Slashdot Mirror
CA Secretary of State Bans Diebold Machines
Etcetera writes "The CA Secretary of State has just announced that they're pulling the plug on the use of Diebold voting machines (thank you KNSD) as a result of the flaws that came up where they were used during March's elections. More background on the issue (not updated yet) from the Secretary of State's perspective is available here."
Myself and my family are from Napa, CA (one of the cities that had some serious problems with Diebold), and I can't explain how frustrating it is to not be sure if your vote was counted properly or not. For democracy to work, you must have faith in the security and validity of the elections. Diebold has seriously undermined this, especially in my hometown. The jokes and grumblings have been raging, not to mention the rumors of the end of our Registrar of Voters' career. Although "no harm, no foul" has been claimed, confidence has been undermined, which IMHO, is one of the most important aspects of a good democracy.
Maybe things do work in this system.
The basic problem here is that, in 2000, the public was shocked, SHOCKED, to learn that if you ask a person to "choose one", there is about a three percent chance that he or she will attempt to make an invalid choice, such as two candidates (overvote), or 1/10 of a candidate (unintentional "undervote"). Sometimes this is a mechanical problem (e.g., hanging chads) and sometimes it's just a combination of poor directions and voter stupidity.
Prompt ballot scanning can prevent some overvote problems. If two ovals are obviously marked, or two chads are obviously punched, then the ballot can be rejected, and the voter can have a do-over. (Prompt scanning is common for optical ballots, and rare for punchcard ballots.)
But this does not solve all problems. For optically scanned ballots, a voter can make a faint mark that is apparent to a human but not readable to the machine. Or a voter can make a circle around an oval that the machine disregards. (The "fill-in-the-arrow" style largely avoids this problem.) For punchcard ballots, a chad can be incomplely punched due to some defect in the card, the stylus, or the voters.
Electronic voting machines and lever voting machines prevent these problems. With these, it is not only impossible to cast an invalid vote. It is also impossible for anyone to look back at the evidence and say, "Well, it looks like the voter meant to pull this lever / touch the screen here, but didn't try hard enough."
Of course, certain voting errors cannot be prevented by any scheme. If the voter indicates a different choice than he or she intended, and then does not check the results, then the wrong vote is cast.
I would guess that (on average) voters have this kind of core logic fault at least a few percent of the time. People tend to vote in a hurry, and many do not take it very seriously. This is inherently uncorrectable.
Of course this has nothing to do with the basis for this particular decision regarding Diebold. The stated basis of this decision was the set of serious problems experienced in getting their machines operational. Unstated, but probably about as important, was the fact that Diebold has been demonized by every liberal and semi-liberal mouthpiece in the country for the last two years. No other voting machine manufacturer is going to be held to the same level of scrutiny as Diebold at the moment. Not that this is necessarily a bad thing...
of all the intelligent and objective people on Slashdot, and in so many other forums, who saw all of these issues two years ago:
WE TOLD YOU SO!
Now... if you're ready to implement a reliable, trackable, scalable and secure electronic voting system, I think we're all still willing to help you. If you'll just listen this time.
Article X: The powers not delegated... by the Constitution...are reserved...to the people
I used one as well.
I was in a hurry, as I was voting before heading off to work. After I finished voting, I was walking to the table to hand the card back when my cell phone vibrated. I walked outside with my phone to take the call, away from the other voters... with the card still in my hand.
Ok, so that would have just meant that ~I~ didn't get to vote... which would have been bad, but not the end of the world. That's not the interesting bit though! What I later heard was that there is only one card per voting machine. Had I not returned the card, that machine would have been out of order for the rest of that election. I can't confirm this to be true, but if it is, that's really scary!
I think any third rate magician wouldn't have a problem substituting a card of their choice for the community card in this system.
Actually, come to think of it, you could swap the card out in private right at the booth.
I wonder how long it would take for someone to come up with a pirate card for this (assuming tha the machines ever see the light of day again)? A read-only card that would cast the same set of votes over and over again...
That's an interesting point, but the paper ballots are counted in public by public officials following rules created by your state legislature. For elections in Michigan, anybody who wants can observe the election procedure, and can complain if they think it's done poorly; I'd imagine you could do something similar to monitor a vote recount.
The difference here is that the electronic ballots are tallied in secret by secret software written by a private company, with no observation allowed.
