Charles Stewart III of MIT and the Voting Technology Project in their milestone 100th Working Paper (emphasis added):
Nowadays, when I am asked “could Florida happen again?” I answer, “We won’t have any more problems of hanging chad, but I actually think the chance of a large-scale meltdown in many parts of the county are greater now than they were. I at least expect ‘another Florida’ in my lifetime.” The reason I answer this way is that innovation in the core technology of voting has failed to keep up with the challenges of the voting environment. At the same time, the “new” machines purchased with HAVA (Help America Vote Act) money have proven to have shorter life spans than initially estimated. Just as the pregnant chad problem was caused by the failure to keep up the maintenance of old technology that inevitably degrades, the “next Florida” is likely to come when a cash-strapped county somewhere in America lets its maintenance contract lapse, or fails to update its software in time.In other words, low-tech human failures.