Voting Isn't Easy, Even if Cheating Is

The Open Voting Foundation's disclsosure that only one switch need be flipped to allow the machine to boot from an unverified external flash drive instead of the built-in, verified EEPROM drew more than 600 comments; some of the most interesting ones are below, in today's Backslash story summary. Expressing a common sentiment, reader cmd finds nothing innocent about the inclusion of such a switch:

Diebold also builds automated teller machines (ATM), the definitive model for reliability and accountability.

The AccuVote machines are what they are, not due to poor design or unintentional mistake. They are the result of a deliberate intent to enable fraud on a massive scale. Viewed from this perspective, the AccuVote design is very good. The real problem comes when Diebold realizes that it needs to become better at obfuscation and makes it harder to detect the fraud.

"Electronic voting machines with no paper trail are an insult to democracy," writes pieterh. "That they come with switches to bypass even the dubious 'safeguards' provided is hardly a surprise."

Paper trails, of course, are only as good as the people guarding the paper; readers familar with more recent allegations of vote manipulation may be interested in the 1946 confrontation in Athens, Tennessee (pointed out by reader William J. Poser) between WWII veterans and the election officials.

Reader Soong, though, provides a conspiracy-free explanation for the presence of such a switch:

The ability to boot from different sources is a normal debugging feature, not in itself sinister. Should they have cleaned that up on the production model? Yeah, sure. But verifiability is ultimately a human concern anyway, not a tech one.

It all comes down to who you trust.

If you don't trust the polling place, make the voting machine tamper proof. But then you have to trust the guy who built the voting machine. You have to trust the guy who loaded the software on it at the factory or the elections office. You have to trust the guy who wrote the code. Even if you inspected the code, you have to trust him to give you a binary based on that and not pull a fast one. You have to trust his compiler to give him a binary without compiled in back doors. I feel like I probably haven't listed all the points where this voting machine chain of trust can break down.

Several readers pointed out that voters might better trust the machines as well as the process of electronic voting if regulation were more rigorous; as reader Animats puts it, "slot machine standards are much tighter":

The Nevada Gaming Control Board has technical standards for slot machines. They've had enough fraud over the years that they know what has to be done. Some highlights:

• ... must resist forced illegal entry and must retain evidence of any entry until properly cleared or until a new play is initiated. A gaming device must have a protective cover over the circuit boards that contain programs and circuitry used in the random selection process and control of the gaming device, including any electrically alterable program storage media. The cover must be designed to permit installation of a security locking mechanism by the manufacturer or end user of the gaming device.
• ... must exhibit total immunity to human body electrostatic discharges on all player-exposed areas. ...
• A gaming device may exhibit temporary disruption when subjected to electrostatic discharges of 20,000 to 27,000 volts DC ... but must exhibit a capacity to recover and complete an interrupted play without loss or corruption of any stored or displayed information and without component failure. ...
• Gaming device power supply filtering must be sufficient to prevent disruption of the device by repeated switching on and off of the AC power. ... must be impervious to influences from outside the device, including, but not limited to, electro-magnetic interference, electro-static interference, and radio frequency interference.
• All gaming devices which have control programs residing in one or more Conventional ROM Devices must employ a mechanism approved by the chairman to verify control programs and data. The mechanism used must detect at least 99.99 percent of all possible media failures. If these programs and data are to operate out of volatile RAM, the program that loads the RAM must reside on and operate from a Conventional ROM Device.
• All gaming devices having control programs or data stored on memory devices other than Conventional ROM Devices must:
  1. Employ a mechanism approved by the chairman which verifies that all control program components, including data and graphic information, are authentic copies of the approved components. The chairman may require tests to verify that components used by Nevada licensees are approved components. The verification mechanism must have an error rate of less than 1 in 10 to the 38th power and must prevent the execution of any control program component if any component is determined to be invalid. Any program component of the verification or initialization mechanism must be stored on a Conventional ROM Device that must be capable of being authenticated using a method approved by the chairman.
  2. Employ a mechanism approved by the chairman which tests unused or unallocated areas of any alterable media for unintended programs or data and tests the structure of the storage media for integrity. The mechanism must prevent further play of the gaming device if unexpected data or structural inconsistencies are found.
  3. Provide a mechanism for keeping a record, in a form approved by the chairman, anytime a control program component is added, removed, or altered on any alterable media. The record must contain a minimum of the last 10 modifications to the media and each record must contain the date and time of the action, identification of the component affected, the reason for the modification and any pertinent validation information.
  4. Provide, as a minimum, a two-stage mechanism for validating all program components on demand via a communication port and protocol approved by the chairman. The first stage of this mechanism must verify all control components. The second stage must be capable of completely authenticating all program components, including graphics and data components in a maximum of 20 minutes. The mechanism for extracting the authentication information must be stored on a Conventional ROM Device that must be capable of being authenticated by a method approved by the chairman.

Those standards cover the possibility of an "alternate program" in a slot machine, and provide a way to check for it, with logs and an external program check capability.

The Gaming Control Board of Nevada was asked to take a look at Diebold, and Nevada rejected Diebold equipment as a result.

Voting machines need tough standards like that. They don't have them.

Even if e-voting machines had a spec list that would pass at the Gaming Commission, Midnight Thunder is puzzled that tamper-proofing techniques aren't more evident on the Diebold machines:

Given taxi meters and electricity meters both have tamper seals, you would have thought that these would have visible tamper seals as well. If in doubt you could even have two tamper seals: one from Diebold and another from the voting commission, in order to ensure that both parties are satisfied with the state of the machine.

