Over at Ars, I’ve got a story about the version of the Holt e-voting bill that was passed out of committee on Tuesday. In addition to making more money available to comply with the law’s requirements, the legislation significantly weakens the previous strong source code disclosure rules. The original Holt bill required that the source code be available to anyone who asked. Although the new version does make it available to a lot of people—including parties to election litigation and security researchers—it adds some red tape that’s likely to substantially reduce the number of people who are willing and able to jump through the relevant hoops.
Although I’ve said before that open source voting is no panacea, I find it hard to see an argument against requiring the disclosure of the source code for voting machines. It’s essential that all aspects of an election be transparent and accountable, and the source code of a voting machines is as much a part of the election process as the election judges’ handbook or the rules for recounts. Moreover, voting machines are almost entirely purchased by large institutions, so it’s not like there’s any risk of the disclosed source code showing up in someone else’s voting machine. The only reason I can think of to keep the source closed is to protect voting machine vendors from public scrutiny, which is obviously a reason not to allow them to keep it closed.
Over at Ars, I give a qualified endorsement to the Holt e-voting reform bill:
Serious concerns were raised regarding the flaws with the printers used to produce paper audit trails. Norris cited a Las Vegas survey in which fewer than 40 percent of voters actually checked the paper record of their vote before leaving the polling place. An election official in North Carolina reported that there were hundreds of printer failures in that state during the 2006 election. He cited a Georgia study about the logistical challenges of storing, tracking, and manually counting thousands of votes recorded on unwieldy rolls of paper tape.
Continue reading →
Another person who testified about HR 811 on Friday was disability access advocate Harold Snider. He makes some good points about how DREs improve the accessibility of elections to disabled voters, and raises concerns that the requirement for a paper trail will delay the arrival of fully accessible voting. But then he veers off into hyperbole:
I am very proud of the fact that I was able to complete a Doctorate at Oxford University in 1974, where I studied 19th Century British History. I learned that in early 19th –Century England, a group of people called Luddites attempted to destroy early industrial production machinery because they perceived it as a threat, and had no confidence in it. I believe that the same is true with those who favor H.R. 811. In the 21st Century there are still people who have no faith in modern technology and in its ability to deliver a secure electronic voting process.
This argument is extremely silly, and the supporters of DREs are only shooting themselves in the foot when they make it. The most vocal critics of DREs are computer geeks. Jon Stokes, for example, writes in-depth reviews of new computer chips for Ars Technica. The idea that computer science professors, free software enthusiasts, and the Electronic Frontier Foundation are luddites doesn’t pass the straight face test.
I’ve been reading through last week’s testimony on the Holt bill, and I’m learning that one of the major concerns for designing an election system is ensuring accessibility to non-native voters with limited English skills.
I’m normally pretty hostile to nativist English-only movements. If people want to speak Spanish, or Chinese, or Klingon in their private lives, that’s their business. And if a significant number of citizens are most fluent in a language other than English, I see nothing wrong with the government offering services in other languages. Just today my colleauge Sarah Brodsky did an excellent post about a protectionist effort to require English fluency to be a commercial driver in Missouri.
However, I still have trouble seeing a strong argument for accommodating voting systems for non-native speakers. American politics, at least at the federal level, is overwhelmingly carried out in English. If your grasp of English is so weak that you have difficulty deciphering a ballot, then chances are you’ll have an equally difficult time following contemporary political debates. And if you can’t follow the debate, you’re not likely to make very sensible choices in the ballot box.
I certainly don’t think the federal government should prohibit states from offering multi-lingual voting systems. But I also don’t think it makes sense to require states to accommodate non-English speakers. For states whose politics are carried out almost exclusively in English (which I believe is all of them outside of Florida and the Southwest), I think it’s perfectly reasonable for ballots to be exclusively in English.
The Miami Herald is predicting that newly elected Florida governor Charlie Crist will be the latest elected official to come out against DREs:
Gov. Charlie Crist will recommend on Thursday that Florida’s problem-plagued touchscreen voting machines should go the way of the butterfly ballot–the trash heap–and his proposed budget will recommend replacing them all with optical scan machines, which produce a paper trail, at a cost of up to $35 million.
