin of victory" you mean "vote differential". If instead you mean "percentage differential", it would have actually increased once the repeat votes were removed. This is an application of what an economics prof of mine called "the Pete Rose Theorem". It works like this: If Pete Rose enters a game batting .300 and he goes one for four, his batting average at the end of the day will decrease; this is because "one for four", or .250, represents a smaller ratio of hits to not-hits being added to his existing total, which in turn reduces his overall ratio, which is expressed here as his batting average. Had he gone one for three (.333), his average would have gone up. In this case, since Willet's percentage before the repeat votes were added was above 50%, adding the same number of votes to his total and to Smith's (i.e., 50% to each) would have lowered Willet's ratio of votes to not-votes, while raising Smith's and thus do the same to each person's percentage of the vote.

To make up an absurdly simple example, imagine Willet beating Smith two to one. That means 67% for Willet and 33% for Smith. Now add one vote to each person, so that it's 3 to 2 in favor of Willet. Willet's percentage then declines from 67 to 60, while Smith's climbs from 33 to 40. Do it again so that it's 4-3 for Willet, and the percentages become 57-43. Play around with it yourself and you'll see what I mean. Bottom line, Willet's percentage of the vote was higher, and Smith's was lower, after the extra votes were removed, not the other way around.

Anyway. Thanks to commenter Chito for pointing me to the original Laredo Morning Times link.

Posted by Charles Kuffner on March 15, 2006 to Election 2006 | TrackBack

What? A post on E-voting and not one single comment? Where is everybody?

Now any time one blogs about electronic voting machine problems, one will get an earful of suspicion and doubt about them with varying ranges of intensity.

There's always a chance that one race or another was hacked, but in this case, the failure actually reduced the chance, since it's plausible that a hacker would choose to alter only the PEB's. Of course, if a hacker were able to compromise the software installed on each machine, then he could alter both the flash cards and the PEB's, so going to the flash cards doesn't eliminate the possibility completely.

In this case, the only thing that raises any suspicion at all is that the early vote was a surprisingly high percentage of the vote total in Webb County. But with the Cuellar-Rodriguez race, at least, there was no huge difference between the candidates' early vote percentages and their primary-day vote percentages; if there were any "phantom" votes added to the totals, they were added to all candidates proportionately. Of course such tampering would've still benefited Cuellar, since
Webb County is his base, but the number of "phantom" votes needed to swing the election would still be implausibly large.

It's nice to know that if the PEBs fail you could still count the votes from the individual flash cards, but what happens if a flash card fails?

Good point. As I understand it that's pretty much what happened in NC in 2004. Election officials were able to determine the total number of votes cast at the failing machine, but the vote differential in one race was less than this number, so they were forced to rerun that election!

I support equipping all electronic voting machines with printers like the ones on ATMs so that each voter also creates a paper record of his or her vote.... Admittedly, this would not have sped things up in Laredo, but at least it would have put to rest any questions about the actual vote totals.

If the printed ballots could be optically scanned, then it might have sped things up, as well as putting to rest the concerns about tampering.

Posted by: Mathwiz on March 15, 2006 5:04 PM

Did these type of problems exist before the Wappointment or is the process really improving?

Posted by: Charles Hixon on March 15, 2006 7:49 PM

E-Voting machines and Lever machines are non-evidentiary. Unaccountable. This is a blatantly fatal flaw to the notion of transparent, honest elections on the face of it.

And, scanning machines like e-voting machines are hackable in many different ways, on site and off site, and only some of the ways are detectable. However, scanning machines do require the physical evidence of paper ballots.

However, in many different ways, scanned paper ballot evidence is prevented from being hand counted on purpose. In Florida 2000, the now apologetic Sandra Day O’Connor helped to stop the never counted over-votes (when both chads are punched and the same name is written in. You know the voter’s intent.)

Now in Ohio where scanning machines were used in 2004, challenges by the Libertarian and Green Party arose because of numerous problems. Senator Kerry had the paper ballots made secure. Even now, Ohio passed laws to try to prevent the counting of those ballots.

And to repeat:

Abramoff's Diebold > Ney's HAVA + Taxpayer Funding > Non-Evidentiary E-Voting = ?

95% of the democratic nations of the world vote with paper ballots which are hand counted in a few hours. Low cost, no complexity, no specialized training, no company personnel, no machines breaking down or hacked, no long lines, no distrust.

Just cheap and easy.

There is no reason for the expense, complexity, secrecy, failure rate, unaddressed problems, partisan personnel, lawsuits, doubt, fury while media and politicians push ...unless...


"If the American people had ever known the truth about what we Bushes have done to this nation, we would be chased down in the streets and lynched." GHWB


Why am I running for Congress?

My name is Clint Curtis. I am considering running for the U.S. House of Representatives in Florida's 24th Congressional District. I will need your help.

I am a former Republican who became a whistleblower that came forward and testified to Congress under oath, regarding the request from a sitting congressman to build a computer program that could, without detection, flip your vote in an election. This manipulation would allow a person or political party to take control of voting machine and decide who wins the election, regardless of the true vote count. In March of 2005, I passed a polygraph test as to the accuracy of my statements.

