Slashdot Mirror
Avi Rubin and More on Electronic Voting
jgo writes "Johns Hopkins Computer Science professor Avi Rubin, posted his experience as an election judge on his website. It's an interesting read and exposes some potential security problems with electronic voting. At one point he held in his hand the five memory cards containing all of his precinct's votes." Rubin had posted his experience in the primary election earlier.
Glitch gave Bush extra votes in Ohio.
Franklin County's unofficial results had Bush receiving 4,258 votes to Democrat John Kerry's 260 votes in a precinct in Gahanna.
Records show only 638 voters cast ballots in that precinct.
How's this for a way of safely conducting electronic voting...
Give everyone a GUID, a complete random key of sufficient length that you can't simply guess and get a valid GUID. Mail it to them.
When a person votes, their vote is stored against their GUID, in a publically accessable database. Anyone can check that their vote has been correctly counted by looking up their GUID in the table.
Voting would effectively be pseudonymous instead of anonymous. (With a new pseudonym for every election).
Please watch this free 30-minute film about black box voting machines.
We have all been scared about Diebold and other black box voting machines, and for good reason. Apparently one of the central machines from Election Systems & Software Inc. tallied 115 votes for Bush in a certain county, while another machine tallied 365 votes for that same county. Which one was right? There is no way to tell, because "it is too hard" to add a printer to a counting machine. It is not like they have been doing that for 30 years. But who needs to do a recount when the machines are infallible, right?
Most infuriating of all is that Republican Senator Hagel, the former Senate Ethics Director, resigned after admitting that he owned Election Systems & Software! That's right, the same voting machine maker that 60% of ALL VOTES in the U.S. are counted on, the same one that provably miscounted votes in Ohio and other states, and the same one that refuses to print receipts to recount these votes. No wonder legislation trying to require printers on voting machines is taking so long to get through congress when congressmen can vote themselves into office without a paper trail.
We walk into a grocery store and usually buy stuff instead of stuffing it in our pockets and running. We know it's wrong to leave without paying.
Why do votes need uber security check technology? Whatever happened to scrutiny by peers?
IMO, paper ballots are best because it is just tougher to destroy them. But, we should get receipts showing how we voted for our own records.
But, trying to turn the entire election process into zero possibility of error or fraud undermines the election itself and goes against the ideals of our society. People in general are honest - and those that aren't get caught eventually by honest people.
Suggesting that 'one person' should not be able to hold an entire precincts' votes just doesn't make much sense. People are often responsible for others. I suppose twenty people should all carry a piece of the nuclear football too..
--- We need more Ron Paul!
There is a question, of course, about how long you might be locked up for doing so.
It seems that no one really wants to come forward and raise this as a serious concern for this election, despite the fact that it's entirely plausible. Yeah, that worries me a bit too. A wrinkle is the fact that all the early exit polls pointed to a Kerry victory, the republicans were depressed, the democrats estatic. Then when the real results started coming in the situation was reversed. Especially when you have the president of Diebold a very strong Repubican. There is probabally nothing in it, and for whatever reason the exit polls were wrong. Normally they are pretty accurate though. It is probably for the best to try to forget about it, and make sure that these stuff is fixed for the next election. The other huge problem is the amount of gerrymandering that goes on. You really need to get the partisan officials OUT of the redistricting. The house of representitives elections are becoming insane, with a lot of stupidly safe seats. only something like 10% of house seats are competetive, and that is really really bad for democracy. If the only way you can lose your seat is if you get deselected by the party faithful, then it polaizes the politics, and noone moves to the centre, and it becomes a real mess.
Time to put the tinfoil hat back on, you paranoid pinko!
Seriously, someone has cast doubt. Blackboxvoting.org blanket the country with freedom of information requests on election night. They currently need $50,000 to complete the audit. I gave $100. Let's see what we can do together as slashdot.
Computers are useless. They can only give you answers.
-- Pablo Picasso
H.R.2239 and S.1980, discussed further here, will amend the Help America Vote Act (an act designed to ensure consistent voting systems that meet certain standards be available to ALL voters in ALL jurisdictions), such that there is "a voter-verified permanent record or hardcopy" attached with each and every ballot cast by every voter.
.[1] Don't believe people who make it seem like companies like Diebold are resisting. They aren't. They'll build - and sell - whatever municipalities will buy.
Please, simply support this legislation.
