Monday, September 22, 2003

Manipulating the vote

I've been tracking the issue of the potential abuse of electronic voting techniques as a way of altering the outcome of elections, but haven't blogged about it till now because I wasn't sure if there was any there there.

It's starting to appear that there is indeed.

Salon has a piece up tonight that nicely brings together a lot of the pieces of the puzzle:
An open invitation to election fraud

The bulk of the piece is an interview with a fellow Seattleite named Bev Harris who has enough expertise in the systems to uncover serious questions about the systems that may be in place in many voting booths come November 2004. Here's what Harris says:
Well I work with about 22 computer programmers who have been looking at this stuff -- I'm not that brilliant. Immediately when they began looking at the GEMS program they began commenting on the fact that it has no -- it's something called referential integrity. And what that means is that there are many different ways that it can become vulnerable to hacking. It has to do with how one part of the database is hooked into the next part.

I got a call from one of our more brilliant computer programmers -- he's got quite a few advanced degrees -- and he called me on a weekend and he said, "I want you to go to your computer." And he walked me through it just like a support tech does -- open this panel, click this, do this, do that. And as I'm doing this it was appalling how easy it was. Once you know the steps, a 10-year-old can rig an election. In fact it's so easy that one of our activists, Jim March in California, put together a "rig-a-vote" CD. He's been going around showing it to elections officials, and now this CD has been making its way to Congress members.

It's shocking. All you do is double-click the icon. You go backwards through the Internet to that county computer, and if you have Microsoft Access on your machine you can walk right into that election database while it's open. It's configured for multiple access at the same time. You can be in there changing things and you can change anything you want.

Coupled with this are the recent posts from Mark Crispin Miller, who points us to an incredibly disturbing memo apparently authored by the fine folks at Diebold:
On January 17, 2001, Lana Hines, a county elections official sends out an inquiry as to how Al Gore ended up with a vote-count of -16,022. That's NEGATIVE 16,022—which just happens also to have been the total number of votes cast for various independent and third-party candidates who also ran. (It was the largest number of such votes cast in Volusia County's history.)

Is it really possible that a candidate could be recorded as having negative votes in any precinct? How is this possible? And did this in fact affect the vote totals in Florida? I'll be watching for answers.

I would be remiss if I didn't note that my friend Maia Cowan has been bugging me to get on this story, and have to admit I've only shied away because I've been too damned busy. Maia has been compiling data about these and related stories at her site, Failure is Impossible. As you can see, it's also a very complex story, and separating fact from speculation from fiction isn't going to be easy.

Also worth reading is the material that's being posted in Salon's Table Talk, in particular Maia's posts and those by Fergus, in the Touch Screen "Voting" thread.

But this is, I think, a potentially important story, with really deep implications for the integrity of democratic institutions. It also speaks, again, to the conservative movement's growing antipathy to democacy and the voting process.

Of course, I thought L. Jean Lewis' hiring should be an important story too, so take that assessment for all it's worth.

No comments: