Law Prof Argues Cell Location Records Shouldn't Need Warrants Because Cell Phones Have Encryption

from the towards-a-more-government-friendly-reading-of-the-4th-Amendment dept

As the Supreme Court readies itself for an important Fourth Amendment decision, the supporting briefs are beginning to trickle in. At stake is a potential redefining of the expectation of privacy under the Fourth Amendment, something that was diminished by the 1979 Supreme Court decision that created the so-called Third Party Doctrine.

In Carpenter's case, the third party records in question are something gathered by all cell phone companies: location data. The government used months of cell site location data to retrace Carpenter's movements, all without a warrant. This warrantless access turns cell phones into proxy tracking devices for the government. The government is perfectly fine with this turn of events and is asking the Supreme Court to uphold the lower court's decision.

A brief [PDF] siding with the government has been submitted by George Washington University law prof (and Volokh Conspirator) Orin Kerr. In it, Kerr makes some strange arguments.

The least weird argument is Kerr's assertion cell site location records shouldn't be covered by the Fourth Amendment because they are the equivalent of "observation in a public space." This is undoubtedly true, but it does allow the government to perform these "observations" without actually having to use its own eyeballs. Instead of tracking someone's movement through direct, in-person surveillance, the government can serve a subpoena to phone companies and use constantly-collected data to perform retrospective tracking.

Kerr goes on to serve up an analogy to buttress his assertion the Fourth Amendment should provide no protection for ostensibly "public" activities. And that's where the arguments start going off the rails.

[I]magine a world without communications networks. If Alice wants to communicate with Bob, Alice has to leave her home and travel to Bob’s house. If the police suspect that Alice and Bob are conspirators planning a crime, and they assign an officer to watch Alice’s whereabouts, the police can collect only some information without triggering the Fourth Amendment. The police cannot learn the contents of what Alice and Bob said inside Bob’s home without a warrant. On the other hand, the police can observe Alice and see what she did in public – when she left home, where she traveled, when she arrived at Bob’s house, and where they both live – without triggering the Fourth Amendment.

Next imagine that Alice calls Bob on her cell phone instead of meeting him in person. Alice no longer has to travel to meet Bob. The cell phone network delivers the call from Alice to Bob, making a remote transfer that eliminates the need for a public trip. But, critically, the same information exists. What was previously the contents of the conversation in Bob’s house is now the contents of the phone call between Alice and Bob. And what was previously Alice’s publicly observable trip from her house to Bob’s house is now a record that the phone provider generated and may keep about when the call was made, to and from what numbers, and what cell towers were used to deliver it.

What's glossed over in this analogy is the existence of landlines. This middle step is instructive and its absence from Kerr's brief seems almost disingenuous. For years, criminal collaborators used landlines and payphones to converse. This is what the Third Party Doctrine is predicated on: phone records. The Supreme Court's 1979 decision forced companies to comply with (and provide technical assistance for) pen register/trap-and-trace orders. These captured numbers dialed and length of conversations. The only location of interest was already known: the residence/phone booth containing the targeted phone. If law enforcement wanted information on suspects' movements, they still needed to deploy some form of additional surveillance.

Kerr is arguing law enforcement should have access to people's public movements without having to do the actual legwork. And he starts this argument by ignoring the fact law enforcement has -- for years! -- been unable to do anything more than collect phone records sans location data. But now phone records also contain information about people's movements, and Kerr believes they should be inseparable and easily-accessible. This assertion is made despite Kerr attempting to draw a straight line from the good old days of walking from house to house directly through CSLI and email header info.

To maintain the balance of the Fourth Amendment, courts should treat the same information in the same way in both the physical and network contexts. The contents of phone calls should be protected, as they are the telephone equivalent of protected inside space. This means, in the Internet context, that the contents of e-mails, text messages, and files that users place in cloud storage should receive full Fourth Amendment protection. On the other hand, non-content records generated by network providers – the business records they generate about how they delivered the communications – should not be protected because they are the network equivalent of the publicly observable trip that is outside such protection in the physical world.

Going on from there, Kerr says the court -- along with legislators -- must maintain an "equilibrium" between expectations of privacy and lawful access. But in this case, the equilibrium must shift towards the government. Why? Because criminals use cell phones.