U.S. Rep. Robert Wexler, a Boca Raton Democrat, will join the governor in Palm Beach County to announce that his proposed budget will include the money to pay for replacing the machines in 15 counties, said Josh Rogin, Wexler’s deputy chief of staff.
”It’s something we’ve been working on for a long time,” Rogin said of Wexler’s six-year battle to require counties to have a paper trail.
Rogin said the governor’s office will recommend replacing the touchscreen machines with optical-scan machines, even though counties have spent millions acquiring the equipment.
The governor’s office would not confirm or deny the reports. The secretary of state’s office said the cost to replace the machines would be at between $30 million and $35 million.
It’s remarkable how quickly the conventional wisdom on this issue has shifted. When researchers raised alarms less than four years ago, they were regarded by many as paranoid eccentrics. Even in the wake of alleged e-voting problems in the 2004 elections, it was still challenging to get elected officials to pay attention to the issue. Now, as more and more politicians begin to publicly question the security of e-voting, it’s beginning to look like it’s only a matter of time before states start scrapping their DREs en masse.
IT&T News is a great publication that features many excellent articles by a variety of free-market policy experts. But I found this article on e-voting, by PRI’s Vince Vasquez, rather disappointing:
The e-voting experience has been a resounding success that has generated relatively few complaints from the electorate. To be sure, there were some legitimate problems with DRE machines on November 7, but many have been found to be man-made, such as innocent user error, inept poll workers, or ineffective planning by local election authorities. Unfortunately, these human-based fumbles have opened the doors for open-source zealots, wide-eyed activists, and crafty politicians who want to scrap DREs for the 2008 elections.
I’m not a politician, and I actually don’t think that open source would solve what’s wrong with e-voting, so by process of elimination, I must be a “wide-eyed activist.” I bet Ed Felten and Avi Rubin–both widely respected computer scientists–would be surprised to learn that they, too, are “wide-eyed activists.”
After busting out that sort of inflammatory rhetoric, you would think that Mr. Vasquez would have some pretty compelling refutations of us wide-eyed activists. But he doesn’t even mention–much less address–any of the actual arguments that e-voting critics make against computerize voting. No mention of the fact that DREs are less transparent, harder to audit, or more susceptible to wide-scale (rather than local) fraud than paper ballots. No mention of the current debacle in Florida, the various reports of problems with e-voting machines, or the fact that computer security researchers have actually demonstrated that some e-voting machines are vulnerable to vote-stealing viruses.
Nope, all we get is vague arguments about how “digital red tape and risky industry requirements jeopardizes the value of these innovative machines.” (Why are they innovative? Because there are computers in them!) And overheated rhetoric about “feeding the country’s voting system to ideological lions.” There might be some good arguments for using DREs, but Mr. Vasquez doesn’t seem to have any.
Ars has a good write-up of a recent report on problems with the 2006 elections:
The report puts to rest the commonly held belief that screen calibration problems could account for all of the reported instances of vote flipping. Some of the offending machines were not touchscreen models–voters used a selection wheel to make their choices. In other cases, the voters would make a touchscreen selection for one slate of candidates, only to have the summary screen (and in some cases the paper tape) report that half or more of the selections had been flipped.
Notably, there were reports of vote-flipping in the hotly contested FL-13 race in Sarasota County, FL. Most recently in the legal contest over that disputed race, a federal judge has declined the Jennings camp’s demands to see the source code to the voting machines used. (In effect, the judge has declared that America’s citizens are not allowed to see how the votes were counted in this very close race, because to reveal that information would violate the voting machine company’s “trade secrets.”)
Other problems described in the report included difficulties with printing the voter verified paper trails (VVPATs) that are required by law in a few states, and that may soon be required by federal law in all states. In some cases, the VVPATs didn’t match voter choices, but the most common problem was that they were simply unavailable, typically due to printing problems.
