Sequoia Threatens Over Voting Machine Evaluation

enodo writes "Voting machine manufacturer Sequoia has sent well-known Princeton professor Ed Felten and his colleague Andrew Appel a letter threatening to sue if New Jersey sends them a machine to evaluate. It's not clear from the letter Sequoia sent whether they intend to sue the professors or the state — presumably that ambiguity was deliberate on Sequoia's part. Put another clipping in your scrapbook of cases of companies invoking 'intellectual property rights' for bogus reasons." Sequoia seems to be claiming that no one can make a "report" regarding their "software" without their permission.

No permission should be needed

I am sure the state of New Jersey can tell Sequoia to accept this investigation or say good-bye to any certification. Sequoia is just making themselves look bad and like they have something to hide.

Re:No permission should be needed

I am sure the state of New Jersey can tell Sequoia to accept this investigation or say good-bye to any certification. Sequoia is just making themselves look bad and like they have something to hide.

I agree however New Jersey probably has already paid for the machines. I can't see Sequoia telling the state they can't test them, have them tested scientifically, if they haven't already been paid for. This case can however be used to show other potential buyers just how the company operates.

Falcon

Re:No permission should be needed

With the company threatening legal action, why even consider going forward with the investigation? Threats usually come from feeling that what you're threatening is dangerous in some way, so it's highly likely there are security problems to be found. Additionally, the state shouldn't even consider using a company that opposes a comprehensive security evaluation.

I Agree with Sequoia on this

and you too.

If their contract declares as such then they are in the right.

However, it should be a requirement at the state level, if not federal, to require this sort of outside verification and study. If a manufacturer does not agree to it then they should be considered for the application. Fair is fair.

Don't want to be hold accountable then don't expect our money.

I am quite sure some other company will step forward if there is money to be made and their intellectual property rights are protected. I am all for testing and certification by outside groups but I also realize that there is investment here and that needs protected first. What must come first is OUR rights, our rights to know that outside experts have certified a solution and future implementations will protect our vote. Surely some company will step up to this for the money. Maybe it will be the kick in the pants for some group already in place to do so.

Re:No permission should be needed

If they do not have enough confidence in their system's security or accuracy to allow it to be tested, then it is not good enough to be used for e-voting. They have just demonstrated that their system can not be trusted.

Sweet.

Yes, that is *exactly* what we want Sequoia to do. Sue a Priceton professor & *security* researcher.

Bullet. Meet foot.

Re:

Actually, threatening Felten with a lawsuit is quite effective. Now actual lawsuit needs to be filed. Since this would be consulting work, not research, he would not be covered by Princeton's Lawyers and will have to fend for himself, which is quite expensive -- so the threat of litigation might be enough to deter him. New Jersey, being a state, cannot be sued without its consent.

Re:Sweet.

But... New Jersey might file a countersuit against Sequoia claiming that they interfering with New Jersey's ability to verify that the machines are suitable for use in actual elections (before the elections are held and found to be invalid). Imagine GM suing a customer because they took their car to a non-GM mechanic to verify that some work had been done properly. I doubt the courts would look kindly on that.

If I were a New Jersey state elections employee, I'd be looking seriously at returning the equipment, informing Sequoia that it was unsuitable for use, and demanding my money back.

Finally, I'm guessing that Sequoia isn't all that interested in staying in the electronic voting machine business. This act will pretty much have everyone who's still considering even using electronic voting equipment making a mental note to cross Sequoia off their short list of vendors.

The ambiguity is a dead giveaway.

It's not clear from the letter Sequoia sent whether they intend to sue the professors or the state -- presumably that ambiguity was deliberate on Sequoia's part.

In other words, this is a scare tactic with nothing to back it up, pure and simple. If Sequoia thought the would have had actual grounds to sue, you can bet that they would have been chillingly specific in their letter.

When people resort to these sort of tactics to attempt to dissuade you, you can be assured you're doing something right.

Re:The ambiguity is a dead giveaway.

