from the a-failure-of-knowledge dept
[T]o comply with the government’s subpoena would have either required Lavabit to perpetrate a fraud on its customer base or shut down entirely. That is the key point, and the resulting harm goes far beyond a mere inconvenient search for records. Just as requiring a hotel owner to install glass doors on all its hotel rooms would destroy the hotel’s business, Lavabit cannot exist as an honest company if the government is entitled to take this sort of information in secret. Its relationship with its customers and business partners depends on an assurance that it will not secretly enable the government to monitor all of their communications at all times. If a mere grand jury subpoena can be used to get around that (in secret, no less), then no business—anywhere—can credibly offer its customers a secure email service.But Kerr points out that this is a "really weak argument":
This strikes me as a really weak argument. Lavabit is essentially claiming that its anti-government business model trumps the subpoena power. That is, it is arguing that the subpoena is “oppressive” precisely because it would work: It would allow the government to conduct the surveillance it is allowed to conduct under the Pen Register statute.Further, Kerr argues that to accept Lavabit's argument would mean that any company that announces an "ideology or business strategy" that opposes government surveillance could then resist legitimate government subpoenas simply by arguing that they are oppressive and abusive.
I respect Kerr and always look forward to his legal analysis, but I think he's wrong at a variety of levels here, and, tragically the judge in the case seems to have the same confused view of what Lavabit is actually arguing (though, one could argue, that is actually the fault of Lavabit in not making its case clearly). Lawyer Scott Greenfield does a good job explaining why Kerr has mischaracterized Lavabit's defense -- first noting that being pro-privacy is hardly being "anti-government" as Kerr implies. Then pointing out that Lavabit's argument isn't that the government's demand for its private keys was merely oppressive because of its business model, but because it would put Lavabit out of business -- which is not the same thing.
This isn't really a fair characterization of Lavabit's point. Initially, the argument is that revelation of the private key would be the ruination of the business. By exposing every customer to government disclosure, and covert disclosure at that, the government would take a viable business, making money and delivering a service as businesses are allowed to do in America, and destroy it. Poof, company gone. Business gone. Revenue gone. Wham, bam, thank you, Ladar.But there's an even bigger point in here, which I think Kerr misses entirely, and Greenfield skips over: from a technology standpoint, what the government is demanding of Lavabit is absolutely oppressive and abusive. And, for that, it helps to look at Ed Felten's discussion of the case, in which he notes that the judge and other DOJ supporters in this case (including, it would seem, Kerr) are basically arguing that "If court orders are legitimate, why should we allow engineers to design services that protect users against court-ordered access." But Felten points out that requiring "court ordered access" is tantamount to requiring a massive vulnerability to insider attacks:
To see why, consider two companies, which we’ll call Lavabit and Guavabit. At Lavabit, an employee, on receiving a court order, copies user data and gives it to an outside party—in this case, the government. Meanwhile, over at Guavabit, an employee, on receiving a bribe or extortion threat from a drug cartel, copies user data and gives it to an outside party—in this case, the drug cartel.Now, go back to the judge's order or Kerr's analysis, and revisit it with what Felten pointed out, and you realize how far off-base both the Judge and Kerr are in their analyses. Lavabit didn't design its system to be setup the way it was because it was "anti-government," but rather because it wanted to create secure email that protects against a variety of different kinds of attacks, both insider and outsider. That's why it found the government's request so "abusive" and "oppressive." Not because of an ideological disagreement, but rather because of the technological reality that handing over Lavabit's private keys absolutely wrecks any real security of Lavabit's system, which is Lavabit's entire business.
From a purely technological standpoint, these two scenarios are exactly the same: an employee copies user data and gives it to an outside party. Only two things are different: the employee’s motivation, and the destination of the data after it leaves the company. Neither of these differences is visible to the company’s technology—it can’t read the employee’s mind to learn the motivation, and it can’t tell where the data will go once it has been extracted from the company’s system. Technical measures that prevent one access scenario will unavoidably prevent the other one.
Insider attacks are a big problem. You might have read about a recent insider attack against the NSA by Edward Snowden. Similar but less spectacular attacks happen all the time, and Lavabit, or any well-run service that holds user data, has good reason to try to control them.
So, while Kerr and the judge in the case seem to think it's a mere ideological issue, that's simply not true. It's a technological issue, on which Lavabit's entire business was based. If Kerr and the judge are correct, then, as Felten properly notes, it becomes effectively illegal to build a really secure communications system. That seems positively ridiculous, especially in a time when we're told (by the very government agency that wants to do all this spying) that we need better online security to protect against attacks.