The ways that cell phones can facilitate crime and avoid detection counsels against creating new Fourth Amendment protections for cell phone records. Obviously, most people don’t use their phones to commit crimes. But most people don’t have their records collected by court order under the Stored Communications Act, either. The key point is that the effect of cell phone technology on the “often competitive enterprise of ferreting out crime,” Johnson v. United States, 333 U.S. 10, 14 (1948), operates as a two-way street. The ability of cell phone companies to deliver communications quickly and silently over any distance cuts both ways. It can lead to records about the delivery that helps the police, and it can aid in the commission of crime that helps wrongdoers. Both should be considered.

He's right that both should be considered. But his argument doesn't suggest both should be considered. Kerr believes the government's view should be given priority -- a view that allows for no narrowing of the Third Party Doctrine. No one should be granted a higher expectation of privacy because criminals use cell phones. That's basically the argument. And it leads directly to this argument, which isn't any better. The Third Party Doctrine should remain intact because people -- including criminals -- use encryption.

It is too early to tell how far encryption will interfere with government investigative powers. But because users generally can’t encrypt non-content records such as historical cell-site records, the collection of such records may take on a more important role in future surveillance practices. The Court should be reluctant to introduce new constitutional protections for non-content records when the existing constitutional framework for access to contents may be impeded by new encryption technology.

I'm honestly unsure what to make of that argument, which seems to imply the Court should only view the Fourth Amendment as an avenue for law enforcement access, rather than its true purpose: protecting citizens from their government. Kerr talks about maintaining a balance, but posits that relevant technological advancements should work for the government, rather than against it. If people can have encryption and little metal-and-glass rectangles that allow them to hold private, long distance conversations, then the government should have uninterrupted, warrantless access to anything the government deems to be a "third party records."

This isn't the way to maintain balance. The Fourth Amendment isn't the government's enabler. And it never has been. It was written to curb government overreach and abuse. The government has been dealing with unobservable conversations for decades now. That it can now track people without ever leaving the office may be handy, but it doesn't necessarily follow it should always be able to do this without a warrant. The Supreme Court should take a close look at the implications of allowing the status quo to remain in place. Thousands of electronic devices generate millions of third party records every day, all less than a warrant away. Ignoring these implications in favor a simplistic rehashing of a forty-year-old decision is only going to cause further difficulties down the road.

Filed Under: 4th amendment, cell site location info, csli, encryption, location info, orin kerr, scotus, third party doctrine

The Indictment Against Malware Researcher Marcus Hutchines Is Really Weird

from the why-is-that-illegal? dept

So, yesterday, we wrote a quick post about recently-famous malware research Marcus Hutchins (famous for accidentally stopping the WannaCry attack) being detained by the FBI as he left Defcon. An hour or so later, we updated it with the details of the indictment which had been released. That had my quick response, which noted that the "evidence" didn't seem very strong. It just claims (without anything else) that Hutchins wrote the Kronos malware, and most of the indictment and most of the activity focuses on a second defendant (whose name is redacted) who apparently was out selling the malware. I was planning to write up a more thorough look at the indictment and its problems today, but last night, Orin Kerr beat me to it, and he (famed lawyer, law professor and former assistant US attorney) has a bit more expertise in the subject, so let's work off of his analysis.

The crux of the indictment is that Hutchins and the unnamed "co-conspirator" worked together to create and sell malware, leading Kerr to ask the fairly obvious question:

This raises an interesting legal question: Is it a crime to create and sell malware?

After all, as many others pointed out, there are lots of folks out there who build and sell malware of one kind or another -- and, indeed, the US government is often a large purchaser of malware sold by others. Kerr's initial gut reaction was more or less the same as mine: that the actual amount of evidence in the indictment is pretty minimal, though obviously they may have a lot more that just hasn't been shared yet (or they may turn up more).

Do the charges hold up? Just based on a first look at the case, my sense is that the government’s theory of the case is fairly aggressive. It will lead to some significant legal challenges. It’s hard to say, at this point, how those challenges will play out. The indictment is pretty bare bones, and we don’t have all the facts or even what the government thinks are the facts. So while we can’t say that this indictment is clearly an overreach, we can say that the government is pushing the envelope in some ways and may or may not have the facts it needs to make its case. As always, we’ll have to stay tuned.