In other words, this is a scare tactic with nothing to back it up, pure and simple.

I'm wondering "how thay can threaten their customer?" Who do they think they are, the RIAA and that New Jersey is a mom on food stamps?

But perhaps they can. This cynical old man thinks there's a lot more here than meets the eye - Sequoia may surreptuously funnel cash to the campaigns of some of the high ranking New Jersey goons, er, excuse me, lawmakers/bureaucrats.

But then I'm in Illinois where the last Democrat Governor [wikipedia.org] went to prison, and the last Republican Governor [wikipedia.org] then went to prison. I'm thinking if a Republican wins the next Governor election, Blago [wikipedia.org] will join Ryan in a cell.

Rich powerful people don't play nice. You don't get to be a rich, powerful man by giving a rat's ass about anyone or anything except your money and power.

-mcgrew

(background on Illinois Politics): [wikipedia.org]

Politics in the state, particularly Chicago machine politics, have been famous for highly visible corruption cases, as well as for crusading reformers such as governors Adlai Stevenson (D) and James Thompson (R). In 2006, former Governor George Ryan (R) was convicted of racketeering and bribery. In the late 20th century Congressman Dan Rostenkowski (Dem) was imprisoned for mail fraud; former governor and federal judge Otto Kerner, Jr. (D) was imprisoned for bribery; and State Auditor of Public Accounts (Comptroller) Orville Hodge (R) was imprisoned for embezzlement. In 1912 William Lorimer, the GOP boss of Chicago, was expelled from the U.S. Senate for bribery, and in 1921 Governor Len Small (R) was found to have defrauded the state of a million dollars.[58][27][15]

Check, Meet Balance

A voting apparatus without a clear path for auditing of every system and sub-system is an invitation to corruption. Comparable mechanisms in governance have checks and balances to ensure fidelity.

Why do these shifty porkchops think they ought to be exempt? Because it may make their investors nervous?

This is definitely a situation where the bottom line should be drawn by logic, not by dollars.

Re:

Maybe that's how this happened. Some clueless board member demands something to be done, slamming his first on the table. The task gets assigned to the CEO, who assigns it to the lawyers, who do "something" that will appease the chain of command all the while knowing how useless it is.

Re:Check, Meet Balance

Any voting system should be verifiable by any member of the voting populous. Having a PhD in computer science should not be prerequisite for understanding the voting system. You also shouldn't have to take somebody else's word for it either. Pen and paper hand counted ballots make sense, because anybody can see exactly what's going on, and fully understand the process. If voters don't understand the voting system, then they might as well not even be voting.

Re:Check, Meet Balance

The problem is the lack of even a single impartial ballot counter. Whereas an expertly designed and reviewed machine can be reasonably guaranteed to be bias-free.

Fixing elections is not something that has been enabled by new technology. The problem here is that the technology was supposed to reduce the fraud and inaccuracies, but it turns out it's just as hackable as the old pen & paper or punchcard systems despite the higher cost.

Re:Check, Meet Balance

That's why the ballots can be counted, or viewed by multiple parties, who should all agree on the final counts. Anybody should be allowed to stand around and watch the counting. I'm not saying that no fraud would happen, because it's happened in the past, and it will happen in the future. I'm just saying that it should be obvious to the voting public when fraud is happening. The problem with machines, is that even if they are verified, it's impossible to know what code is running on it when you walk up to it on election day. Think of game consoles. They try to make it so you can only run licensed content. But people always find a way to run homebrew/pirated games. You can verify the machine all you want. There's no guaranteeing that the same machine will actually be used on election day. You could probably even put a completely different machine in front of people on election day, possibly in the same casing, although that's not even necessary, and people wouldn't even know they were using the wrong machine, because each polling district uses different machines.

Re:Check, Meet Balance

No, it's not the same problem. It's easy for people to watch the movements of pieces of paper, make sure they end up where they belong, and count them. It's much harder to do the same with electrons.