For electronic voting to work, it's going to need to use completely open software, where many experts can verify that it will work properly. Since so many people have an interest in the system working perfectly, there will be lots of people reviewing the code, and I think that very few serious bugs would be there for long.
verifiedvoting.org is advocating this same position. I'm not sure if they're actually writing software.
My Web Page
I'm not registered, but per the second article:
In fact, I recall reading the first article in the San Jose Mercury News when it was printed, and evidently the machines Diebold installed were a second-generation set. Their first-gen. machines had been approved a while ago, and so they evidently tried to cut corners, assuming the second-gen. ones would certify as well, and went ahead and installed the machines before they were certified.
On the other hand, I think it's interesting to wonder whether or not they really would've certified. Is it possible that the circumstances that led to the failure of these second generation machines may've also lead to the failure of the first generation machines, as well? I suspect the CA Gov't officials are dodging a bullet here, since Diebold seems to come out as the only fall-guys here (and rightfully so, as far as they're concerned).
I defer to anyone who has read more about this than I, which isn't much to begin with.
Moo
What if the insecure voting machines aren't the result of incompetent programmers -- what if they've been made insecure on purpose so they are easy to manipulate. You can't tell me right now some republican hitman doesn't have a machine and isn't figuring out how to walk into a polling station and cast 5,000 votes at once. After all, this is the most important job in the world we're talking about here.
And if anyone ever finds out -- theres no paper trail, no audit, no log, no way to know what really went on, and it was done on purpose by a company whose president swore to deliver electoral votes to bush.
Religion is a gateway psychosis. -- Dave Foley
A quick thought on overvoting. This is an "error" that should be available. It indicates that you don't agree with any of the canidates, and unlike an "undervote" this cannot be argued as voter error later.
Error conditions should always be avoided. If a voter truely wants to not vote in a race, they should be asked to fill in an oval that indicates such.
In some places in the world, a "None of the above" or "None of those listed" option is on the ballot in what is called a "Turkey Ballot" format. Basically, if the non-option should obtain a specfied in the rules number that indicates a critical mass, the election must be reheld, but all of those who appeared on the ballot in the first vote are disqualified because voters have declared them all turkeys. The parties are basically each being told to nominate somebody else because the voters didn't like any of the options they were presented with.
I agree that the "grand conspiracy" theory that Diebold is actively trying to sweep Republicans into power is a bit much...
However, Diebold played right into those people's hands with what they did, running an uncertified product which skipped all of the checking processes that were in place to make sure that nobody is attempt to cheat the system. The first set of canaries in this mine are dead... we shouldn't allow this operation to go any further because once they go arround the certification process, there isn't really much more needed to taint the vote counts and get away with it.
Quess what, I'm still against electronic voting of this sort. The machine doesn't produce a paper trail, and there is now way to find out, ever, if fraud has taken place by the machine's supplier or operators. The fact that a paper ballot system is unwieldy
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
"The basic problem here is that, in 2000, the public was shocked, SHOCKED, to learn that if you ask a person to "choose one", there is about a three percent chance that he or she will attempt to make an invalid choice, such as two candidates (overvote), or 1/10 of a candidate (unintentional "undervote")."
Can't say I was particularly shocked at this - sorta expected it. If I think about my (and most likely our) field of expertiese look at e-mail worms/viruses. Wow, how can someone be so stupid to open said attachemnt. Given that the overwhelmingly do, how can we expect them to vote correctly. Then take into consideration 9as the parent notes) mechanical failures and what do you expect?
And no, I do not intend this as "funny". Just think about the level of stupidity that has someone opening a "I love you" attachment (or even "I'm a virus" which people where I worked opened) and ask how you would design a fool-proof voting mechanism for them. Especially given that there is a certain amount of error from even competant people that you can not avoid.
And I will agree with the parent that this doesn't exclude Diebold from being incompetant.
------- Sorry about the spelling, I suffer from two problems. Dyslexia makes it difficult to spell well, lazy makes it
It should be quite possible to make ballot readers and verifiers in a similar way. The verifier can be operated by the voter (so the vote stays secret) to verify that the ballot will be read correctly.
Mind you the 9th circuit is the largest out of any district, and definitely the most liberal out of all of them.
Some of non-9th circuit folks are somewhat glad that the 9th circuit pushes the limits on regular basis. For someone has to, and for each time the Supreme Court over-rules them, they have to justify why, which becomes even more interesting.
Sunny Dubey
We use papertrail-less electronic voting in Belgium too, and there was an incident where a spontaneous bitflip in the counting machine's memory caused a miscount of 8192 votes. It was blamed on cosmic radiation. And no, I'm not kidding.
