Keeping Levers - as a Strategy

[WillYourVoteBCounted]

Strategy wise, keeping the levers is a smart move.

KEEPING THE LEVERS BUYS YOU TIME:

Time until vendors are made responsible for their equipment and its problems,
Time until the federal certification program is more meaningful,
Time until decent voting systems are certified,
Time until open source is available,
Time until laws require ballot definition files to be public,
Time until robust audit laws are in place,
Time until federal legislation passes (don't count on it).

KEEPING THE LEVERS LIMITS RISK TO VOTES

One lever machine counts up to 999 votes. That is the max that can be corrupted or lost.
With levers, there are no mass programming errors or ballot definition errors that can impact tens of thousands of votes in one fell swoop.

Switching from Levers to the new generation of voting machines - the DS200 or the Insight -
makes states become gigantic beta tests. We saw the problems with the DS200 in Florida.

WE NEED MORE ACCOUNTABILITY FOR VOTING SYSTEMS AND THE VENDORS THAT SELL THEM.
BETTER YET, GET THE VENDORS OUT OF THE VOTING SOFTWARE BUSINESS FIRST.

ENACT THE NEEDED STANDARDS AND REQUIREMENTS FOR VOTING VENDORS AND SOFTWARE SO THAT THE LEAP IS NOT FROM THE FRYING PAN INTO THE BON FIRE.

[Bill Bored]

1. There aren't going to be any "robust audit laws."

Even reform-minded election officials can't or won't get it done. They talk a good game, but in the end, they just trust the computers.

And to do audits right is much too complex, unless some serious effort is put into simplifying rigorous methods. This just isn't a priority for the current crop of would-be technocrats. And besides, too much hand counting is required to check all elections anyway.

When all is said and done, what we will end up with is, for the most part, pretty much what we started out with: faith-based elections.

Levers are different. For the reasons you mention, there is every reason to believe that the votes are being counted as cast. Large errors can be prevented even before they occur, upon inspection of the machines, or after the election in the rare event of a machine failure. If an error is large enough to affect the outcome of a race (also rare), a new election could even be held. Many have suggested that this would be preferable to a recount in which the paper ballots are presumed to have been tampered with, or just seem to appear and disappear. That's still better than trusting computers, but it's nowhere near as secure as a lever election.

The most compelling argument for keeping levers is simply that they are not physically capable of switching votes during elections. But that's not good enough for some of the techies who'd rather spend their time, and other people's money, figuring out how to keep computers from switching votes during elections. (Of course, that's a good idea if computers are running your elections, but otherwise, it's kind of silly.)

There's room for all this stuff in the E.I. community, but not at the expense of replacing non-computerized systems, which are in fact, the gold standard against which all e-vote counting systems should be judged. So far, they have all failed to measure up.