Thursday, June 22, 2006

A rant on voting

There are a number of common miss-perceptions about voting that I have seen of late in the media. Pardon my rant. Here are some of my recent thoughts on how the e-voting problem is being misrepresented.

Three false precepts:

1. Electronic voting needs a "paper trail" or "receipt"
2. Mail in paper ballots are a good idea.
3. A conspiracy "theory" is required to corrupt an election.

Starting with the last point, a conspiracy doesn't require that it be a theory. We've seen conspiracies throughout our history, most times in situations involving lower stakes then a presidential election. That having been said, the actions of local poll workers can alter the counts of untraceable voting totals, for example those produced by closed source systems. This doesn't require a conspiracy, just the motivation of a group of loyal supporters. On top of this you simply do not know what is going on in the box. The ease of hacking some of these systems is simpler then putting a virus on your PC. You can get anti-virus software for your PC. I've never seen anti-voting-fraud software. What makes this worse is that you can't legally update a voting system after it's certified, although this has happened illegally more then once. In other words, fixing a voting security hole is often determined to be "too late for the election".

On my second point, when you think about it, mail in paper ballots are *not* a good idea. Let's say I'm your employer, abusive spouse or you're a child of a strong supporter of candidate XYZ. Perhaps someone offers you money to vote for XYZ. In a traditional polling place that person could not tell if you voted for who they said to. With mail in ballots that is quite easy. You show them the ballot, sign it and hand it over so they can mail it. In the best scenario you get $50 and go home. In the worst you might not get beat again. See the problem?

If the polling place survives, voting may well be done via touch screen computers. When no one, or even a limited group can analyze how the proprietary software and hardware in these machines works, it seems obvious that a few folks could manipulate the results. This applies to the machines that count the votes, tabulators, as much or more then the individual voting stations. Note that the tabulators are also used for mail-in votes. This is the strongest argument for open source voting machines. One person can manipulate a central vote counting system. One programmer can change all of the voting stations. Without a way to see what the voter really intended, it can be done without a trace. Knowing the voters intent, beyond manipulation, can not be achieved without many eyes on the software source codes.

A paper trail doesn't help, it's a ballot of record that will determine the election. If a computer is used as an aid to fill in a paper ballot so be it, just put that paper in the hands of the voter so that they are more likely to verify it. It's the voters intent that counts here. A "paper trail" behind a glass door isn't something the voter is likely to even notice. A receipt is something that again allows coercion or buying of votes.

Although I'm glad the some in the media are starting to understand that elections have and will be manipulated, in large part due to HAVA, again the solutions may well worsen the problem.

"The people who cast the votes decide nothing. The people who count the votes decide everything."


Joseph Stalin

I hope this fairly well spells out the problem. Please add on anything I've missed here.

0 Comments:

Post a Comment

<< Home