Still, the vote gets recorded to a separate magnetic card for each voter, so it is possible to retally, even though you have to take the voting machine's word for it that your vote was recorded correctly.
Of course, the problem is that there are so many candidates elected by default in the U.S., it's likely that nobody would ever get elected to some offices, and the number of elections would be horribly expensive.
However, it's a fun game to play to think about whether, should this system come about, whether actually competent people might have to start running for office. We might even have to start drafting people to serve, as in one of Heinlein's speculative libertarian future Americas.
Because someone might like to pay you for that receipt.
Ita erat quando hic adveni.
CAN Diebold appeal this? I know the Supreme Court currently (and Justice Bush-Is-God Scalia in particular) tends to be very business-friendly, states-hostile and... Liberal in their interpretation of the law. So if it does get to them, I have no doubt that they'll rule that CA cannot, in fact, use any voting machines other than Diebold's. But this was a direct order by the (Republican!) CA Secretary of State. I'm not sure Diebold can appeal that.
Or, at the very least, don't they have to work their way up? And thus, the 9th circuit court can refuse to hear the case...
The Wired article shows that many of the system's vulnerability were due not to the GEMS software itself, but to the W2k operating system.
So from now on, if anyone insists on choosing MS over other solutions for mission-critical system, and says "nobody ever got fired for choosing MS", we can point them to the Diebold debacle. Not only were they fired, they got it rubbed in and on national headlines!
--
Mad science! Robots! Underwear! Cute girls! Full comic online! http://www.girlgeniusonline.com/
Well, it was the shenanigans BEFORE the election that really hurt Gore. Katherine Harris, the florida election comissioner took pains to rig the election *before* anyone knew that florida would choose the president. One of the strangest things they did -- they wanted to remove felons from the voting roles (illegal), but there was no way to do this for some reason. So they took the list of felons, and deleted EVERYONE who shared a birthday with any of them -- some 40,000 people.
Other strange things -- the counties that did have electronic voting machines (some did), in white counties, the machine wouldn't allow you to form an incorrect ballot, it would warn you and ask you to recast the ballot. In black counties, the SAME machines accepted the ballot and threw it away letting the person think they were voting.
That's just the begining of the weird antics in florida -- the problem is it demonstrates an unwillingness to play fair in the democratic process
Religion is a gateway psychosis. -- Dave Foley
"Good point. Is the guaranteed accountability of a paper system worth the additional inaccuracy it might introduce?"
They are not mutually exclusive. A DRE that has a paper trail has the best of both worlds: the theoretical 100% accuracy and auditability. After each election do a random sample of precincts to audit and count the paper trail; if they are the same or off by a margin within the margin of error of the paper machine, accept the electronic total. Otherwise you can have a full-blown recount with all the care you want.
"How many lines of code will a voting machine take? For it to be "fully audited", somebody needs to spend that $1000 per line of code.
"And don't forget that the voters are expecting a shiny antialiased GUI"
I'd argue that this is unnecessary. Look at, for example, ATMs. Do people have issues with how ATMs work? I hardly consider this a shiny GUI. I'd go so far as to say that a touchscreen is unneccessary; ATM style buttons down the side should be plenty. In case there are some people who are technophobic and don't like even ATMs, you can provide a paper ballot counted in the traditional way.
"...everything down to the disk access, kernel and filesystem needs to be verified."
So base it off of something. Surely there is already at least some sort of OS that has this degree of auditing, so build on top of it.
"It's a big job, just to reach the same standards that paper ballots already offer."
No, because once you have the audited system you have all the benefits of a DRE system (instant vote tallies, 100% accuracy, no overvotes/undervotes, easily verifiable results, etc.)
So the Supreme Court said that the first amendment guarantees the right of association in political parties. I don't see why Democrats or independents should have a right to vote for my Republican nominee. I am sure you would hate to see non-Democrats choose the Democrat nominee.
For example, I would've chosen Zell Miller as the Democrat presidential nominee. That would've left a competition between Zell Miller and President Bush. What choice would people have then?
The open primary system was flawed from the beginning because of this. The parties would actually direct their members to cross the party line and vote for a weaker candidate for the opposing party. (This is how the Socialist party hijacked the Democrat party, in fact.) This is not representative government, and it is not fair to have to put (R) next to a person nominated by Democrats.
The radical sect of Islam would either see you dead or "reverted" to Islam.