This is why it’s critical that the paper be the official record for the election result. If I were going to steal a DRE election, the first thing I would do is cause a “malfunction” with the machines’ printer–something that’s rather easy to simulate in software. Therefore, the paper trail is absolutely useless if voters are allowed to continue using machines that are unable to produce paper records. Only if pollworkers know that the paper ballot is what will actually get counted will they ensure that every voter’s vote is properly recorded on paper.
I don’t know about the legal merits, but as a policy matter, this seems like a terrible decision:
A judge ruled Friday that congressional aspirant Christine Jennings has no right to examine the programming source code that runs the electronic voting machines at the center of a disputed Southwest Florida congressional race.
Circuit Judge William Gary ruled that Jennings’ arguments about the possibility of lost votes were “conjecture,” and didn’t warrant overriding the trade secrets of the voting machine company.
Democrats in Congress meanwhile, said they’d allow Republican Vern Buchanan to take the seat next Thursday, but with a warning that the inquiry wasn’t over and that his hold on it could be temporary.
The state has certified Buchanan the winner of the District 13 race by a scant 369 votes.
The ruling Friday from Judge Gary prevents for now the Jennings camp from being able to use the programming code to try to show voting machines used in Sarasota County malfunctioned. Jennings claims that an unusually large number of undervotes _ ballots that didn’t show a vote _ recorded in the race implies the machines lost the votes.
The source code to a voting machine in a DRE election is analogous to the voting procedures manual in a traditional paper election. It would clearly be absurd to run an election in which votes were counted in secret and no one was allowed to know what rules were used to determine the winner. Transparency demands that anyone be permitted to inspect the voting procedures to make sure they are fair and accurate.
Precisely the same principle applies here. The source code is the vote-counting procedure. It’s absurd to ask Jennings to prove there was a problem with the voting process before she can be given access to the details of how the election was conducted. Obviously her allegations of miscounting are only “conjecture” at this point, because she hasn’t been given the opportunity to examine the process by which the votes were counted.
My friend Kerry Howley has a fantastic piece on the bizarre state of the law regarding ownership of human body parts. Self-annointed “bioethicists” claim that we can’t give people the right to control what happens to their own body parts, because that could lead to a world in which body parts “become nothing more than chattel going to the highest bidder.”
And that would be bad because… well, it’s not clear why. Certainly, the research community hasn’t been shy about using their control over patients’ tissues to enrich themselves:
In the end, disputes of this kind always come back to John Moore’s million dollar spleen. Twenty years ago, UCLA School of Medicine Dr. David Golde told Moore his leukemia-ravaged spleen would have to go, and Moore agreed to have surgeons remove the organ. For years afterward, Moore would fly from his home in Seattle to UCLA, where Golde would check on his progress and take samples of sperm, blood, and bone marrow aspirate.
Unbeknownst to Moore, his supposedly trashed spleen was teeming with biomedical treasure. Golde derived a commercial cell line from the disembodied organ, and proceeded to patent it. Eventually, Moore became suspicious at the steady stream of vague release forms he was being asked to sign. He investigated, caught Golde, and sued. In 1990, a California court ruled that Moore had no proprietary right to the blood and tissue taken from his body. By that time, Golde had sold the patent for $2 million.
So apparently, it’s ethical to make large profits from human body parts, but only if the patient who provided the body parts doesn’t get a penny of it. Sounds ethical to me.
Kerry’s article is worth reading in full.
It seems that the Technical Guidelines Development Committee has accepted the NIST recommendations against e-voting after all. As a compromise, the new guidelines apparently emphasize that existing voting machines are grandfathered in.
This strikes me as a reasonable compromise. Obviously, I’d like to see us scrap paperless voting machines for the 2008 election, but there is likely to be substantial political resistance to that idea. Since TGDC doesn’t have the authority to ban paperless voting in any event, the important thing is that they come out with a strong statement that machines without paper trails are insecure. That will give the good guys in individual states the ammunition they need to accelerate the process of phasing out insecure machines.
The Technology Liberation Front is the tech policy blog dedicated to keeping politicians' hands off the 'net and everything else related to technology.