Voting machine vendor threatens computer science researcher
TOUCH-SCREEN voting machine vendor Sequoia Voting Systems emailed a vaguely threatening letter to Princeton University computer science professors Edward Felten and Andrew Appel last Friday.
Writing in his bog, Professor Felten noted that the email has been circulating on various mailing lists and has prompted questions about its authenticity. He therefore published a slightly redacted copy of it and said, " Yes, it is genuine."
The email, apparently sent by one Edwin Smith of Sequoia Voting Systems, notes that "certain New Jersey election officials have stated that they plan to send to you one or more Sequoia Advantage voting machines for analysis."
Smith then says that, if the County pursues such an analysis, "it violates their established Sequoia licensing Agreement for use of the voting system."
As though academic researchers at a private University might care about some contract agreement -- that they did not sign, and to which they are not parties.
That's simply ridiculous on its face, but wait, it gets even better. Smith goes on, "Sequoia has also retained counsel to stop any infringement of our intellectual properties, including any non-compliant analysis."
That brings up the question: What might be a "compliant" analysis? Would that be an analysis conducted under the watchful eyes of Sequoia's representatives, precisely according to their instructions? Would a "compliant" analysis be such a tame subservient inspection that it, conveniently enough, totally failed to find any functional or security problems with the voting machines examined, that is, nothing untoward in the view of elections officials or embarrassing to Sequoia?
Then, not content to merely mention lawyers in a vaguely ominous way, Smith says, "We will also take appropriate steps to protect against any publication of Sequoia software, its behavior, reports regarding same or any other infringement of our intellectual property." [emphasis added]
Here, Sequoia seems to be claiming that publishing any description of its voting machine software behaviour, or any report about its functionality, any security vulnerabilities it exhibits, and so on, is tantamount to copyright infringement.
That's so far from what the US Copyright Act actually says that it's laughable.
Sequoia must be really terrified of what independent, objective examination of its Sequoia Advantage electronic touch-screen voting machines might reveal.
Readers' comments appended to Professor Felten's posting of this email are entertainingly scathing, and you can find them all at the link provided below. µ
L'Inq
Prof.
Felten's blog
Share this:
ShareBusiness Intelligence Software
Customer Relationship Management (CRM)
Enterprise Accounting Software
Enterprise Resource Planning (ERP)
Enterprise Systems Management
I would have thought that a company would welcome independent analysis of their system. Any faults found by the prof could then be rectified.
If no faults are found you now have independent verification of how secure your device is.

But no, they call in the lawyers, which to me screams "We know we're going to be torn apart, we know out system is cr*p, and won't stand up to any kind of examination".

For me that's enough grounds to chuck all of the damn things into a landfill!
El Presidente Hugo Rafael Chávez Frías Wins Again! Go figure? These Sequoia (Cherokee peace be upon him) disAdvantage'D Voters have been certified by *the Company?* Any attempts by Princeton Pedagogues to peek beneath the Sequoia's bonnet tantamounts to rape! 
Pursuant to the BBC's publication
(http://news.bbc.co.uk/1/hi/technology/6200005.stm) 

the UK Office of Science and Innovation's Horizon Scanning Centre, which addresses Robo-rights: "the robots... have certain responsibilities such as voting... Conversely, society would also have a duty of care to their new digital citizens".

Careful New Jersey and Ivy-League, Cyberocidic debauchers! After the criminal adjudications, civil suits are bound to follow.
I would bet that they don't want them to find the algorithm that makes McPain win.
First, Princeton is a public university, owned and operated by the state of New Jersey.

Second, as purchasers, the state of New Jersey is absolutely within their rights to have their experts examine any item comtemplated for purchase for public use to ensure that it meets their requirements.

As long as the investigators swear legally to maintain trade secrets, the vendor is in no position to threaten the state with any legal action, which would have to go through NJ courts, by the way.

If I were in charge of the procurement, I would just tell the vendor that without passing muster by our (i.e., the state's) chosen employees (as those professors are, working for Princeton), then NO SALE!

BTW, I think in NY state here, we are going to use optical scanning mark cards, as there is a real issue with using electronic voting machines with no physical record for post-election auditing. Maybe NJ should do the same.
I have to agree this stinks to high heaven . I would hope NJ tells vendor to pound salt .
I agree, any company having made a product in the spirit of helping its customers should certainly welcome any independent analysis that aims to improve said product.
Unless, of course, the "customer" is not the voting public the machine will be used by, and the real "customer" has already validated that said machines will rig the vote as intended. In which case, of course, the University will shortly develop some serious biological complication that will require a relocation to Guantanamo and a long period of "medical" isolation - for their own safety, of course.

Note to the moderators : could you people PLEASE do you job and remove posts (like the previous one) that have NEXT TO NOTHING to do with the subject at hand ?
I thought electronic voting machines were generally required to have such features as public source code and generating a paper trail.
The ONLY SURE WAY to restore integrity / accuracy / honesty to the U.S. election system is to scrap / junk ALL electronic voting machine systems in the U.S. and return to the paper ballot. The public has been duped by the fiasco of the 2000 election, whereby the public, was told that paper ballots had caused all of the trouble (hanging chads). That was only partially true, as Clint Curtis (software engineer hired by Tom Feeney) testified to the Florida State legislature regarding electronic voting machine irregularities whereby the voting machine software can and WAS designed in the code to erase itself once the vote numbers had been manipulated within the voting machine. The key to this type of software manipulation in the electronic voting machine is the voting software's 'device drivers' buried in the Windows Operating System. The code within the voting machine's 'device driver' can stay hidden from normal checks / analysis by less-computer-savvy individuals responsible for that State's voter oversight. The corrupt device driver's code can be written to erase any evidence of itself, thereby leaving no traces for later analysis. You can see detailed discussion of this on YouTube. Device driver manipulation is an old trick that hackers have used since the DOS days (late 1980's) It is time we all educate ourselves and spread the word about electronic voting. 

The source code (operating software) for the various manufacturers of electronic voting machines should be in the public domain - since electronic voting machine apparently serve the public interest - or, are assumed to serve the public interest. The fact is, the manufacturers of these voting machines including: AccuPoll , Diebold, ES&S, Sequoia, & Science Applications International Corporation (SAIC) do not allow public analysis of their source code or operating systems in any manner. WHY??? See this article http://www.nj.com/news/index.ssf/2008/03/voting_machine_maker_threatens.html

And Google: 'voting machine software analysis'

Also Google: 'Clint Curtis'

I am so fed up with all of this I may sit out the November elections. It would be quite interesting to see the national voting number results if millions of us sat it out. I'd bet the numbers would still be 51%-49% as designed into the code.

Brian