A vote of confidence for open source elections

Salt shakers
February 7, 2008

A vote of confidence for open source elections
Do you remember the hanging chad fiasco of 2000? How about lesser known examples of election voting problems? Alan Dechert keeps a close count of them all. Dechert is out to reform the way America votes and boost civic participation and democracy in the process. His organization, the Open Voting Consortium , calls for a national “open voting system” employing the latest open source technology and the simple notion of complete electoral procedural transparency in U.S. elections.

How did you decide you wanted to reform the voting system?
I got started in all this after the presidential race in 2000. The system was antiquated and broken. We put a man on the moon and we can’t count votes. It’s absurd.

What exactly is an “open voting system”?
Basically, we’re trying to get at the whole process. Every aspect of it needs to be fully open to public scrutiny.

First, the software used on voting computers needs to be open source. Second, after a person casts their vote, there needs to be a paper ballot printed out that confirms their choices that they can verify. Lastly, the voter would physically cast their ballot into a ballot box, and all ballots would be hand counted.

And how does that differ from traditional voting systems?
The first difference is the cost. Traditional voting systems spend a lot of money on hardware that we don’t really need to. We recently held a test trial of our system in San Luis Obispo, California. The system we employ uses old computers and printers. That hardware cost was zero. The software cost was zero. And we also don’t use pre-printed ballots, which cuts down on the cost considerably as well.

Another difference is we don’t have issues with understanding voter intent. We print out in plain text for the voter to read and verify their selections.

Explain how “open source software” differs from the software used in traditional voting systems?
Those systems are proprietary, they’re incredibly difficult to run testing on and the public has no reason to trust that they function the way companies say they will. It’s security by obscurity.

With open source, the code is available for anyone to see, test and verify.

But you’re a programmer. Most people are not. Will this system be too complicated?
I hear that a lot. We’ve written the software so that it is verifiable by non-programmers. Basically we’ve written it so that it renders the program like a picture file. In other words, it creates images of the screens that you, as a voter, will see you when go to the polls to cast your vote. Anyone, therefore, can go online in advance, load the program, and verify that those screens are accurate.

hat’s your goal in all of this? What are the benefits for the public?
There are two things. One is trust. We know that a lot of people don’t have that much confidence that their votes are being counted correctly. Two, the public tabulation of votes increases public participation in the election process.

I think it will increase civic participation because it gets people involved in a real, tangible way. At our recent trial in California, we interviewed some people who sounded thrilled by the whole procedure. They found the whole process very enjoyable.

When will we see the first presidential election run on an open voting system? How far off are we?
We're $2 million dollars away. Real testing of computer software is really expensive. The lab fees you have to pay to get your software certified, both federally and at the state level, will cost a million dollars or more. On top of that, we probably need another million for research and development.

But I’m optimistic. I’d say, we'll be there by 2012.

Back to page top