California Said no to Die bold electronic voting machines. Could not prove th at they were secure so the state said no.
Ohio investigators are treating a warehouse where 15 electronic voting machines have been quarantined as a crime scene following a report someone may have illegally tampered with them to remove a candidate's name from the ballot. Officials from Ohio's Franklin County Board of Elections asked for a forensic analysis of the …
If it's what's-his-bucket (I forgot the name already), it probably wouldn't matter even if he won all of Ohio.
I think the government should force the sellers to be Open Source for the good of the nation. How do we really know that behind the scenes the voting machine isn't just adding one to a certain candidate? At least there's the every-so-often check by university people, yet they find a ton of holes; think how many the Open Source community could find.
I fear that these things are completely backwards. Instead of adding a paper trail to an electronic system why not the following?
You use a touch screen electronic machine to select your preferred candidate. The machine then prints a ballot paper (machine readable) for you to check. If you are happy that it has printed a giant X in the box of the candidate for whom you wish to vote you then place it in the reader. The machine can then optically verify that it has the same ballot, that the ballot reads the same as was entered electronically and the paper copy is then secured in the internal ballot box the same as a manual vote. This also avoids problems with "difficult to read" votes made by black voters in Florida when doing a recount as each piece of paper will have been correctly printed - if not it would be aborted and you get another chance to print it.
I guess the machines *were* made by Diebold (or whatever their election systems unit is called now) as they are headquartered in North Canton, Ohio. I would be surprised if their home state bought a competitor product.
I've had some contact with Diebold's ATM business - there are lots of good people there, good to work with. E-voting still scares me though.
Old school polls using paper ballots and ballot boxes, with people counting the ballots works fine. Scrutineers can validate the count and document questionable counts. There is a literal paper trail, and the system is easily understood and verified.
I've worked as a scrutineer, and a poll clerk and as a deputy returning officer: this is not a complicated way to hold a vote, and I doubt that it approaches 10% of the price of the electronic method.
Electronic records and computer systems are far too vulnerable to surreptitious alteration to be trusted with an election.
The USAers harp on their 'right to bear arms': when did they forget about the right to observe (scrutinize) their elections? Electronic voting obscures who casts ballots and obscures the accuracy of counting of ballots, it eliminates a physical record of the vote.
With paper ballots, the returning officer holds up the ballot and calls out 'one for party X': if the returning officer is wrong it is obvious and the citizens present can say so. I can see no way to unambiguously do a count or recount using electronic voting.
Placing the election mechanism beyond the oversight of the electorate is a bad idea in a democracy.
Good old paper and pencil ballots. Considering the cost of voting machines, it's not that expensive. They're used in Australia exclusively - and there is almost always a result "on the night", plus scrutineers from parties can, literally, look over the shoulders of official counters to verify the count as it occurs and accompany the sealed ballot boxes to the counting room - that is, everybody watches everybody else and every step in the process - the only thing that's secret is the voter placing their marks(s) on the ballot paper(s). Any re-counts, although they are quite rare, are also reliable. Usually, about 70% of the vote is tallied by around 4-5 hours after polls close (which gives a quite valid result, though, of course, every vote is counted). With a system as good as that, why even consider electronic machines?
---- Released Under GPL2----
To make paper voting even more secure, consider this:
a. Print invisible water marks on each ballot paper (on paper stock), and, as is currently done, keep the ballots under lock and key until election day, when they're distributed to polling booth officials to be handed to each voter at the polling booth;
b. Embed artificial DNA in the 'lead' of the pencils - a unique sequence for each electorate - and, similar to ballot papers, keep them under lock and key until election day;
c. If the election count is challenged in any electorate, or an official notices a discrepancy, then both the DNA sample left by the pencil and the watermark on the ballot paper(s) can be checked for validity.
(C) Anonymous Coward
----End GPL2 Release----
I often think that proponents of electronic voting are one of the following:
- Manufacturers if voting machines
- Naive tech-heads, yet to learn much about the 'real world' and its pitfalls
- People with a vested interest in tampering with elections.
It's getting chilly in here - I might need my coat.
Of course in Australia we also have a national body that
1. manages the elections,
2. makes sure that the process is simple,
3. runs national advertising campaigns explaining how to vote,
4. Is apolitical and incredibly professional.
If you have a mishmash of a system, you expect to get a mishmash of results.
(but they still managed to lose my voter registration update last time)
They are not installing voting machines *despite* the potential for fraud, they are installing voting machines *because of* the potential for fraud. You can rig a paper-based election, but there are so many annoying inconveniences involved, like what to do with that heavy box full of paper for the opposition candidate. Enough boxes dumped down sewers and eventually someone is going to notice your thugs doing it. And if you want to create ghost voters, then considerable medium abilities are necessary to get them to walk into a polling booth and grab a pen with their ectoplasm. And if you tell a man in the street that "The SFD protocol was probably compromised leading to data diarrhoea in the FDK backup" and show him a log you'll get a blank stare, but tell him that government troops have been ripping up ballot papers and show him photos and he'll be ready to riot.
Coincidentally the front page story on the Times is about widespread postal vote fraud. The excuse for that is to 'increase turnout'. What a stupid idea. When a vote results in a crap government it's not because we were too lazy to go and vote for the good politicians. There aren't any. It's their fault, not ours, no matter how much they'd like to pretend otherwise - and how much of our money they're willing to spend in the hope that we'll boost their egos.
good idea. Actually I had the very same idea some time ago already. More details:
- vote (page) would contain printed cryptographic signature, signed by the private key of the voting machine with included location and time (from builtin GPS - they are very cheap these days). This signature would be encoded in the form of two dimessional barcode. This barcode would not contain copy of actual vote, but its hash - this way vote will have to be scanned from the actual ballot and verified against signature.
