Slashdot Mirror
States Throw Out Electronic Voting Machines
Davide Marney passes along an AP story about the thousands of voting machines gathering dust in warehouses across the country after states such as California, Ohio, and Florida have banned their use. Many of these machines cost $3.5K to $5K each. Local election boards are struggling to find ways to recover any of the cost of the machines, or even to recycle them. The picture in Ohio is the most confusing, as multiple court cases limit the state's options and result in a situation in which the discredited machines will nevertheless be used in the presidential election coming up in November. The state's new (Democratic) attorney general has just issued a rule banning the practice of election workers taking the machines home with them the night before elections.
Of course not. With Linux it would not need antivirus software.
It is dangerous to be right when the government is wrong.
Actually, it's not Ohio Governor Ted Strickland you need to really thank for this, it's Ohio Secretary of State Jennifer Brunner who came in with Strickland (who had previously specialized in election law).
By comparison, her predecessor Ken Blackwell was one of those involved in guaranteeing the electoral votes of Ohio would go to Bush. Which of course had nothing at all to do with the fact that white suburban precincts had plenty of voting machines and about a 10 minute wait while poor black urban precincts had 5 hour waits and college campuses closer to 6 hour waits.
I am officially gone from
Wide use of these machines was adopted in the 2000 election: Winner=George Bush
Even more are used in the 2004 election: Winner=George Bush
Now they throw them out just in time for the 2008 election because George Bush might win again if they didn't.
No, wide use of these machines was implemented after the 2000 election.
The truth is that all men having power ought to be mistrusted. James Madison
And now the Democrat Secretary of State is redefining election law to say that registration and then absentee voting up to 5 days before the election is legal because by the time the paper ballots are recorded the voter would have been registered for the legislated 30 days.
Why stop there, why not let 17 yr olds that would turn 18 by the time the ballots are counted vote too?
What if in 30 days you plan to be a resident of the state?
What is this "waiting" thing? We vote using paper in the UK and IIRC I have never had to wait even a single minute. I just walk to the desk with the people on it and speak to one of them to get the blank paper. The five minute walk to the polling station takes longer than the entire voting process (getting verified, voting in a booth, posting the vote in the ballot box).
Or... It's all about the (implied?) parenthesis. (t*9.81) m/s, where the parent is showing the units of the result not the units of the constant.
"Uh... yeah, Brain, but where are we going to find rubber pants our size?" --Pinky
First, the paper-ballot-counters aren't connected to networks, generally.
Second, a system where a paper ballot is counted by an electronic machine has a critical feature that an all-electronic system lacks: it's auditable, with an independent backup of the vote (the paper ballot).
touchscreens ARE cheap. Hell I buy LCD's with touchscreens on them for $25.00 on ebay all the time.
It's not like these voting machines have laptop ready, 1024X768 16 million color TFT displays with high end touchscreens in them.
They have the cheezy LCD's and Cheezy resistance based touchscreens. you can get them everywhere for dirt.
My hope is some scrapper buys them all parts them out to pieces and sells the parts on places like sparkfun and allelectronics for almost nothing.
Do not look at laser with remaining good eye.
Which is exactly why I favor the fill-in-the-oval type of ballots. They are easy to understand (no butterfly-ballot, "I pushed the wrong button" sort of problems), they are easy to read (Is the oval black or is it white?), they are machine-readable (so machine counting is quick and easy), and, most importantly, they leave behind a directly human-filled-in paper trail (so physical, manual, by-hand, honest-to-goodness recounts and cross-checks are actually possible, when needed).
Touch-screen voting machines, on the other hand, are problematic from the start. There is not and cannot be any true record of what was showing on the screen and where the user touched the screen or what the user intended by touching the screen in that spot. That's the fundamental flaw and no amount of regulations, printouts, or open-source software can fix that. These machines are dangerous precisely because they are so rife with such massive potential for invisible, unverifiable voter fraud and, I truly fear, that is the only reason they were ever even considered in the first place.