Several readers are for canning electronic voting for U.S. elections completely. Reader Iamthefallen wants to know

Has anyone answered the question regarding need for automated vote counting in a satisfactory way?

Seems to me that manual counting of votes would be vastly more secure as it would take a huge conspiracy to affect the result either way.

Counting a hundred million votes is hard, counting a thousand votes in a hundred thousand locations is easy.

Similarly, slofstra writes

Sorry, I have never seen the point of these machines. Paper ballots are auditable, user friendly, and if electronics is put into the reporting system, can be counted in a few minutes and submitted. Voting machine are a perfect example of a technology fetish at work. It would make an interesting case study to examine the economic and sociological reasons why we sometimes buy technology that we don't need, don't want and further, serves no useful purpose.

(Augmenting electronic voting machines with a paper record is a frequently raised idea; reader megaditto, for one, asks "Is it that hard to put a thermal printer behind a glass shield?" A similar system is required in Nevada voting machines already.)

Paper ballots and electronic ones aren't the only options, though; lever-based voting machines have dominated recent American national elections. Mark Walling writes

My district switched to electronic- from lever-based. in 2004, at 7:15 when I voted on lever machines, there was no line, and just about as many signatures in the book. In 2005, the line was out the door and around the corner at the same time. The person in front of me took 5 minutes to use the electronic machine. People knew how to use the old machines, and they were reliable. These new things take the old people forever to use, and then they complain that they were hard to read ...

Reader WillAffleckUW suggests skipping in-person voting completely; absentee voting is a good idea, he argues, not only in light of the flaws (demonstrated or alleged) in electronic voting methods, but because

absentee voters get a paper ballot that is not only delivered by a trusted source (the U.S. Post Office) who have a verified date/time stamp — and that the ballots can be audited, traced, and verified — now that is a reason to register permanent absentee.

Not so fast, says reader JDAustin:

I suggest you take a look at the research into the recent Washington state elections done by SoundPolitics.com. They verified close to a 20% error rate in absentee balloting. The signature verification on absentee balloting is no verification at all due to non-verification being done by those who count the ballots. Additionally, the USPS is not a trusted source, they are just another government bureaucracy. The ballots themselves cannot necessarily be traced nor verified — and even when the signatures are completely different, they are still counted. Due to the nature of voter rolls, duplicate ballots are sent out all the time due to slight variation in a person's name, and the duplicate ballots counts are not caught until after the final tally has been done and the election finished. Finally, mischievous government officials can always delay sending the military their ballots so those serving overseas do not have time to get their vote in on time. This actually happened in 2004 in Washington state.

Permanent absentee is not the solution. Neither is electronic voting.

The true solution takes elements of the recent Mexican election to prevent fraud (voter ID cards, thumb inking, precinct-based monitoring and tallying) and combine them with the best paper-based voting machine.

Many thanks to the readers (especially those quoted above) whose comments informed this discussion.

Comments

Couldn't the FOSS community

[lawpoop]

Make some kind of open-source, secure voting system with an auditable paper trail? AFAICT, such a system would need:

• A private, confidential paper receipt, for each vote, that has:
  • a voter-legible ballot that the voter verifies before leaving the vote,
  • a bar-code computer scannable version of the vote, and
  • some kind of code or a non-serial 'serial' number that will indicate any missing paper receipt, or blocks of paper reciepts. We don't want a true serial number so that the vote remains secret and no one can tell who voted for whom by the serial number. Perhaps hashes of hashes?
• A secure, electronic, computer version of this receipt that has some kind of data integrity -- not just a tally of bits, but some binary sequence that has some kind of verifiable, tamper-evident integrity. Perhaps this digital ballot would have a hash stored in a seperate log.

This is just a preliminary brainstorm. Perhaps encoded into each vote's serial number would be a running tally? That would be one method of tamper-evidence -- by going through the votes, we should be able to tell where and when exactly the fraud happened. The tally should be consistent all the way through, and by the time the polls are closed, we have tallies for each booth.

Why not dual-count?

[LinuxDon]

Isn't the most safe option to have 3 separate company's develop -one- machine?

- One company develops the casing and only uses old fashioned electronic push buttons.
- The other two other company's each develop a counter module which are both connected to the same buttons.

This way, the final results should match.
If they do not match, the device is broken, or one of the two company's are attempting fraud.
By keeping the push button system simple, the connections to the counter modules can easily be veryfied by looking at them.

If the whole thing would be sealed and shielded by a glass plate and the wires would be clearly marked, everyone could in theory check the correctness of the machine.

This way, for fraud to be commited, the three company's would have to work together which is more unlikely.
Also, it is possible to prevent the company's from getting in touch with eachother.

A very important point here is: Keep it stupid simple.

You didn't prove what you thought you did.

[Ungrounded Lightning]

We've had the Diebold machines in Georgia since 2002 ... and ... Georgia has gone from one of the worst states ranked around 49 regarding voter fraud and miscounted ballots (with 94,000 undervotes in the 2000 election) to one of the best.

Since there is no way to check that the Diebold machines are counting corretly - or even that they're not making up votes on the fly to be a close match to the number of voters using them - all you've proven is that now that youv'e switched to Diebold machines you no longer can FIND fraud.