Monday, December 13, 2004

Update on the Clint Curtis story

[Note: This is a revised version of a posting from Dec. 7]

blackboxvoting.org

Update on Clint Curtis story,
SATURDAY, DEC 11, 2004

Why the Feeney vote-rigging story sounds like disinformation, as Wayne Madsen writes it

The story hangs together better at BradBlog.com.

ABOUT DISINFORMATION: Like a good lie, it has elements of truth. Trouble is, the truth in Madsen's story doesn't relate to the nuts and bolts of the story.

DISINFORMATION IS DANGEROUS TO THE CLEAN VOTING MOVEMENT: Getting the facts is tedious, unexciting work, consisting of auditing and personal interviews, and it takes time. Many Americans want a magic bullet, a single shot that will blow the lid off everything at once.

That's risky. If the mainstream media continues to be bombarded with stories that sound credible, but aren't, when the real thing comes down the pike it will be ignored.

While MSNBC's Keith Olbermann and I had a run-in recently, I agree with Olbermann's earlier critique of the Madsen homeland security story, and this new Madsen story is just as weak. Most of both Madsen stories are bait and switch. Madsen wanders all over the place, recapping unrelated information from real news agencies, piggybacking onto their credibility, with only the most tenuous ties to what he is actually trying to prove. The work done on BradBlog is much more focused, and Brad seems to be a responsible researcher.

========================================

In my original critique, I raised questions about the Feeney vote-manipulation story; some of them related to Madsen's work. Brad Friedman, the author of BradBlog and the primary researcher for more credible work on Curtis, answered my original questions here. I have updated this section.

1. Madsen's article implied that Curtis's vote-rigging program was used in elections. Brad Friedman correctly points out that the Clint Curtis affidavit explains that he designed a prototype and did not put it into machines. (Many people have written vote-rigging prototypes, and the writing of a program doesn't prove anything about the integrity of the 2004 election.) The issue then becomes: Are Curtis's allegations about Tom Feeney correct?

- Documents do confirm that Curtis worked for Yang Enterprises, and that Feeney was involved with Yang. Documents do not confirm that Curtis met with Feeney and discussed vote-rigging. Curtis names witnesses in his affidavit, which is a good sign. The witnesses have not confirmed the story, yet.

2. I mentioned a second problem, in that several of the Florida counties used different software in 2000 than they do now, and that various Florida counties use different manufacturers and different systems. Writing one program that would tamper with ES&S punch cards and Diebold optical scans at the same time is unrealistic. However, since Curtis says he did not insert the software into any voting system, this is (almost) a moot point.

- The counties Curtis alleges Feeney wanted to rig were Miami-Dade, Broward, Palm Beach. The first two used punch cards in 2000, switched to ES&S touch-screens in 2002, and used ES&S touch-screens in 2004. Palm Beach County used the infamous "butterfly ballot" in 2000, and switched to Sequoia touch-screens in 2002, and used those also in 2004. The Sequoia system has significant differences from the ES&S system, and the same software would not likely work for both

- Note that the Wayne Madsen article does a bait and switch when he discusses Volusia County. He starts by saying it is Feeney's district, and then actually goes on to report a story broken by Black Box Voting in October, 2003, about minus 16,022 votes for Bush in Volusia -- which appears to have nothing to do with the Feeney story. BradBlog takes care not to draw conclusions that aren't supported.

3. The techniques used to program a vote-rigging system in the affidavit by Clint Curtis still have some technical problems. Candidate-switching is not difficult, and there are a number of ways to accomplish it. Programmers have pointed out the the use of VB5 doesn't match use of Unix systems, but several programmers I spoke with were unaware that the Sequoia touch-screens, used in Palm Beach, create their ballots from WinEDS, and that program runs on Windows, and is so replete with security problems that the state of Texas refused to certify it. Now, when I get a high-speed document scanner, I'll post the Texas FOIA documents that show how susceptible the Sequoia WinEDS program is to tampering.

4. Most political shenanigans are not conducted by the candidate himself, but by operatives. It is certainly possible for a politician to hold several meetings in which he commits a felony in front of several witnesses, but that's not usually how it is done. A more common technique is an envelope full of cash left in a drawer of an operative, with at least one, sometimes more, buffer layers between the operative and the politician.

Clint Curtis says Feeney himself had meetings to directly discuss election rigging software. Could happen, certainly, but this seems unusual.