Yes, the hand-filled paper ballots are counted by machine, but they are also a directly-verifiable indication of the voter's intent and can be counted by hand when the need arises. Touch-screen voting, by design, will never have that.
Bah! A trebuchet's too good for them!
If there is a 5 hour queue at the time when the voting shall end - will these be disqualified from voting?
No. If you're in line when the polls officially close, you must still be allowed to cast your vote.
OMG give it a rest. He won and that is history. Now we have to be careful the high and mighty democrats don't go back to doing what they did in Chicago. Inventing votes from people who have been dead for years!!! Yes that is right. People who were dead their votes were cast in Chicago. The democrat haven of Illinois. So before you say votes were invented or slanted look at both sides.
Or worse yet. The hanging chad incidents trying to guess the voters intent. God that one will be in the history books for a long time. Should be in a book called "how to steal an election".
A trebuchet is a form of catapult.
Nobody expects the British Columbia Human Rights Tribunal.
This is, in fact, the current solution used in California, which has moved to optical-scan ballots.
Note that you can save a lot of time and money if everyone who doesn't need assistance just fills out the optical-scan ballot by hand. People who need the assistance can use a machine to produce a properly-filled-out optical-scan paper ballot.
The counting is then done by an entirely different machine. A random set of districts are required to perform a hand count of the ballots to verify the results. The important feature is that the paper ballots provide a later fallback measure if the optical-scan system fails.
Incidentally, the mechanical lever systems bear the same major problem as electronic voting systems: they can be undetectably modified.
Except that you can detect their modification by testing them.
Electronic voting machines could have their modification triggered in dozens of undetectable ways. They do run the clock forward, and do a bunch of votes on it, so it's not going to switch on a specific time, but it could easily be triggered by specific ballot, or pushing the screen in a weird spot, or anything. And then, at the end, copy the correct software into place and wipe itself.
I'm not a huge fan of mechanical voting machines, but they at least either work all the time, or fail to work all the time. If you push a button 1000 times and it counts to 1000 in a test, you can be sure, mechanically, that's correct.
Yes, in theory, there could be some sort of clever linkage built into the machine as a backdoor, same as in an electronic voting machine, but they also do a quick visual inspection of the machinery before the voting.
Whereas there's no way to detect that in a computer:
a) The 'software certification' demonstrably not work. There is too much access to the machine, there are repeated instances of last-minute updates that demonstrate that no one takes it seriously, and poll workers have no idea how to check that.
b) Even if we could prove the software actually running was the right software, that does not exclude very clever backdoors that are actually in the certified code. Although at least we could track them down later...but they could easily be hidden as bugs, and unless there was some evidence they were actually triggered, what do we do?
c) Even if we know the software is perfect, and is actually what is on the machine...these things run Windows and the rest of the machine doesn't have to be certified. All you have to do is combine a 'rootkit' and a game 'trainer' and rewrite memory addresses. (Of course, is ignoring the fact that half such machines run Access, and that's easy enough to modify anyway.)
If corporations are people, aren't stockholders guilty of slavery?
This isn't some theoretical attack, it's a known weakness in the mechanical voting systems. You're just not familiar with the voting machine, so my description doesn't make sense.
The mechanical voting machine has a single large lever, a board full of toggles, and a curtain. On entering, you pull the lever, which closes the curtain. We'll call that lever state "voting". You then flip the toggles for the items you want to vote for. For example, you might flip "John" under the "President" column and "Yes" under the "Proposition 127" column. You can change your mind, but the system (ostensibly) prevents you from marking two toggles in the same column simultaneously. You then flip the lever to the other state, "not voting". This transition causes all set toggles to be unset and, in the process, increments an internal counter for each set toggle. This is all done with gears. It also reopens the curtain, so you cannot vote multiple times.
A toggle can be easily jammed with pencil lead. The toggle will appear to set and unset normally, but until the piece of lead is worked out of the system, the "gears turn, incrementing an internal counter" part fails.
They weren't. A hart was a male deer.
You're making me think. You won't like me when I'm thinking.