Dan Wallach gave a great, if disturbing, talk yesterday in Rice’s Scientia series. He is in the Computer Science department at Rice University and specializes on the mechanics of voting. A good hacker can apparently foil just about any system, by changing the vote, letting you think you have voted for a certain person, but sending in the wrong vote. The hack can be machine based, software based, or hardware based. They can overcome a lot of post voting verification attempts. For example, the machines could have ten votes recorded early for a specific person that didn’t actually get the votes. We the people apparently like best a system we can understand and verify if something goes wrong, and that is paper, with pencil marks, an optical scan system.
It’s true that optical scan paper ballots can be verified, and there is a solid record, if the ballots don’t get lost, or disappear. But there could be computer-based miscounting, with all the other kinds of fraud. The paper trail advantage kicks in only if there is already a suspicion of a problem.
The system Dan favors is way cooler than that, but harder to understand. It involves electonic voting with all votes sent immediately to a central place. The cool part is that the system itself has to tell you how you voted, and sometimes these can be challenges that don’t result in an actual vote, but demonstrate instead that the vote cast was the vote recorded. No doubt you can find a clearer description on his web page.
Internet-based voting sounds good, but may be impossible to secure. The best story is a Washington D. C. online voting trial that invited hackers to test the system. Paul Stenbjorn from the University of Michigan, my alma mater, set his students to test the system, and one broke in successfully, even causing it to play the Michigan fight song (had to ask my husband to remind me of the melody).
Sadly, it seems like the political will for accurate voting isn’t out there. People are happy with what they have, and don’t very often worry about it.
I love Scientia because it brings together the curious and critical from across campus, and the community. I should say I love the monthly talks, but not the odd structure with members and fellows, and voting, and a Bochner lecture that took many decades before they found a woman they deemed interesting enough to invite. But it’s mostly about the talks, and they are great.
It also has a crowd of undergraduates in there for credit, writing their papers on it as they listen, then crowding out for the cheese, fruit, and crackers, even before the questions. Good thing they don’t know about the wine in the back room. They come, they probably learn something, and only one of the ones I sat behind was web surfing.
How do we get our students to pay attention these days?
Rick Wilson, David Queller, and David Leebron enjoyed the talk.