Additionally, the electronic voting manufacturers, such as Diebold, already have the ability to add permanent, individual voter-verified paper audit trails to their products
The roadblock, as it turns out, is often local election boards. First, the new paper verification systems NEED to go through the government certification process - remember, it's the e-voting watchdogs who are chastising non-certified patches/updates being put into place; the paper audit systems need to go through the same certification process. Further, many municipalities can't understand why they should be forcing paper audit trails; after all, they think, they are just getting away from paper ballots - why should they be arguing for paper ballots (and all the headaches that go along with them, ON TOP of the headaches they already have from learning to deal with e-voting), so why should they go back to them?
Folks, so many people are involved in elections at so many different levels that there is literally no way that any central entity could rig an election across an entire state. Experts dealing with e-voting don't even have this on their radar. Their concern is more errors and failures. E.g., most of Ohio is still punchcard as it is (the majority of the 35 counties moving to e-voting pushed off the transition until AFTER the election because of problems), and someone like Diebold doesn't even have access to this equipment after the fact. Yes, an unscrupulous election official or enterprising hacker might be able to breach individual machines and potentially even a county - it's possible. But the likelihood of something like that happening on any significant scale, ESPECIALLY without being caught (the articles we're talking about here actually prove that the audit processes, be they what they are, do work) is very, very low.
That said, we absolutely should be ensuring that there is a permanent, voter-verified, paper record. It is absolutely critical to our voting process, even if the software is still proprietary on these systems (though it, too, should be open for public inspection). But the permanent voter-verified paper record alone eliminates the chances for any widespread fraud with the counting process itself, and at the very least makes any fraud easily reversible and/or detectable.
Contact your representative and senators, and urge them to support the above bills. It will be a lot more productive that imagining fantasies about Diebold "handing" Bush the election. (If ANYTHING remotely like that happened, there are a shitload of professors, campaign staff, scholars, journalists, and researchers who know a LOT more than you do who would be all over this in a heartbeat. Kerry's $300 million, two-year campaign didn't just roll over for no reason. Bush won, whether anyone likes it or not, and it wasn't because electronic voting handed anyone anything. The POINT here, is that instead of inventing wild conspiracy theories, we should be ensuring that there is voter verification and a permanent paper record for all future elections, because HAVA will require a shift to electronic voting for everyone - before that happens, we should make sure that it's veri
The problem it solves is that a paper ballot is great when you're only asking a few questions, but not when it's used to ask all the questions your government wants to ask you all at the same time.
I saw earlier(trying to remember where it was exactly) a pretty reasoned explanation as to why the Canadian paper ballot solution couldn't be applied to the states. The consensus was that until the vote tally for the presidential election(and perhaps house/senate) became a federal responsability, it was unpractical, after that maybe.
I'm wondering why so few people are interested from "Decoupling" the presidential elections from the local school board, any opinions?
BBV is soliciting donations icw the largest FOIA request ever submitted
stolenelection2004.com
votergate.tv
Outrage in Ohio Was the Ohio Election Honest and Fair?
Kerry Won
Shoplifting the Presidency?
Ultimate Felony Against Democracy
Surprising Pattern of Florida's Election Results
votes for party president versus voters registered
exit_poll(gif)
Florida2004chart
openvotingconsortium.org
verifiedvoting.org/eirs
electionprotection2004.org
The Rise of Open-Source Politics
http://www.cpsr.net
Presume once congress & the administration are aware to the purported problems they respond rapidly with "Help America Vote Act - II".
I voted on an electronic machine here in Atlanta, GA. Previously, I have voted using mechanical machines in NY and Pennsylvania. One big difference: less privacy with the electronic machines. It's not a particularly big deal to me, but some might feel weird about that. Especially if they intend to vote for a candidate that is very unpopular in their district.
I felt the process and UI was fine (clear, minimal opportunity for human error, etc.).
Main complaint (other than security concerns): the potential of the electronic machines was not realized. For example, there were several initiatives on the ballot here. One was a widely publicized gay bashing, er, I mean, marriage protection ammendment. Another was a lesser publicized amendment relating to judicial jurisdiction. (Both described here) I knew a great deal about the gay bashing measure, but hadn't heard of the proposed amendment about the courts. All they put on the ballot was a yes or no to the following statement: "Shall the Constitution be amended so as to provide that the Supreme Court shall have jurisdiction and authority to answer questions of law from any state appellate or federal district or appellate court?" Um, how about maybe?