The solution for paper ballots is based on four principles: transparency, adversarial conditions, counting everything in a way that, if done right, makes double-entry accounting look like a random number generator, and decentralization.

Transparency means that every step of the process is done in the open, with multiple people watching. Adversarial conditions means that the people watching include representatives (i.e. campaign workers) of all candidates, who are highly motivated to ensure that the others don't cheat. In Scotland, for example, each candidate can even apply their own seal to the ballot boxes in addition to the electoral commission's seals, so they can verify for themselves that boxes haven't been swapped, opened, or lost. A fraudster would have to be able to duplicate the seals of every political party in addition to the electoral commission.

As for counting, every ballot paper must be accounted for. Polling stations start with a known number of blank ballots (verified by all candidates) and they must count the number of ballot papers issued, used, spoiled, and not used, as well as the number of ballots that end up in the ballot box, and if the numbers don't add up right then one can deduce that funny business is going on.

Finally, decentralization is important. With safeguards in place, it may still be possible to cheat in a few locations (although you'd have to get campaign workers from all sides to look the other way), but widespread fraud serious enough to steal a whole election becomes extremely difficult. It is difficult to compromise the process in many locations at once. And even though the central counting facility receives the counts from each polling station and adds them up, it can be made to echo the numbers back and discrepancies can be spotted. A centralized electronic system, though, can be compromised at the center, you don't have to take over every polling station.

Re:Check, Meet Balance

"It is the exact same problem."

Hardly.

Think of the issues in logistical terms. In a paper-ballot system, tampering with a ballot box in such a way as to make any appreciable difference in the vote result would require tools, materials, physical access and a certain amount of time and effort within a relatively small window of opportunity.

With an electronic system, if the state isn't allowed to run simulated elections or do detailed inspection of the voting machine and software, the window of opportunity for tampering is huge, the potential for altering the result is high, and the risk of detection is minimal.

Let's call a spade a spade:

and see that the only reason Sequoia is pissed off is that they either
a. are afraid that there are gaping security holes in their machines
b. KNOW that there are gaping security holes in their machines

all the privacy zealots will no doubt say that my "if you have nothing to hide you have nothing to be afraid of" mentality is misguided, but let's take a step back and see what is on the line here. this is NOT about personal data, this is about objectively evaluating the security of a device that is going to be used in a VERY public fashion. do lamp makers threaten Underwriters Laboratories for wanting to make sure their device works as intended?

Re:Let's call a spade a spade:

I for one say we need to amend the patent and copyright acts to make devices used to voting unpatentable, and exempt from copyright, or barring that, that the certification process requires all rights to be signed over to the government. These machines by their very nature should be open to the most detailed scrutiny imaginable by anyone who feels so inclined. If the companies want to make money on them sell maintenance and manufacturing contracts, but there should be no way to claim trade secrets on anything used for voting.

Re:Let's call a spade a spade:

Someone mod this guy up. I find it unconscionable that it is possible to patent or copyright something that is absolutely critical to the fundamental processes of democracy. Sequoia might as well demand that they be made Emperors of America.

Re:Let's call a spade a spade:

I find it unconscionable that it is possible to patent or copyright something that is absolutely critical to the fundamental processes of democracy.

Also unconscionable is the notion that the underlaying software algorithm would essencially boil down to a very simple statement that looks something like the following:

vote++;

You could argue that there is some finesse involved in getting that data from the machine it was cast at to the central tallying point where it is counted and tabulated, but NOTHING in that process is any more complicated than the Automated Teller Machines which function in a similar way to take data from client nodes and send it up to the hub... so "prior art" in the realm of basic concepts of networking makes those patents unattainable.

Re:Let's call a spade a spade:

Lamp manufacturers pay UL for the privelege of being certified. UL doesn't randomly grab equipment and force a cert upon it. The voting machine manufacturers should be more than interested in having third-party audits of their equipment ... unless they are either crap or corrupt. I suspect the former, but the latter isn't too far behind.

