it's not a conspiracy, it's just poor design

[waider]

problems with touchscreen voting machines. Yep, I recognise this. I have four touchscreens around my desk at the moment; one resistive screen and three capacitative. The resistive is the most accurate of them, and it's still rough, even when you go to the trouble of a full 25-point calibration. The capacitative touchscreens are a joke; one of them won't work properly if the machine is fully assembled for reasons that aren't clear to me, and the other two have pretty bad drift toward the screen edges. Also, all three of the capacitative screens will point to the wrong place if you happen to be touching (i.e. leaning on) the metal casing at the same time as you're touching the screen.

Now, why the pieces of crap being used for the elections were approved for use may be open to your choice of conspiracy theory, but since it seems that it's aribtrary about which way your vote goes it seems like it should just throw errors more-or-less equally on both sides of the line.

Comments

[wisn.livejournal.com]

I'm waiting for proof that dynamic machine voting solves any problems it was meant to solve. As far as I can tell it can only complexify existing processes. (The old-fashioned lever balloting system was notorious for non-subtle riggability undetectable by the voter)

This is to be distinguished from electronic vote counting. My county has been using a scanning tally system for at least as long as I've lived here; the voter has to carefully mark a field in a sheet. The marking method could be better designed, but it's not bad; ones votes are very clearly expressed and any user errors require nothing more than asking for a new tally sheet. The counting is electronic, but the scanning machine neatly stacks the voted papers, so retallying can be done electronically or manually.

The reporting of tallies electronically is a different issue, but would also be a different problem.

[wisn.livejournal.com]

Taking quick issue with your subject: Bad design can be evidence of conspiracy. Which is not to say conspiracy is at play here, although other aspects of design sloppiness (eg, Diebold's voting machine's database and account management) indicate a calculated indifference to tamperability.

Or, in other words, who cares whether the voter selects A or B when the votecounter will log A anyway?