I believe our district is currently represented by one of the most corrupt politicians in history. I have attended closed-door meetings where Congressman Tom Feeney spoke candidly about altering election results. Feeney is deeply involved with individuals like super-lobbyist Jack Abramoff and has been an active participant in this unprecedented decade of corruption. Listening to Feeney has given me the impression that he never had a vote he wouldn't sell or a law he wouldn't side-step. District 24 deserves an alternative to reelecting a corrupt congressman.

I am running because our democracy insists that Americans know that their vote will actually count and that elections must not be pre-determined by manipulating electronic voting machines. Feeney believes that elections are nothing but sporting events. He believes it is acceptable to fix elections through voting machines, suppressing the vote through all types of creative methods, or in simply letting those in power declare a winner regardless of the vote outcome. He attempted to do this in the 2000 Election.

Feeney's actions show he is a big believer in a controlled concept of democracy --where elections are pre-determined. Feeney's actions would seem to support: "The people who cast the votes decide nothing. The people who count the votes decide everything. " I humbly --and wholeheartedly --disagree". Your vote must be counted!

I have put my name and reputation on the line for several years. I have spoken to the press, law-enforcement officials and even members of the U.S. House Judiciary Committee --under oath and on video tape. Now I am putting my name on the line as a candidate against Mr. Feeney. His deliberate corruption forces me to provide this district with an alternative. I will need your help to achieve honesty in our district.

Feeney personally designed the boundaries of this district for himself when he was Florida House Speaker. It will take all of us to break his hold over our lives and future! Your contribution to my campaign will make a difference. Don't lose your vote as others have. Send your check today, for a better tomorrow. Mail to:
"Clint Curtis for Congress Exploratory Campaign"- P.O. Box 1456 - Titusville, FL 32780 or online.


Blogged by Brad on 3/8/2006 @ 7:18pm PT...

E-Voting Woes: You Ain't Seen Nothing Yet...
Texas Primary Augurs Upcoming E-Trouble in E-Merica...

Texas held their first round of Primary Elections last night since adding new electronic voting equipment to loads of counties around the state. It was one of the first Primary Elections to occur since the Help America Voting Act (HAVA) kicked in this year.

If today's 'Daily Voting News' by John Gideon is any indication, America is in for one hell of a mess this year. Take a look at just a few of the headlines from just the Texas section of today's DVN...and keep in mind that usually the real problems held in 'E-Lections' don't even begin to surface until some time after Election Day...

• TX: Galveston County - New voting system, same problems (Hart eSlate and eScan) LINK

• TX: Jefferson County - Voting woes mar debut of electronic ballots (ES&S iVotronic) LINK

• TX: Jefferson County - Confusion mars new voting process LINK

• TX: Robertson County - Robertson County vote tallies delayed (ES&S AutoMark) LINK

• TX: Tom Green County - Computer problems delay election returns (Hart eScan) LINK,1897,SAST_4943_4522634,00.html

• TX: Webb County - Cuellar defeats Rodriguez in congressional grudge match (Counting problems – ES&S) LINK

• TX: Webb and Bexar County - ‘Glitches' get blame in Webb (ES&S) LINK

We have a feeling it's gonna be a very long year...

March 13, 2006

Is There a Case for Impeachment?

I appeared last week at a Harper’s Magazine forum which addressed the conduct of the Bush Administration and whether it has committed crimes that rise to the level of impeachment. A crowd of 1,500 enthusiastically heard from several experts and me as we discussed the tyranny of a President acting to commit a fraud against the United States.

I need your help to build the case for an impeachment investigation.

I have conducted an exhaustive study on this subject and believe that we must begin the process to investigate this administration for possible impeachable offenses. As former congresswoman Elizabeth Holtzman noted at the forum, a groundswell of public support is necessary for impeachment proceedings to succeed. There was overwhelming public support for pursuing a case against Richard Nixon, while these same conditions did not exist when the Republicans tried to impeach President Clinton. Public opinion is on our side and now is the time to make a difference.

Become a Citizen Cosponsor
I have introduced House Resolution 635 to create a special committee to investigate whether the president's misconduct rises to the level of impeachment. 29 Members of Congress have already joined me as cosponsors. Lend your support. Help build momentum to enact this legislation by joining the more than 42,000 concerned Americans who have already signed on as Citizen Cosponsors at:, or encourage your friends to sign up.

Watch the Harper’s forum
You can watch “Is There a Case for Impeachment?” online here:

Read the Report
Read the Constitution in Crisis, A 182-page report with over 1,000 footnotes covering the entire spectrum of deception, manipulation, torture, retribution and cover-ups by the Bush Administration. It can be found online here:

Write a Letter to the Editor
Visit this link to write a letter to the editor in support of House Resolution 635.

Thank you for working to help build a better democracy.


John Conyers


Posted by: Prove Our Democracy with Paper Ballots on March 16, 2006 4:35 PM