From there, Kerr digs into each of the charges. The first is "conspiracy." This one struck my layman's mind as somewhat odd. Two people working together does not a conspiracy make. Kerr similarly calls it "odd" and notes that for this charge to work, the government has to argue that selling malware is the same as using malware to damage a computer. And that seems... difficult. Kerr points out that there are two conditions that must be met for this to work:

First, the government must prove that Hutchins and X had an intent to damage a computer. That is, the goal of their conspiracy must have been to impair the availability or integrity of a program or data. Maybe there are facts that support that, but at the very least they don’t appear in the indictment. The indictment makes it seem that the purpose of selling the malware was to, well, sell malware. It’s not obvious that Hutchins and X cared what the buyer did with the malware after so long as they paid. If Hutchins and X didn’t care what the buyer did with the malware, it’s hard to see how they could have a purpose to impair the availability or integrity of a computer.

Second, the government must prove that the agreement was to cause the result of damaging a computer. In an ordinary 1030(a)(5)(A) case, causation is easy. The person sends the malware and the malware damages the machine. Here, though, the government’s theory adds an intermediary: The theory seems to be that Hutchins and X conspired, and the goal of their collective activity was to cause damage, even though the actual act of damaging a computer (if it happened) was to be caused directly by the buyer using the malware rather than by Hutchins and X.

That second point is especially interesting to me. We've seen more and more attempts to charge "intermediaries" with crimes based on actions of third party users of their tools (the Megaupload case being one big example). And that seems like a very dangerous path to go down. One of the reasons why we talk about "intermediary liability protections" on Techdirt so much is that they're so important on a basic "blame the person who actually did the wrong" spectrum. It's not the intermediary, it's the user. Go after the user, even if that's more difficult. Here, the DOJ seems to be going after the intermediary. Because.

The next three charges are all similar, and I didn't quite get them at first, but Kerr explains. They're making use of 18 U.S.C 2512 which Kerr describes as, "a rarely-used law that criminalizes making, selling, or advertising for sale illegal wiretapping devices." Yes, wiretapping devices. Here, Kerr focuses on the question of whether or not a piece of malware software is a "device" under the law, and argues that may be difficult as well.

In Potter v. Havlice, 2008 WL 2556723 (S.D. Ohio 2008), the plaintiff sued the defendant under Section 2512 for making and selling “Activity Monitor,” which was billed as “an ideal spy software package to ensure you have the control you need over your child or spouse activity when they are online.” After rejecting Section 2512 liability because there is no civil cause of action under the statute, the court added an alternative holding that “Activity Monitor is not a device as contemplated by Section 2512.”

Section 2512 makes the manufacture and/or trafficking of “any electronic, mechanical, or other device” illegal. The phrase “electronic, mechanical, or other device” is defined in 18 U.S.C. § 2510(5) to generally mean “any device or apparatus which can be used to intercept a wire, oral, or electronic communication….” Clearly, Activity Monitor alone cannot be used to intercept communications. It must be installed in a device, such as a computer, to be able to do so.

Also, the definition of the word “device” does not encompass software such as Activity Monitor. Merriam Webster Dictionary defines “device” as “a piece of equipment or a mechanism designed to serve a special purpose or perform a special function.” Activity Monitor alone is not a piece of equipmentor a mechanism.

So... that's going to make this interesting. Of course, then there's the further question of whether or not the malware itself is really intercepting communications. Either way, this feels like a way to try to twist a law targeting older technology to pretend that it applies to a very different kind of technology. I know this happens semi-frequently, but it always troubles me. You get bad results this way, because the technology that was originally being regulated, and what it's now being used against, are very different, and should be treated differently. But when you try to shove something like malware into laws created to stop wiretapping devices... you end up with bad results, where rulings can be made about something being "bad" without realizing the wider reverberations it may have.

And, finally, there's a CFAA claim, because if there's a criminal case that could be summarized as "behaving badly on a computer" you have to expect an eventual CFAA claim.

This count raises the same challenges as count one. The theory seems to be that that selling a copy of malware is akin to using the malware to damage a computer. But to get there, they need to show that Hutchins and X had the intent to impair the availability or integrity of information on a computer and not just intent to distribute the malware to a paying customer. The government also needs to prove that their act of distributing the malware was the proximate cause of the resulting damage even though a third party’s intentional act of sending the malware was required for that to happen.

Again... this seems quite difficult to actually show, though perhaps there's more evidence that the DOJ hasn't yet revealed.

In the meantime, others are insisting that the DOJ has the wrong guy. A friend and colleague of Hutchins, Kevin Beaumont, insisted that the DOJ is simply wrong, and that Marcus has more or less dedicated his life to fighting malware, not creating it:

On top of that, the BBC spotted the fact that Marcus asked on Twitter if anyone had a sample of Kronos after it first was discovered:

Now, of course, that alone is not evidence of much. After all, if he really had created it, why not tweet something like that to make sure people think he hadn't? But, still, it is worth pointing out, along with multiple other folks saying that they simply don't believe Hutchins would have been behind the malware, let alone the broader legal question of whether or not making and selling malware is even illegal in the first place.