Regardless, I don't see how the manufacturer could impose restrictions on the equipment if it has been sold. Leased? Yeah, that'd come with a use restriction because title never was transferred.

Re:Yes. And why does UL do it?

Neveda Gaming Commission [nv.gov]

I hear they are pretty good a doing hardware/software system audits and design reviews.

Ok, I RTFA, but still...

...have I got this straight?

Their voting machines are paid for by public dollars, used by the majority of the members of the public, to elect public officials, and they claim evaluation of their software cannot occur without their "permission"?

(Even my 9 year old nephew read this and thought it was "dumb")

Re:Ok, I RTFA, but still...

The DMCA cripples reverse-engineering and software analysis. There was another "unified copyright" bill around the same time that forbid the review, or publication of any review, of software without the permission of the company selling it. Never heard if that made it into law. If it did, then yes, Felton's review would be illegal. If it didn't, but the review process would necessarily violate the DMCA, then it is also illegal.

Is this stupid? Yes. Is this hostile to the interests of Americans? Yes. Was it voted for by a Congressman you have re-elected since? That I don't know, obviously, but it's something you may want to check on before this election. If you're not planning onvoting for whoever is currently elected, you might also want to find out the views of the opponents, particularly if the region is tech-savvy enough (or even tech-phobic enough) to be suspicious of voting machines. The candidate you're looking at might enjoy playing around with Sequoia's attitude problem as a (minor, to them) campaign issue.

Re:Ok, I RTFA, but still...

The DMCA was passed in the House by voice vote, and the Senate by Unanimous Concent.

See the "Major Actions" section of this address
http://thomas.loc.gov/cgi-bin/bdquery/z?d105:HR02281:@@@L&summ2=m& [loc.gov]

So, if your congresscritter was in office in 1998, you really have to assume he/she voted for it. I suppose you could troll the attendence logs for the day in question and see if they were absent, but I don't know how much that washes their hands. And potentialy, House reps may have voice voted against it, but it's unlikley.

So, Pretty much any Senator/Rep who's been in office for more than 10 years is responsible.

Re:Ok, I RTFA, but still...

...have I got this straight?

Their voting machines are paid for by public dollars, used by the majority of the members of the public, to elect public officials, and they claim evaluation of their software cannot occur without their "permission"?

Their voting machines are SOLD (or maybe leased) to the government agencies that operate elections and have a contract specifying terms of use. They're claiming the contract forbids the sort of investigation that is proposed.

Now perhaps there are indeed such terms in the contract. In which case the New Jersey secretary of state (or a past one) made an unwise decision. Nevertheless, the state has a duty to insure that the system is not defective. Inspecting its operation, including that of the software if there is any question about its functionality (or even if there isn't, just to check), is obviously a part of that duty, and being inspected is obviously part of what it is to be a voting machine. So such contract terms, if present and interpreted as Sequoia claims, are clearly unconscionable. On that basis the state should be free to ignore the clause.

Alternatively, if the clause were to stand the resulting terms of use would make the machines "unsuitable for the intended use", violating the implied warranty of fitness. So the state could return them for a full refund. B-)

It would be interesting to see what would happen if Sequoia actually sued. If the contract specified interpretation under the laws of New Jersey (or didn't specify jurisdiction) the state might just refuse to be sued. B-) If it specifies another state (or they sue in their own) it would still be a funny show.

As for suing the professor, either he's acting as an agent of the state (in which case they're suing the state) or he's not (in which case he has no contract with them to enforce.) In the latter they'd have to go after him for something like DMCA violations or some part of contract law I'm unaware of.

Of course IANAL - and especially not a contract lawyer. So take the above with a suitable quantity of salt. B-)

History lesson

Mebbe they need a history lesson...

Anyone care to guess when the last armed revolt against government was here in the US and the reasons behind it?

Answer - Battle of Athens, Tenn. 1946. And it was over voting issues...

http://en.wikipedia.org/wiki/Battle_of_Athens [wikipedia.org]

Re:History lesson

The wiki page is pretty short on details.

