San Diego Diebold Poll Worker's Report Posted
James Renken writes "I was a poll worker in San Diego for this year's primary election. It was the county's first using Diebold voting machines, and as you may have heard, we ran into some problems! My full report of the goings-on can be found at Live from the Nuke Free Zone. Enjoy!"
"I think that current electronic voting systems are better than the traditional systems in terms of security, and also in terms of usability for most people."
But how are those Diebold machines at allowing a recount? Do they finally create a paper trail, or is it still "faith-based voting"?
Wouldn't the loser be more likely to complain?
It seems like these Diebold systems have all sorts of features like smart cards and locks that make them look secure, but when you actually kick the tires you realize things are not as secure as they should be.
We'd be much better off with a system that produces prints a human readable and machine readable piece of paper, and then put those pieces of paper into a ballot box. At least, when the security of a box in plain sight gets compromised we know that something happened... the worst case here is swearing in a losing candidate.
All the places where he said
"It appeared to have recorded all of the votes properly, but I can't be 100% certain" or apparently.
With 1,000,000 people voting, an error 1/1000 is enough to change the results of election for the whole state. We need paper ballots.
It is even scarier, because he was a poll worker and did't realize this.
So, I'm going to argue against the seemingly overwhelming majority of people out there about voting machines.
:)
Yes, there are issues. Are the issues any worse than what can/has been done in the past with non-electronic voting? Probably not. I think that Florida proved that you can tamper with the old system just as well as an electronic one.
Eventually what voting comes down to is trusting that the people who run the system are honest. If you have dishonest people anywhere in the chain; you're going to get bogus results. The only solution that I can see to this issue is that the process must become more open.
1) Diebold needs to modify the machines to produce a printed slip that shows the party X voted for. X is then responsible for ensuring that the slip makes it into the collection basket. Bar codes can be used to correlate votes in the machine and votes on paper, and verify that they match. Because the electronic vote must match the paper vote, and because the user can verify the paper vote themselves, it becomes harder to cheat.
2) Make the vote counting process open. If anyone wants to, let them watch votes being counted. Canada does this, why not? Votes are counted to verify election results. In the event of a discrepancy, the paper votes would be used. The electronic tallies would be used for "quick" results.
Well, it's an idea, anyways
What, me worry?
And then you'll have a paper trail
Indefinitely Detained US Citizen
This never seems to be addressed, but cost.
If these machines are more difficult to operate and more expensive to maintain, and require the hiring of additional personnel to administer, why are they being used?
Paper ballots seem exponentially cheaper in all respects, and I haven't seen a piece of oaktag crash in many years.
Ok, I admit Diebold's systems are flawed. No Paper Ballot(tm), no trust. I agree.
/.ers are so frequently decrying "Method Patents" if we cannot fathom the creation of a *fair* automated tallying system (something more 'complex' than paper), why not question what the system itself is?
Diebold and its advocates are bent and determined to use them in elections. OK, lets do that.
The Comprimise(TM): Change the voting Method.*
If your country is split so close, so narrowly through the center, that the *POSSIBILITY* of tampering is not 100% obvious (that causes those riots in the streets...) why not look to garner a better concensus? Why not consider altering the *structure* of the debate? Why not consider the method?
We
If your public discourse is incapable of discussing *that* issue -- Real Reform of Government (like, I dont know, maybe more than a Democracy of the Republicrat Party). If your paperless ballot system was meant to build concensus, you wouldnt have this debate in the first place.
NO LARGE GROUP WOULD BE UNHAPPY WITH THE RESULTS. Maybe the "one person one vote, winner take all" system is just a little dated? Lets start communicating. Lets focus enough to discuss our governance...if we cant, why build all these "communication tools?".....oh, look, a shiny thing...
*Woha, woha, woha. Before you go flaming me, or modding me down, I am not delivering a flippant "this is the solution" answer, im suggesting a place to start thinking. I am not for, or against, *that particular method*. There are many, how about PR? (Use Google))
If you don't trust Diebold, request an absentee ballot. In California, at least, these are still old fashion recountable paper.
The cake is a pie
The central problem with electronic voting lies not with bugs, hardware failures, or security, but rather than with our modern concept of democracy and where we see ourselves going as citizens of a common nation.