Filed Under: cfaa, conspiracy, doj, indictment, kronos, malware, malwaretech, marcus hutchins, orin kerr, selling malware

Encryption Workarounds Paper Shows Why 'Going Dark' Is Not A Problem, And In Fact Is As Old As Humanity Itself

from the you-don't-know-what-I-know dept

It was October 2014 when FBI Director James Comey made his famous claim that things were "going dark" in the world of law enforcement because of the increasing use of encryption. Since then, Techdirt has had dozens of posts on the topic, many of them reporting on further dire warnings that the very fabric of civilization was under threat thanks to what was claimed to be a frightening new ability to keep things secret. Many others pointed out that the resulting calls for backdoors to encryption systems were a stunningly foolish idea that only people unable to understand the underlying technology could make.

One Techdirt post on the topic mentioned a great paper with the title "Keys Under Doormats: Mandating insecurity by requiring government access to all data and communications," which ran through all the problems with the backdoor idea. It was written by many of the top experts in this field, including Bruce Schneier. He's just published another paper, co-authored with Orin Kerr, who is a professor at George Washington University Law School, which looks at the other side of things -- how to circumvent encryption:

The widespread use of encryption has triggered a new step in many criminal investigations: the encryption workaround. We define an encryption workaround as any lawful government effort to reveal an unencrypted version of a target's data that has been concealed by encryption. This essay provides an overview of encryption workaround.

The various possibilities are largely self-explanatory:

We classify six kinds of workarounds: find the key, guess the key, compel the key, exploit a flaw in the encryption software, access plaintext while the device is in use, and locate another plaintext copy. For each approach, we consider the practical, technological, and legal hurdles raised by its use.

What's interesting is not so much what the workarounds are, as is the fact that there are a number of them, and that they can all work in the right circumstances. This gives the lie to the idea that we are entering a terrible new era where things are "going dark," and it is simply impossible to obtain important information. But as the authors point out:

there is no magic way for the government to get around encryption. The nature of the problem is one of probabilities rather than certainty. Different approaches will work more or less often in different kinds of cases.

Schneier and Kerr go on to draw an analogy:

When the police have a suspect and want a confession, the law gives the police a set of tools they may use in an effort to persuade the suspect to confess. None of the interrogation methods work every time. In some cases, no matter what the government does, suspects will confess. In other cases, no matter what the government does, suspects will assert their rights and refuse to speak. The government must work with the inherently probabilistic nature of obtaining confessions. Similarly, the government must work with the inherently probabilistic nature of encryption workarounds.

That analogy reveals something profound: that the supposedly new problem of "going dark" -- of not being able to find out information -- has existed as long as humans have been around. After all, there is no way -- yet, at least -- of accessing information held in a person's mind unless some kind of interrogation technique is used to extract it. And as the analogy shows us, that is exactly like needing to find some encryption workaround when information is held on a digital device. It may be possible, or it may not; but the only difference between the problems faced by those demanding answers thousands of years ago and today is that some of the required information may be held external to the mind in an encrypted digital form. Asking for guaranteed backdoors to that digital data is as unreasonable as demanding a foolproof method to extract information from any person's mind. We accept that it may not be possible to do the latter, so why not accept the former may not be feasible either?

Follow me @glynmoody on Twitter or identi.ca, and +glynmoody on Google+

Filed Under: bruce schneier, encryption, encryption workarounds, going dark, orin kerr

Law Enforcement Freaks Out Over Apple & Google's Decision To Encrypt Phone Info By Default

from the we're-all-gonna-die dept

Last week, we noted that it was good news to see both Apple and Google highlight plans to encrypt certain phone information by default on new versions of their mobile operating systems, making that information no longer obtainable by those companies and, by extension, governments and law enforcement showing up with warrants and court orders. Having giant tech companies competing on how well they protect your privacy? That's new... and awesome. Except, of course, if you're law enforcement. In those cases, these announcements are apparently cause for a general freakout about how we're all going to die. From the Wall Street Journal:

One Justice Department official said that if the new systems work as advertised, they will make it harder, if not impossible, to solve some cases. Another said the companies have promised customers "the equivalent of a house that can't be searched, or a car trunk that could never be opened.''