It would be great if a more clear explanation could be added to the ballot. The electronic medium makes this crazy easy. It's no more expensive to do. The website linked above even has a very clear description that could have been used. (Of course, this opens up questions about potential bias that can be worked in to the description. However, I think something is almost certainly better than nothing.)
I think electronic voting will be a good thing if the security concerns are worked out. Will they be? That's hard to say. In the near future will most Americans think they are? Yes, almost certainly.
We always get our results in a timely manner and, to my recollection, there have never been any problems with the vote counting.
More than doubts, check out this PDF
Graphs.pdf Clicking this link will start a download.
photosMy Photostream
In Arkansas, your ballot is numbered and the number of your ballot recorded next to your name in the voter registration book. They can look at will to see how you voted. Entirely unamerican if you ask me.
Why are we introducing the chances for errors into our most important civic institution? This is insanity! As another poster wrote there is no reason that a printout will accurately reflect how the machine handles your input, it's only showing you what was sent to the printer. We have so many other obfuscating problems as well, like magnets and code tampering and using phone lines to transmit results.
.5% in the polls. It'll be 10 people and 10 ballot boxes per precinct - tops. Wood is not expensive so don't go there.
/. so you've got more insight than most folks.
The real problem is taking the physical stylus out of the hand of the voter. I would only consider eVoting for disabled persons, and I would think the majority of them have few problems.
1) To avoid fraud, why not submit the ballot into more than one ballot box. One for each candidate on the ticket. If democrats and republicans have their own ballot box - they'll likely have the same number of votes - the incentive to cheat is removed without duopoly.
2) Allow all candidates nationwide to be on the ballot if they garner
Here's a nice page to Federal Contact Information http://www.eff.org/congress/ - tell them what you think - you're on
Stuff that matters.
What's wrong with a pencil, a piece of paper, and a count process to which the candidates (and their lawyers) can be invited to ?
Because that would be, like, so untechnological. If ever there was a thing that *should be* untechnological, it'd be voting.
The US, in its wisdom and reliance on expensive stuff, thinks that plain old paper is not good enough.
You deserve what you allow the computers to get away with.
Avi's not the only election judge recording his experiences. So are his minions: http://cs.jhu.edu/~mgreen/election_judge.html
Let's assume the worst-case scenario (from an effort point of view, not from an accuracy point ov view) and say that the votes are challenged every time and the paper ballots end up having the final say. How has the electronic counting helped?
Given that computers are less prone to make careless errors (OK, they don't make careless errors), even if they might be more prone to systematic errors, they give you a number to compare against. Let's say that the computer told you it had printed out 2,523 votes for Bush and 2,427 votes for Kerry. When the vote-counters counted it, however, they counted 2,525 votes for Bush and 2,425 votes for Kerry. The first thing that one should assume is that the vote-counters miscounted, and should recount. If a second recount (by different people) got the same result as the first human count, then we have a problem. The error could be: (1) the computer mis-counted, or (2) the computer mis-printed. Unfortunately, either one is possible. However, since the voters would be encouraged to look at their ballots prior to them entering the box, it would seem more likely that the computer mis-counted, in which case the human count should trump the computer count.
However, notice that the computer count still helps. It gives us a number to compare against. If the human count on the first count matched the computer count, there is little reason to suspect that both counts are wrong. (Although, theoretically, the computer could still have mis-printed and mis-counted in a matching way. This would be an unlikely accidental error, and a very risky deliberate hack since the voters can verify their votes before they go in the box.
Of course, this only works if the printed version can be viewed by the voter prior to it going in the box.
Ben Hocking
Need a professional organizer?
Big deal... I can watch the guy count. I can understand what he's doing. Without actually recounting it myself, I can't make sure he's actually counting it correctly. The fact that it's simpler to understand doesn't make it simpler to verify.
How do you determine whether your voting machine is working or not ? You have to employ an engineer ($$$), and then you have to trust that he's not lying to you or incompetent when he tells you it's working just fine. Why bother ?
There's no need for an engineer that costs a lot of money. You just hand recount the ballots. The proof of the pudding is in the eating. In the end, it doesn't matter if the machine is working properly or not. It only matters that it produces the right numbers.
Define "serious challenge". I double-dare you.