But this gets a bit more interesting. As I was checking this out, I got a report from someone completely unrelated, on an entirely different kind of vote-manipulation endeavor, and Feeney's name came up in that, too. So the issue of Feeney's behavior is about as clear as mud.

5. The author says it will be difficult to write a program that will escape notice if the source code is examined. That's not quite true.

I originally wrote that putting a trigger into a program can involve a very small amount of code, hard to detect -- and you can comment the code such that it looks like it is there for another purpose. Also, the certifiers do a slipshod job of code analysis, and you could probably drive a greyhound bus through their examination of the source code.

But I've been receiving e-mails from programmers that point out something even more obvious: by slipping the rig into a .dll, a program that runs in the background in the operating system (which is never examined at all) you can certainly achieve vote-rigging and survive a source code review.

Programmers pointed out to me that Curtis, as a programmer, should have known that. However, according to his affidavit, Curtis got his degree in Political Science and History, not computer science. He was apparently a self-trained programmer. I won't go into the technical merits more here, because if he didn't put the program into voting systems, they aren't relevant.

6. Now, my most significant objection to the story, which goes to Curtis's credibility, still involves his statement on the affidavit saying that he filed a "QUITAM" whistleblower suit, that is "pending." First, he doesn't spell it correctly. The correct spelling is two words, "Qui Tam." Next, Qui Tam cases MUST be filed under seal. If a Qui Tam is filed in Florida, both the evidence and the existence of the case must be sealed, and only the Florida Attorney General can unseal it.

People have written to me to explain that Curtis did file a whistleblower suit, but did so a day after the deadline. That is not a Qui Tam, but an employment-related suit. In his affidavit, Curtis refers to filing lawsuits two different places. One is an employment suit, the other is a "QUITAM" suit. I found documentation of the employement suit and its dismissal, but saw no documentation at all about a Qui Tam suit. That means it's either still under seal, and therefore, by talking about it, Curtis just invalidated the suit and violated a court order, or there is no Qui Tam suit.

Please do show it to me, if you can find it in the dockets and it has been unsealed.

Black Box Voting board member Jim March and I filed a Qui Tam suit in California in November 2003, against Diebold Election Systems. Using a California law, we refused to seal the evidence, but still had to keep the existence of the case under seal. It did not come out from under seal until the California Attorney General got the court to unseal it, and the Associated Press covered the unsealing of the case. You cannot keep the unsealing of a Qui Tam case away from the press. The press has mentioned no FDOT Qui Tam.

This goes directly to Curtis's credibility. I was not able to get hold of him today, and I will keep trying tomorrow, so that we can learn the answer to this.

There are two other credibility-checking questions I need answered. First, a small scrambled egg on my face: I wrote "Court documents refer to a judgment against Curtis for copyright infringement. Actually, the court documents referred to might have been papers filed by an opposing party, i.e. Yang Enterprises, alleging the copyright infringement without proving it. According to the Daytona News-Journal, Yang says Curtis was "successfully sued" over copyright infringement.

Parsing words here: "successfully sued" may mean there was a judgment entered, but according to Brad Friedman, Curtis says the case was settled out of court without either side paying the other. ("Each side paid their attorney's fees and went their merry way.") "Successfully sued" could also mean an out-of-court settlement in which Curtis paid a settlement to the other party. It is really stretching it to interpret "successfully sued" as simply filing a case. The term "successfully sued" could be a smear by Yang or Feeney, or it could be that Curtis didn't fully disclose the problems with the copyright infringement case.

Because this goes to credibility, and in this case credibility is extremely important, the next two questions that must be answered are: Who was the former employer and what were the real terms of the settlement or judgment?

One more credibility test: When Curtis lived in Illinois, he ran for office as a Republican. While discussing his run for office in a letter to the Bloomington Pantagraph he accused a local attorney of stealing $28,000. The accusation might be accurate, since it apparently was an embezzlement, and those are much more common than people realize. I'd like to know the names of the attorney and the injured party.

My gut tells me -- but this is only speculation -- that Curtis was correct in blowing the whistle on the attorney for misappropriating $28,000. I say that because I've seen written up several financial fraud cases, met the embezzlers, it happens frequently. If he blew the whistle on a $28,000 theft and his charges were correct, that would shore up his credibility on the Feeney story.

I announced on a national radio show Friday that I will be happy to take what you folks throw at me, if I am wrong on these points. In the mean time, because the implications of this story are so significant, I think we need to continue to exercise caution and get the story to the point where it is truly bulletproof.

-- Bev Harris