Andrew Weissmann, a former Federal Bureau of Investigation general counsel, called Apple's announcement outrageous, because even a judge's decision that there is probable cause to suspect a crime has been committed won't get Apple to help retrieve potential evidence. Apple is "announcing to criminals, 'use this,' " he said. "You could have people who are defrauded, threatened, or even at the extreme, terrorists using it.''

The level of privacy described by Apple and Google is "wonderful until it's your kid who is kidnapped and being abused, and because of the technology, we can't get to them,'' said Ronald Hosko, who left the FBI earlier this year as the head of its criminal-investigations division. "Who's going to get lost because of this, and we're not going to crack the case?"

That Hosko guy apparently gets around. Here he is freaking out in the Washington Post as well:

Ronald T. Hosko, the former head of the FBI’s criminal investigative division, called the move by Apple “problematic,” saying it will contribute to the steady decrease of law enforcement’s ability to collect key evidence — to solve crimes and prevent them. The agency long has publicly worried about the “going dark” problem, in which the rising use of encryption across a range of services has undermined government’s ability to conduct surveillance, even when it is legally authorized.

“Our ability to act on data that does exist . . . is critical to our success,” Hosko said. He suggested that it would take a major event, such as a terrorist attack, to cause the pendulum to swing back toward giving authorities access to a broad range of digital information.

Think of the children! And the children killed by terrorists! And just be afraid! Of course, this is the usual refrain any time there's more privacy added to products, or when laws are changed to better protect privacy. And it's almost always bogus. I'm reminded of all the fretting and worries by law enforcement types about how "free WiFi" and Tor would mean that criminals could get away with all sorts of stuff. Except, as we've seen, good old fashioned police/detective work can still let them track down criminals. The information on the phone is not the only evidence, and criminals almost always leave other trails of information.

No one has any proactive obligation to make life easier for law enforcement.

Orin Kerr, who regularly writes on privacy, technology and "cybercrime" issues, announced that he was troubled by this move, though he later downgraded his concerns to "more information needed." His initial argument was that since the only thing these moves appeared to do was keep out law enforcement, he couldn't see how it was helpful:

If I understand how it works, the only time the new design matters is when the government has a search warrant, signed by a judge, based on a finding of probable cause. Under the old operating system, Apple could execute a lawful warrant and give law enforcement the data on the phone. Under the new operating system, that warrant is a nullity. It’s just a nice piece of paper with a judge’s signature. Because Apple demands a warrant to decrypt a phone when it is capable of doing so, the only time Apple’s inability to do that makes a difference is when the government has a valid warrant. The policy switch doesn’t stop hackers, trespassers, or rogue agents. It only stops lawful investigations with lawful warrants.

Apple’s design change one it is legally authorized to make, to be clear. Apple can’t intentionally obstruct justice in a specific case, but it is generally up to Apple to design its operating system as it pleases. So it’s lawful on Apple’s part. But here’s the question to consider: How is the public interest served by a policy that only thwarts lawful search warrants?

His "downgraded" concern comes after many people pointed out that by leaving backdoors in its technology, Apple (and others) are also leaving open security vulnerabilities for others to exploit. He says he was under the impression that the backdoors required physical access to the phones in question, but if there were remote capabilities, perhaps Apple's move is more reasonable.

Perhaps the best response (which covers everything I was going to say before I spotted this) comes from Mark Draughn, who details "the dangerous thinking" by those like Kerr who are concerned about this. He covers the issue above about how any vulnerability left by Apple or Google is a vulnerability open to being exploited, but then makes a further (and more important) point: this isn't about them, it's about us and protecting our privacy:

You know what? I don’t give a damn what Apple thinks. Or their general counsel. The data stored on my phone isn’t encrypted because Apple wants it encrypted. It’s encrypted because I want it encrypted. I chose this phone, and I chose to use an operating system that encrypts my data. The reason Apple can’t decrypt my data is because I installed an operating system that doesn’t allow them to.

I’m writing this post on a couple of my computers that run versions of Microsoft Windows. Unsurprisingly, Apple can’t decrypt the data on these computers either. That this operating system software is from Microsoft rather than Apple is beside the point. The fact is that Apple can’t decrypt the data on these computers is because I’ve chosen to use software that doesn’t allow them to. The same would be true if I was posting from my iPhone. That Apple wrote the software doesn’t change my decision to encrypt.

