Talk:2002 Georgia general election
Auditing the patches
Normally, illegal voting machine patches are never available for public scrutiny. But Diebold posted them on an FTP server whose contents were eventually downloaded and shared by Bev Harris. That means that all of the patches, such as the Windows CE modifications, should be available to reverse engineer. If the 2002 Georgia election was indeed rigged, this could provide evidence of fraud.
Harri Hursti apparently looked at rob-georgia.zip, though his conclusions are unknown. Jim Bryant on a Minnesota email list claims to have found evidence of vote rigging in the AccuVote software: a hidden screen for stealing votes (like what Clint Curtis alleged Tom Feeney commissioned in 2000?).
If the GOP wants to rig elections, and for those who think I don't know what I'm talking about, here's an example: 6:11:36am stimpy(29): d rob-georgia.zip -rw-r--r-- 1 jbryant staff - 64537508 Feb 22 2003 rob-georgia.zip 6:11:45am stimpy(30): md5 rob-georgia.zip MD5 (rob-georgia.zip) = e0dac9db0a3d88c8ba3a652776326879 or maybe this might be closer to home: 6:11:51am stimpy(31): d *anoka* -rw-r--r-- 1 jbryant staff - 2881 Feb 23 2003 195anoka.zip 6:12:40am stimpy(32): md5 *anoka* MD5 (195anoka.zip) = 15a85e8b77410e54fefbd83b66e5cfe6 For those of you that don't speak computer, that's just a sampling of Diebold's Accuvote software that was made to steal votes in 2002. Interesting hidden third screen that allows votes to be tampered with without auditing and without changing anything but the counters that get tallied for the official results. Very interesting stuff, and I was one of the first to analyze this. I can't distribute this because of the injunction though. You see, once word fgot out that we found the hidden third screen used to steal votes from Max Cleland, they got one of Daddy Bush's activist Judges to declare it proprietary, and prevent it's further distribution.. After all, the Johns Hopkins and SAIC studies were very damaging to Diebold [they came after us], and the injunction came out about that time so that more peoplke didn't see it, after all we were setting up a website that showed this, and anmnotated it so that even laypeople could see how Diebold steals votes from us.
This claim hasn't been verified. However, the dates on the files do match when the unsecured Diebold FTP server was exposed, and an earlier message by Bryant confirms his work in investigating the Diebold software:
To start with, I'm a bit of a rabble-rouser myself, and have been involved in Democratic politics actively since the 1996 election, and at this point, barring getting a real day job again is about all I do now. My professional background involves like going on 18 years of hardcore experience in the IT field ranging from programming to systems administrator for organizations ranging from the us navy's pacific missile test center to internet companies to Sprint, and generally would bore most of you to tears, so I'll not discuss that here. [...] I was also one of the first source code reviewers of the Diebold 2002 Source Code for AccuVote systems, and in particular the ones used in both the State of Georgia as well as Anoka County, MN. For all of you DU'ers out there, yes, I'm that Jim Bryant, I'm just one in a long line of people Bev Harris has slandered, and although I could say a few words about that [rhymes with "witch"], I won't, because she has done some good with her self-aggrandizement by drawing attention to the voting machines issue. On a side note I would like to thank the people of DU for perm-banning her sorry arse, and also would extend a non-invitation to more than [not]willingly give a copy of what has an injunction against it's distribution to some of the brighter coders/crackers in the party for their own review, they will agree that what I didn't give them is what we said.
It would be worthwhile to obtain the 2002 Georgia patches and reverse engineer them. Unfortunately, I haven't yet been able to find the FTP files.
A very weird pattern showed up in the election results: Cleland won north Georgia's heavily-Republican areas by 14% more than the primaries would indicate, while Chambliss won south Georgia's heavily-Democratic areas by 22% more than the primaries would indicate. What could be the cause of this shift? There seem to be three potential explanations:
- Demographics: Quite possibly, but nobody has yet analyzed this.
- Machine error: The machines may be error-prone and lose votes, something that similarly affected Democratic and Republican strongholds. However, it would take a massive amount of voting machine failures to cause this, and there doesn't seem to be any documentation of widespread breakdowns. Also, how does this explain that the disparity only appeared in 101 counties, with 58 counties matching the primary results?
- Fraud: The machines might have been intentionally programmed to create a result like this that ultimately gave the election to Chambliss. However, why exactly would a fraud scheme increase both candidates' vote shares in the strongholds of their opponent? It would make more sense to unidirectionally steal votes.
Cynthia McKinney defeat
Cynthia McKinney was defeated in the 2002 primary, following controversial comments about fraud in the 2000 election and criticism of George W. Bush (including his role in 9/11). It was believed at the time that thousands of Republicans crossed over into the Democratic primary to vote against her. However, since paperless Diebold touchscreens were first introduced into the state in 2002, and McKinney apparently made an enemy of the Bush Administration, election tampering might have also been the cause of her losses. Several Diebold patches (including the clock fix) were applied before the primary. It should be analyzed whether there were any irregularities in McKinney's 2002 loss. She got her seat back in 2004 and was defeated in the primary again in 2006, after which she hinted that she believed her election was stolen through voting machine rigging.
Uncertified patch motivation
Why was Diebold Election Systems avoiding certification? It doesn't appear to be for the sake of convenience, since Talbot Iredale explicitly said (TODO: link memo cited in Ch.11 of Black Box Voting) that they didn't want Wyle Labs to look at the patches. They had some unknown reason to want to avoid scrutiny, specifically with their Windows CE modifications. And why was Bob Urosevich personally involved with delivering patches?