See http://www.constitution.org/mil/tn/batathen.htm [constitution.org]instead.

"Really"?

Sequoia seems to be claiming that no one can make a "report" regarding their "software" without their permission.

I "beg" to "differ".

"non compliant analysis"?

"non compliant analysis"? whats the betting that the only compliant analysis gives it an A-OK rating.

thats like car salesperson attempting to sell you a car but only if you agree to take his word that it works and he'll sue anybody that you bring in to check the engine. if ever there was a warning bell not to buy their equipment that was it.

whats next? DMCA action against /. for this thread? absolute cobblers.

Update and more details on this

It was Union County, NJ that was planning to send sample machines to Dr. Felten. They were threatened by Sequoia and have backed down from their plan as a result. The whole thing happened because several counties in NJ reported errors with the Sequoia machines in the February primary election. Sequoia reviewed these and just said that it was "user error" and not a problem. The counties understandably didn't find this an acceptable response, and Union County wanted to get an outside opinion. All the details can be found in this NJ Star-Ledger story [nj.com].

In other news...

Four mob families in conjunction with a number of street gangs is suing the US Department of Justice for restriction of trade.

Trivial Workaround

The State of New Jersey just needs to hire Professors Felten and Appel as consultants. Nothing in the so-called "established Sequoia licensing Agreement for use of the voting system" can prevent a State employee from accessing a State machine.

threat of suit

IANAL but I am reasonably confident that Sequoia cannot successfully sue Felten. They may be able to sue New Jersey for breach of contract if in fact they have a contract with New Jersey that forbids such reviews. That may be the case - I believe that the license for MS Windows Server forbids reviews not approved by Microsoft. If there is such a contract, it would be interesting to see if it holds up in court. The "no review" provision is arguably void as being contrary to public policy.

Felten, however, has no contractual relationship with Sequoia and therefore cannot be in breach of contract. Sequoia therefore cannot sue him unless they can come up with another cause of action. Maybe, just maybe, they could sue him for disclosing trade secrets, if New Jersey has really nasty trade secret laws.

Nice to see they are so confident...

in the quality of their product...

Disappointing

This is just disappointing. I had Sequoia pegged as one of the good guys. When we selected them here in Nevada, they willingly submitted to every test, including review by the gaming board (who know a thing or two about fairness and anti-tampering in electronic devices - they have to certify all slot and video gaming machines). Sequoia passed the tests, including having a paper trail. That's not to say they're perfect or infallible, but they sure seemed cooperative and focused in the right direction.

This just doesn't fit with their actions back then. Has there been a change in management?

Re:Speechless.

Just speechless.

That's often the results with certain voting machines.

handy though

If this gets thrown out, it will really dent future attempts to use methods like this to shield shoddy products.

If you ask me, Sequoia has been given some very bad legal advice. Didn't anyone stop to think about the public relations nightmare this would cause? Not to mention damage to their business.

Re:handy though

If this gets thrown out? Surely no court could be so stupid to think that a third party's opinions of an available product can be stifled on the grounds of intellectual property laws. I can't think of anything even reasonably close to a valid excuse to allow it to go through. Perhaps they expect that it will come across as libelous (meaning a completely valid and accurate assessment that portrays them in a negative light - which just means that your company/product sucks), but the review would have to be out there first. I don't think a gag order can be issued in this kind of situation, and an NDA violation (assuming they ever held up in court, which is very rarely the case) hardly applies here.

I don't think a PR nightmare really applies. This kind of stuff rarely if ever hits mainstream media, and us geeks of Slashdot aren't really the type to buy into proprietary tools in general, let alone ones used for voting.

Re:handy though

If this gets thrown out? Surely no court could be so stupid to think that a third party's opinions of an available product can be stifled on the grounds of intellectual property laws. I can't think of anything even reasonably close to a valid excuse to allow it to go through.

That's where "intellectual property" becomes such a vague and dubious notion. Consider if this were a medical device. If health risks were involved, could a company forbid the FDA from investigating the device because someone signed a licensing agreement? Or if the FDA has no interest, what about restriction a state board of health from investigating the device?