I meant "serious" in the sense of formally asking for a recount. If it is asked, my belief is that it should be granted. But full-scale recounts aren't the issue, because you're going to pay the time & effort to hand-count them whether the first pass is done with machines or hand-counted. Your argument that every politician is going to demand a recount is simply not credible. How many politicians demanded a recount in this last election?
Who gets to decide which sample gets tested, and how can we be sure that they don't tip off any would-be vote riggers ?
This isn't a hard problem: you randomly select a few and throw in the outliers in the Republican/Democrat/Independent distribution. You can also allow challenges to certain machines if observers have any questions about them (i.e. there appeared to be 3K votes already on the machine before the polls opened). You have an independent group of observers to do this, so it's not the same people running the machines as testing them.
Another option is just to recount all of the machines by hand. Then in order to rig the election, a worker would have to rig the electronic machines and rig the recount. This is still a time saver because now you only have to hand-count the ballots once.
Your question about how to avoid collusion in order to rig an election is a tough one. But you have the same problems happen with paper ballots, so it's not a problem with e-voting specifically.
And what if your sample finds problems. What do you do then ?
Then you check everything. That's why you have a paper trail in the first place.
I want a transparent, public count process easily understood by the layman...
It's a machine that counts votes. How hard a concept is that to understand?
Hand-counting doesn't provide this either.
To within half a percent, pi seconds is a nanocentury. -- Tom Duff
So far I haven't heard any mention the cost at all. I understand these machines are made with standard PC technology. Windows, access, etc... Who keeps them "windows updated, service packed, compatible with the latest Microsoft Access, and revalidates them every week or so". What happens when Microsoft Access moves on and doesn't do something it used to? How about the hardware, anyone out there still using a computer from four years ago - can you get parts? Is this just a gift to Diebold or what? I like the Canadian method, all you need is a few card tables. No need for such immense sophistication.
Also in Canada.
This only works where there is one thing to choose on the ballot. It would take many hours to tally votes for many positions as I assume is done in the USA. I am custodian in a school that has been used for federal, provincial, and municipal elections. It takes a couple of hours after the polls close to hand count the 'choose one candidate' ballots and finsh the paperwork.
For the municipal election in Edmonton, where we vote for mayor, councillors, public or separate school trustees and any plebicite issues, I feel we have an excellent system. The ballot is paper and your choice is marked by filling a gap in a 1/8 inch wide arrow with a sharpie marker. The ballot is in a privacy sleeve and is immediately fed into a counting machine. The paper ballots are there for verification. After the polls close the machine provides immediate results for the election officials, scrutineers (candidate representatives), and the media - to be compiled at a central location for the official results. The ballots and the voting machine are handled by separate people and transported separately to reduce the likelihood of tampering.
I think it is a simple and elegant solution and it has been used for several elections here.
Do what is right. You will please some and astonish the rest. --Mark Twain
Were you not alive in 2000? There were hanging chads. There were pencil ballots with both Gore and Lieberman marked as their vote for President, and a thousand other little problems with them.
Have you ever ran an election where people use pencils? I ran one at our school where we borrowed voting booths, and had pencils in the booth tied on a string, with big notes to only use that pencil.
People used their own pencils or pens, both of which tripped up the counting machine. Other people made little doodles that tripped up the machine. There's were also problems with people trying to erase a mark they had already made. Some just furiously scribbled over their whole ballot. Many people ripped the pencil out and took it home with them. That was without the problem of sharpening the pencils and making sure we had enough.
Those problems don't exactly instill confidence in me with regards to paper ballots. Each one of those ballots with problems has to be physically inspected, and then there's standards for trying to determine what the voter actually meant, ala Florida. A printout would have been 1,000 times better.
Every voting system has problems, especially pencils. Nothing is going to instill 100% confidence, which is why having multiple systems checking each other is a good idea. We used pencils next year, but we also put a machine in the booth that you'd scan your ballot through that would tell you if your ballot was valid or not. It worked great, but by doing that, were we totally destroying any confidence in the paper ballot method? NO! We just made it more robust, much like adding a paper trail would do.
A singular system by itself will never instill confidence. A couple of systems each checking each other's work goes a long way towards making the voter confident that they will be heard.
Your mileage may vary.
Big deal... I can watch the guy count. I can understand what he's doing. Without actually recounting it myself, I can't make sure he's actually counting it correctly. The fact that it's simpler to understand doesn't make it simpler to verify.
