HAVA Problem?
HAVA -- the Help America Vote Act -- was a good idea, in theory. And I don't believe there is any evidence to suggest that the people who wrote it or voted for it had any bad intentions. Florida 2000 was a black-eye for America. NOBODY wanted to hear the words 'hanging chads' or 'butterfly ballot' anywhere but in history classes. It made perfect sense to encourage states to get rid of the sort of voting systems that made them possible. (And while, as a New Yorker, I disagree with the idea of replacing the clunky old mechanical systems that I have voted on all my life, which are messy and 'old-fashioned' but which are also safe and checkable, the company that made them is out of business, and when they break down it is hard to fix them or get replacement parts.)
And these days the first thought is always towards doing something electronically. Touch screens, computer counts, these were the obvious way to go, if done right.
Unfortunately, short-sightedness, stupidity, haste, and the idea to 'get it done fast, and we'll fix it in the implimentation,' as usual, created as much chaos as pure malevolence could have.
The important thing was not just that the systems created should BE efficient, honest, and secure, but that they would APPEAR so as well. After the Florida Follies, nobody wanted a system that was vulnerable, where the only guarantee of integrity was to trust election officials.
There would seem to be some simple basics here that should be musts in any electronic voting system.
There has to be a REAL paper trail -- an honest one that would include a copy for the voter and a copy of each individual vote for a possible recount. (To ensure voter confidentiality, each vote should be numbered, but the numbers should be assigned randomly and secretly, so only the voter would know the number he was assigned.)
The machine operating code should be open to a non-partisan group of computer experts before the machine was accepted, and the machine needs some form of device to that would set off bells, whistles, and alarms at any tampering with the code, rendering the machine unusable.
Some sort of barrier should be in place when individual machines link up to a central counting device -- is one is absolutely needed, so that if, despite precautions, some tampering is done with an individual machine, it can't affect the whole system.
Any operation of the machine has to be done on a hands on basis, with no wireless ports, no internal wireless modem, etc.
States and cities should be provided with money not just to purchase such machines, but to store them securely, so they would be locked away until the night before the election, brought out, a final test would be run, and then they would be distributed to the polling places. If this cosy extra money, again it should be provided.
Finally, despite the expense, I think that each precint should have four voting machines that could be rotated, for example, primary, general, primary, general, with perhaps a fifth for special elections. That way, the machines could be inspected at any time up to two years after a given election if there are allegations of fraud.
(I would have to ask someone more familiar with technology and costs to answer this, but since the danger seems to me to be the greatest when a machine is reprogrammed for a new election, it might actually be possible to create a 'one-time use' machine that could be used for a given election, be unreprogrammable, be unopenable without rendering it useless, and which could then be stored indefinitely for the use of historians and social scientists.)
None of these seem to be brilliant ideas, merely obvious ones, They must be the sort of ideas that are implimented in current electronic voting machines, the ones that are being used in more and more states. Right?
Unfortunately...
