I am amazed at the problems they are having with the new electronic machines. Plenty of people handle banking online and through IVR systems with a very low percentage of real problems. It may just be the press the voting machines get, but they seem extremely unreliable - and they are conceptually very simple machines.
What amazes me even more is how unreliable all of the US voting systems have been. Here is Connecticut, they moved us to paper ballots similar to the "fill in the bubble with a #2 pencil" tests we all used to take. Some tests were done of the scanning machines that read these ballots and sending the same group of ballots through 10 times yielded 10 different results. Most of the differences were in ballots being readable on one pass, but not another pass.
The old systems of paper ballots and manual counts were obviously going to be a 2-3% issue in just human error (minimum). I cannot believe any vote counting system we have ever had is much better.
So, although the new machines are unreliable, are they more or less unreliable than the old systems? Is it just that we know how unreliable the new ones are and we did not know the problems with the old ones?
Regardless, I think it is only a matter of time before there is a "Google Vote" application built and we can all just adopt that. It will be able to trend the voting statistics right next to the flu spread trend (which they were able to predict faster than the CDC). And eventually they will make an iPhone application for voting directly from your phone - but the app will probably get stuck in the app store waiting for approval and never get released.