We're really talking about finding a technological solution to a social problem. Until the nation as a whole acknowledges in their heart every citizen's right to cast their vote and have it be counted, we're screwed.
Here's a simple low-tech step in the right direction: Keep the polls open for a few days.
Literalism isn't a form of humor, it's you being irritating.
Why are there these cards at all? Shouldn't the electronic voting machine contact a local district with results via phone line. Make it work much more like a fax machine (transmit results to a location). If the phone line is dead then should a AUTHORIZED person only be able to remove data from the device..
Probably nothing to it? Probably nothing to it, you say!
Then I guess it's just a big coincidence that exit polls have been fairly reliable, up until the point that digital voting machines began to be used. Starting then, exit polls stopped being used as a 'reliable' predictor for the vote.
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
I tried to give $100 but their web form required an address with a US state name. Are their restrictions on donations from non-US citizens or from outside the US?
I am a US citizen living in Canada, and I voted. I am happy to contribute to blackbox.org even if I don't get a tax break. Hopefully somebody will see this and fix things.
Here's a procedure using electronic voting machines that has a better average-case time to results:The key points in both protocals:
- Each voter saw a paper ballot with his vote go into a ballot box.
- The first intermediate numbers are at the precinct level - this allows for an easier audit later, and isolates any issues to as small a pool of votes as possible.
-- The act of censorship is always worse than whatever is being censored. Always.
This is too important an issue to become a vehicle for self-promotion.
Just to keep it real, Nixon trounced McGovern (who btw was a highly declarated WWII vetran) with similar attacks on his ability to providea strong defense, McGovern choose to rise above the fray and lost horribly. But just remember only three years later Nixon had been impeached, due to his attempts to thwart the democratic process, and faced possible a prison sentence. Of course his veep Jerry pardoned him, a move that may have cost him the presidency. I say if anything can be proved let's impeach Dick Cheney first or at the same time. Matthew
What do you mean paper failed in Florida ? I think the combination of the machines and people's non-understanding of how they can be appropriately used had a lot more to do with it. If people can't do that how do you expect them to operate an electronic machine correctly ?
Also the ballots here were intended to be counted by machine.
When I say "paper" voting I mean a piece of paper which the voter marks using a pencil in some discernable way next to the candidate of his/her choice. You can't get any simpler than that.
Thus the ballots are easily human readable. You could use a machine to help count them, but it dosn't have to be an especially complex machine or one which contains knowlage of what the marks on the paper actually mean.
Obviously votes are rejected if the mark hasn't been made properly, but you're not going to get too far in life if you can't use a pencil...
Even illiterate people can manage to use such a system.
The powers that be in this country don't want a simple, understandable system. Fraud would be too easy to detect. We have long verbal wars about voting on TV news shows, but it never clears things up. It just makes things muddier. Why would the news and gov't want to make things so difficult? Why would they act like this is so hard to fix?
Think about it.......
True, but it takes much longer to do a recount than to do the initial count. That's because most jurisdictions which use paper ballots count them at the precincts, which are staffed with workers who are already performing other duties. It's quick because there are hundreds (sometimes thousands) doing the work. Keep in mind though, that some elections have many contests (races for judges, propositions, party committees, etc) which can slow things down considerably. Electronic systems have an edge when the ballot is complex.
When it's time for a recount, those workers have long since been paid off and have gone back to their normal lives. So the election administrators must hire them back, or pull people from their regular staff. They do not have the funding to hire as many workers as they had on election night, so the process is much slower (though presumably more accurate).
Election administrators are under enormous pressure to deliver results quickly, even though the law gives them plenty of time to make their final official canvass. That pressure comes from the news media, who are hungry for a story, and from candidates who obviously have a lot vested in the outcome. So the timeliness of results is a large factor in the decision to purchase counting systems.
I don't think that election administrators view speed as more important than accuracy, at least not the ones I have worked for, however they are very sensitive to the desires of their constituencies, and the "need for speed" is a big consideration. I do think that some jurisdictions have been seduced by the "wow" factor of touchscreens, and have been blinded to their faults.
Personally I prefer optical scan ballots, which are easy for the voter to use (one of the big selling points for touchscreens), relatively quick to count, and which can be recounted, since the ballot is the official record. Touchscreens on the other hand have an intermediate layer between the voter and the record of his vote, and what is worse, adding a paper trail creates two separate sets of records, neither of which is directly created by the voter.
Soylent Green is peoplicious!