- voter would take printed vote to the ballot box, and the box would scan vote as soon as it is inserted. This implies that paper would be rigid enough to prevent voter from folding it when inserting to the ballot box (imagine inserting your CC to ATM). If scanner at the box entry is unable to scan the vote or confirm its authenticity, the box would simply push the paper to separate box for manual verification and counting. The box would know public keys of all the machines in the local that are printing votes.
The big problem is that they have a huge number of elections, the same day.
They've been using mechanical aids for a long time.
In Florida:2000 they had a paper trail, and were prevented from checking it.
So, while these electronic machines don't leave a paper trail for checking, the big problem may be that nobody with any chance of winning an election wants the results to be checkable. They want the authority which comes from the magic of the ballot-box, but counting the actual votes? How quaint.
Ohio used to have an automated, computerized voting system that relied on punch cards. Votes were tabulated in a very timely fashion and there was a hard paper trail. Ohioans had an accurate vote tally most times before the 11:00 News and always in time for the morning paper....The mouse trap worked quite well.
The Ohio Republican party decided to moderinze all this with touch screens that leave no paper trail....Did Clintonista victories in the Buckeye state have anything to do with that? Oh well "modrenization" is always good right?
When I explained this to my relatives in Ohio 10 years ago and pointed out they were easily compromised, exploited, etc they looked at me like I told them about flying saucers and the loch ness monster.
was about two weeks ago. The ballot was a two-sided piece of paper with 3 columns. You read the issue or the list of names and scratch in the circle beside the name of the candidate that you want with the pencil chained to the booth. Then you feed it into a Diebold reader fitted to the top of a big metal box.
Maybe I'm wrong, but this seems like a great way to hold an election.
1) Simple, easy to read ballot
2) Said ballot is easy to fill out
3) Electronic scanning of the ballots gives you an instant tally of votes, just like they have been doing for years, except instead of punching out a punch card for a "computer" made in the 1950's, it is an optically scannable piece of input for a more modern counter.
4) It drops your ballot into a sealed bin and keeps it safe - i.e. a paper trail that can be read/counted by humans if need be.
5) If you put a ballot in with more than one mark per race or with a race/issue left blank it tells you and gives you the change to get your ballot back out and finish/fix it.
Seems like it should work to me. Online voting and touch-screen shit really is pretty scary though.
Where there is Foreseeable Harm risk of Secrecy which could possibly be used/abused to hide embarassment and/or professional blundering, rather than hiding jealously guarded industry-specific Knowledge or notions of national security.
You only have to Ask 42 Make IT So:
Dear voters, we apologise unreservedly for this transgression of the voting system's integrity.
We seek to assuage your legitimate concerns by reintroducing a paper-based voting system with immediate effect, until such time as a reliable electronic voting machine can be built by a non-profit company. Further, the design of the machine case, user interface and software will be made internationally available for peer review, comment and subsequent improvements will be fully implemented.
A few questions:
1) The tallies are not announced until after the polls close, so why is an instant tally of votes important?
2) Cannot the instant reading of the ballot be used to correlate how a given person voted?
3) Cryptographically signed papers and pens could be used to correlate votes to individuals too, no?
It seems to me that the old school manual method of holding an election is not broken, and does not need to be 'fixed'. Most of the 'fixes' proposed here, including yours' look a lot like application of computers merely because we can, not because there is a real benefit... and there are some possible serious ill effects.
Several places in the Free World already basically do it this way:
1) Voter uses machine to enter votes. Machine is in a
2) Voting machine spits out a human- and machine-readable paper with votes tallied on it. Machine displays summary of entered votes and maybe an image of what the printout should look like. Voter inspects and verifies his ballot.
3) If the voter decides that the ballot isn't correct, he puts it into another slot in the voting machine, where it is cross-cut shredded and the voter tries again.
4) If the voter approves the ballot, he takes it to the pollworkers' table. There he signs a form declaring that he has inspected and approved the ballot that is about to be put into the ballot box. He keeps one part of the form; the other is put into a box separate from the ballot box (where it can be counted later, to make sure that the number of affidavits and ballots match). There is an ID number on both parts of the form, but that does NOT match or get correlated with any information on the ballot proper....it's just an auditing check.
5) After the polls close, both sealed boxes (ballots and affidavits) get sent to the central counting location, where they are unsealed, counted, and the number of ballots matched against the number of affidavits.
1) If the number of ballots and affidavits for a precinct did NOT match, resolution would be difficult. Even if an electronic record were made by each voting machine as to how many votes had been cast, discrepancies would still risk endangering confidence in the election's integrity.
2) Manual recounts would be entirely practical, though difficult to match against voting-machine electronic records unless identifying/correlating information were added during the counting process - which would have its own set of problems.
I don't think anyone's yet come up with a plan for electronic or electronic-assisted voting thats sufficiently fraudproof and sufficiently idiotproof to trust a real election to it. Americans have learned not to trust "black boxes" in the voting booth, with good reason. While I wouldn't go quite so far as rd232, I am quite confident that highly motivated individuals and organizations are working to perfect his theory. At which point, comparing the US to a third-world banana republic would be even more egregiously unfair than it is already - to the third-world country in question. If we do not have transparent, straightforward, trustable elections, then nobody will trust them - not only the wingnuts on all sides, *everybody*. When that happens -- and if we keep going as we are, it IS 'when', not 'if' -- then the 'great experiment in democracy' will have been conclusively, objectively proven to have failed. Not that the theory is invalid - but in practice, the price of freedom IS eternal vigilance, and what was once the United States of America has been distracted, triangulated, hyperbolized and otherwise marginalized for a generation.