In the past, less than one hundred years ago, it was extremely difficult to investigate dangerous products or workplaces. The government just didn't feel it was important enough to get involved in. There had to be major media effort to bring these problems to the public eye before there was enough pressure on the government to put pressure on companies. Such as hair products that could cause blindness and disfigurement, sometimes death (the reason that we have animal testing for cosmetics today). Companies didn't care if customer's were hurt, as long as they got their money first; but bad publicity made them at least pretend to care.

Or the use of radium on watch dials; a company where the chemists would use lead shields to deal with radium would suggest to the factory workers that they could lick the brushes used to pain the dials. That took a media campaign to publicize slow painful deaths of young women before the public demanded change. It's interesting here that US Radium threatened to sue to prevent publication of a damning report, since the author had signed a confidentiality agreement (deja vu).

But in this case, the danger is to elections, not someone's health. The question is whether this rises to the same level of concern. In the health cases, people had to die or become disfigured before the government and the public took any action. Do we have to wait for a stolen or hacked election before the people decide that the public's interests in this matter overrides intellection property? ("intellectual", hah, there's nothing so special about the technology here; shoddy hardware, shoddy software, slick marketting)

I don't think a PR nightmare really applies. This kind of stuff rarely if ever hits mainstream media, and us geeks of Slashdot aren't really the type to buy into proprietary tools in general, let alone ones used for voting.

But this needs to hid the mainstream media. Otherwise companies just go on with business as usual. Governments will never take any action unless they detect some sort of public outcry, especially in the modern climate where pro-active governments are frowned upon.

Re:handy though

There is an alternative to all of this. New Jersey can simple remove Sequoia from the list of companies they're considering for voting machines. I mean, if they don't have it already, they can simply pass law stating that all those submitting bids for supplying the state with voting machines must make a machine available for independent review by an organization or institution of the state's choice. If Sequoia doesn't like those terms, they don't have to bid.

It's true, however, that IP claims are getting out of hand when a government and/or institution doing some work for the government is threatened with a lawsuit over testing hardware. These events are only going to get more egregious and ludicrous until Washington and the courts start handing these abusers their proverbial balls on a platter.

Here is Sequoia's response from their website...

3/18/2008
Sequoia Voting Systems supports third party reviews and testing of its election equipment

In response to some recent media reports, Sequoia is has issued a statement found at . Through this statement, we hope to educate individuals on the third party review mechanisms already in place in the election industry as well as our company's business practices regarding third party reviews and testing of its election equipment.

An independent review of a voting system is a complex and interdisciplinary process involving a broad knowledge of election law, public administration and technical matters. Many independent reviews have been successfully conducted within the framework of Sequoia's license rights pursuant to appropriate and mutually agreeable arrangements between Sequoia and governmental agencies charged by law with the authority to conduct such reviews. Sequoia welcomes all such responsibly executed review activities.

Please see the Election Technology Council's "Guidelines for States conducting Top-to-Bottom Reviews" found on the organization's website at for additional information.

- Michelle Shafer, VP of Communications

Re:Here is Sequoia's response from their website..

Sequoia welcomes all such responsibly executed review activities.

Obviously they don't. Anyone claiming that Ed Felten is unable to responsibly execute such a review should have her head examined. More to the point however, it is equally obvious that they are very much aware of Professor Felten's reputation, and would very much rather he didn't execute a responsible review of their equipment.

Hey ... if they've nothing to hide ...

Re:Speechless.

Just speechless.

You are speechless for the right reasons but the majority of the American public will be speechless for another and far more unfortunate reason :(

Re:Speechless.

the majority of the American public will be speechless for another and far more unfortunate reason

So - Let's let them be heard. Let's put the decision on whether or not to use the machines to a vote. It can be a trial run for Sequoia's machines. If Sequoia tells us that we voted in favor of using their machines, we know they're lying and ditch 'em.