It is clear that present electronic voting efforts are the first step in a general program of transforming voting as we know it into an online, decentralized process, with the final goal being a system where voting is an activity as simple and hassle-free as ordering a pizza or sufing to a website.
Therein lies the problem. As citizens of a common nation, our involvement in the democratic process should be something that brings us together, together with people we would not ordinarily encounter, for what can only be called our sacred ritual of casting votes. A ballot should not be a screen with virtual buttons floating in hyperspace, it should be a hefty card, symbolizing the hefty decision that lies with each voter, it shows our seriousness that decision needs is embodied in the real physical ballots that are carefully tallied and counted and not simply disseminated into electronic bits.
The process of voting should be a little inconveniencing, with voters having to drive to the polling station, stand in line, and punch a ballot. It reinforces our sense of civic pride to have to make a bit of an effort to vote. It demeans the democratic system for the voting process to be allowed to atrophy into a simple matter of point-and-click, no need to get out of your chair. Choosing the laws and leadership of a nation should be an act more involved than switching channels.
When the once-proud rituals of democracy are reduced to a set of simple gestures, once the paricipants in the voting process are reduced to a mass of isolated individuals typing on keyboards or pushing buttons on PDAs, a sense of togetherness is lost. The insiduous decentralization of the voting system that is the end result of electronic voting can only lead to the erosion of our sense of citizenship, of being equal paricipants in something larger than ourselves. Could the erosion of democracy itself be far behind? It seems the dystopian corporate-run societies of so-called "cyberpunk" "fiction" more than just sci-fi.
I don't remember "civil disobedience" as practiced by Ghandi or MLK including breaking things... it was generally a peaceful type of thing. Maybe a sit-in, demonstrating outside the machines and explaining to the people who come in to vote what's wrong with them, or disobedience like this - but not wrecking the machine. Anyway, these machines seem to do a good enough job of breaking themselves - why go to all the trouble?
Hmmm..
>It's a P4 2GHz with 512mb of ram (wtf?! why on earth does it need that)
"Rich" media ads. You'll need some power to play compressed video of that next hollywood blockbuster while you wait for your cash. Or maybe its just cheaper to buy 'off the shelf' PC commodity stuff. Prob both. Thats probably why WMP was installed.
The speakers are for the future ads and for the interface for the blind. Most have headphone jacks too.
My speculation is that it caches the bloated application(s) into a RAM disk. Makes things quicker. RAM is cheap.
Copyrights, Patents, Trademarks: temporary loans from the Public Domain, not real property ("intellectual" or otherwise)
Man, I totally respect your idealism, however completely ineffective it would be in our society today.
You: smash a voting machine
Media: some nutball tried to destroy the democratic process - cut to clips of Idaho cult training
next election
You and 20 other people you've convinced: smash a voting machine
Media: Terrorists(tm) try to keep Americans from voting, cut to footage of people wearing turbans with "Al Queda" crawl
assuming the nobility of your crusade at the next election manages to recruit more people:
You and 500 others: smash voting machines
Media: Poignant 30-second segment showing the entrails of battered voting machines (soundtrack provided by the currently popular country music whore) spliced with doctored images out of context of evil-looking, misguided perps, if one of the 500 is Middle Eastern, he will be exclusively focused on, cut to additional segment of children crying, cut to trade center attack, cut to Oklahoma bombing
You: electric chair
Media: congratulates itself for once again, protecting the American way of life(tm)
It has to be collected and retained like an old fashioned paper ballot. In which case, electronic voting offers nothing.
Say it prints out the vote. The tally still says what CEO "I am committed to delivering Ohio's electoral votes to G.W. Bush" wants the tally to say.
Canada counts paper ballots under the watchful eye of partisans, and gets it done in a few hours. How many ballots can you count in an hour? Hire enough counters and let the parties watch. Done fsking deal.
I was in the odd position of being called a luddite by a computer science academic. He was Russian, and figured electronic vote rigging was not such a new thing. I have a little more of a security background, and I think the challenges of securing the system end-end are insurmountable. As the systems are now, they are laughably easy to corrupt. And is there any greater incentive to cheat than political power?
Do you think they've got a room full of sweatshop workers chained to desks counting them by hand? No. They count them with a punch card reader. Okay, so I've never used one (I'm not _that_ old), but I really doubt that they're impossible to compromise.
