I’m 2,500 miles from home today, so I voted early. It took me about 10 minutes to get through the process. I suspect that if I were waiting in line today, it would take quite a bit longer.

From an economic cost/benefit analysis, voting is irrational. And given the frequent choice of the lesser of two evils, it’s not particularly satisfying either.

Can my vote really make a difference? Sure, there is the (extremely small) theoretical possibility that it does. In my district there’s only one race that is competitive – the ugly battle between George Allen and Jim Webb for U.S. Senate. If my vote were to make the difference in that race, even that would only really make a difference if it changes the balance of power in the Senate.

Voting is relatively painless for me. My job is flexible enough that if I’m delayed at the polls it’s not an issue. A Wal-mart employee could lose their job and would be docked wages for showing up late. I also don’t have kids to ferry to school or tote to the polls. For many people, the costs of voting are much higher, making the cost/benefit much worse.

Clearly, many registered voters (more than 60% in some states) don’t feel that voting is worth the costs involved. Given that we can’t do much to increase the benefit from voting, we should be looking at reducing the cost.

I can do most things in Arlington County on the Web: renew library books, pay traffic fines, register my car, pay my property taxes, etc. All of these have a direct, quantifiable impact on me. I can do them with a few clicks from the comfort of my living room or at lunch from my desk. It should be just as easy to vote.

Voting is also costly for local governments. Billions of dollars of equipment are purchased with our tax dollars and sit around for much of the year. They are used (typically) twice a year. Governments also must scramble to find people who can staff the polling booths and allocate tax dollars to pay them.

I would love to see open-source e-voting systems developed. Being able to vote from a Web browser would reduce the costs both to individuals and to governments.

One sticking point: the secret ballot. It is technically feasible to separate the issuance of the ballot from the recording of the votes on that ballot. This is essentially what is done with absentee ballots that are sent by mail. It does require that voters trust the system.

Of course, I have no real proof that the vote I cast on the WinVote machine will be counted either. I would rather have a system where the software is available for public review versus one in which companies have economic incentives to hide the flaws in their software.

Control of Congress shouldn’t depend on whether it snows in the Midwest or the 75-year-old poll worker can figure out the newfangled gadget from Diebold, which may or may not have buggy software.