I doubt that they are *impossible* to compromise, too. But, you're omitting a few relevant details:
- These machines work on technology that has been essentially unchanged for many, many years. The code is simple, and likely is public domain, since the government probably developed it. The process is mostly mechanical.
- Access to those machines is very limited. The people who use them are all employees of the election system, which doesn't put them above scrutiny, but does make them far more accountable than Joe Voter. You don't have the punch-card readers out on public display for literally every registered citizen to be able to walk up and stick a smartcard into.
- When there's a question about the results, a *hand* recount can be and is done. This happened in Florida... hundreds of people *did* sit there and go through those cards by hand, tabulating votes.
Obviously, the issues are far from parallel, and there are methods of dealing with them in the case of the old punch cards. Here, there's simply no voter-verifiable record... so no recounts possible.
Don't you wish your girlfriend was a geek like me?
I don't get it, are you saying anyone can go and vote at any polling place if they know the name and address of someone who lives in the allotted area? What am I missing?
True. They were much, much worse. Reread your post, please!
I started poking around the root filesystem looking for a link to the executable.
What are they thinking giving every clerk everywhere root?
I finally found the actual location of the executable--it seemed to be on a datacard of some sort--and started it for them.
Other posters have noted that you have no way of verifying that this is the correct excutable. What if it was a testing version, or something else uncertified by the state?
We had one voting maching give a blank page to someone when it was in large print high contrast mode, but we just hit next and it was fine.
And this is the sort of thing that can be horribly troublesome. People with poor eyesight are mostly (though not exclusively) the elderly--a group that are not known for their comfort (in general) with computers. And here they are with a blank screen.
One of our machines failed to print -- it just cut off in the middle and wouldn't reprint (some paper trail, eh?).
How many people voted at that machine? A hundred? Five hundred? How many votes are now either irretrievable at worst or highly suspect at best? Even though it couldn't print the totals, you expect it to submit electronically the correct tally?
The worst part was that the voting stations give a total number of votes cast onscreen and a total on the printed tape, and on all of our machines but one, these did not match. They were all off by one vote.
How is this problem not very, very serious? First, you lose whatever thin reassurance the total provided. The system now is without an effective check on number of ballots cast. Second, if there was an error--if it was randomly distributed then it's unlikely (though not impossible) for it to affect an election. If it was systematic (deliberately, or just a programming error that inadvertantly doesn't count the first vote for the first candidate, or something like that) then this could be very serious. If five hundred people use each machine, and you lose one of every five hundred votes for a candidate, that's an error of 2000 votes per million ballots. That's appalling--and larger than the margins in a number of states in the last Presidential election.
~Idarubicin
What is wrong with paper ballots ? Maybe someone can enlighten me. I live in a 80M poeple nation where paper ballots seem to work without any problem. We have elections, some of them rather complicated, and usually you have to vote on something every 1-2 years. We always get speedy and dependable results - even with national elections that are way mor complicated than a US presidential elections we usually have a stable estimation before midnight on voting day, ad a provisorical result sometime next day. There never was any problem with vote fraud worth mentioning (though occasionally a politician gets in trouble with the finer points of anti-vote-fraud laws) and basically the perception is that this just works.
Why can't yopu make that work in the US ?
An MIT/Caltech study of voting technology found that paper ballots are the most accurate.
The 2004 Democratic primary had a turnout pattern of primary-specific apathy (lower than expected votes) and caucus-specific inspiration (high and record high votes). Why did the New York primary record a 20-year low turnout on the same day that the Minnesota caucus recorded a 33-year high turnout?
South Carolina's state Dem party fought pressure from the national Dems to institute a loyalty oath for voters, which would have torpedoed Edwards. State officials chose to hand count paper ballots for security, even though machines were available. South Carolina was one of the few 2004 primaries to report record turnout and the only state where Edwards won.
Hand counted paper ballots are the gold standard of voting. Cheaper and faster are neither necessary nor desirable properties of biennial elections.
yes, that is correct. In order to provide anonymous voting to all registered voters, you are not required to have ID. The only thing keeping you from using someone elses name is the possibility that they will show up later (or have already shown up). In fact, this seems like a fabulous way to invalidate _all_ results at a polling place which is not likely to vote in your favor. Simply show up early with a couple of dozen people who claim to be local residents. When the real people show up later there will be no way to determine which votes were valid and which weren't.