Furthermore, he notes that nothing Apple and Google are doing now on phones is any different than tons of software for desktop/laptop computers:

I’ve been using the encryption features in Microsoft Windows for years, and Microsoft makes it very clear that if I lose the pass code for my data, not even Microsoft can recover it. I created the encryption key, which is only stored on my computer, and I created the password that protects the key, which is only stored in my brain. Anyone that needs data on my computer has to go through me. (Actually, the practical implementation of this system has a few cracks, so it’s not quite that secure, but I don’t think that affects my argument. Neither does the possibility that the NSA has secretly compromised the algorithm.)

Microsoft is not the only player in Windows encryption. Symantec offers various encryption products, and there are off-brand tools like DiskCryptor and TrueCrypt (if it ever really comes back to life). You could also switch to Linux, which has several distributions that include whole-disk encryption. You can also find software to encrypt individual documents and databases.

In short, he points out, the choice of encrypting our data is ours to make. Apple or Google offering us yet another set of tools to do that sort of encryption is them offering a service that many users value. And shouldn't that be the primary reason why they're doing stuff, rather than benefiting the desires of FUD-spewing law enforcement folks?

Filed Under: 4th amendment, encryption, law enforcement, orin kerr, phones, privacy, security
Companies: apple, google

Lavabit Case Shows Why We Need Tech Literate Judges

from the a-failure-of-knowledge dept

While there's plenty of attention being paid to Lavabit's temporary re-opening for the sake of letting people export their accounts, a much more interesting issue is the recent development in the legal case. Lavabit has filed its latest brief, and there are some interesting discussions about the details of the case. From my reading, Lavabit makes a very strong argument that the government has no right to demand the production of Lavabit's private SSL keys, as it's an overreach way beyond what traditional wiretapping laws allow. Lawyer Orin Kerr's analysis argues that Lavabit's case is weak, mainly arguing that the federal government can subpoena whatever the hell they want, and just because it conflicts with your business model: too bad. Lavabit argues that complying with the government's order is oppressive because it would effectively mean it would be committing fraud on all its customers:

[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.

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.

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.

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.

Filed Under: doj, ed felten, email, ladar leveson, orin kerr, scott greenfield, secure email, subpoenas
Companies: lavabit

Who Should Be On The 'Independent' NSA Surveillance Review Board?

from the thinking-through-the-list dept

We've been pointing out that President Obama's "Surveillance Review Board" -- which is being put together to provide an "outside" review of the NSA's surveillance efforts -- doesn't inspire much confidence. All are barely "outsiders", having worked closely with President Obama in the past. There are no technologists on the panel at all. Beyond Peter Swire, it's not clear that any have that much experience or focus in civil liberties/4th Amendment issues. Oh, and it all "reports" to James Clapper, the Director of National Intelligence who runs these programs, and a confessed liar to Congress about them. It's the kind of board that you put together when you want to pretend there's been an outside review.

Given all of that, Gordon Mohr suggested via Twitter that we put together our own "shadow" board, comprised of members who we'd love to see on the board instead of who they currently have. Who would we want reviewing these things if there were to be a real "outside" and "independent" review of the NSA's surveillance efforts? This is not an easy list to put together, because there are so many folks who would be good. But I'll suggest a few here and would like to hear from others in the comments. We'll start with the first five (to match the same number on the official board), followed by some potential alternates.

1. Ed Felten. This one is a no-brainer. The Princeton computer science and tech policy professor, and the director (and, I believe, founder) of Princeton's Center for Information Technology Policy (CITP). He has his feet firmly planted in both worlds, and a long history of doing security and privacy research (as well as a long history of not backing away from controversy to stand behind the results of his research). While he recently did a brief stint as the FTC's first CTO, he's still very much an outsider. Honestly, there's no reason he shouldn't be leading this review team.
2. Bruce Schneier. Well-known and outspoken security guru. I've heard some folks in the security world grumble over the fact that he gets so much facetime when it comes to security issues compared to many other qualified security researchers, but he has a way of cutting through the crap and getting to the point in an easy and understandable way that few others possess. Furthermore, his special power is not being fooled by "security theater," but focusing in on not just what the actual risks are, but the actual probabilities associated with those risks. The least likely person to be subdued by "but everyone will die!" rhetoric that comes from defenders of these programs.
3. Chris Soghoian. Another well-known security and privacy researcher, currently working for the ACLU (though also did a stint at the FTC). He's been focused on corporate "back doors" to government surveillance for ages.
4. Dave Farber. Another engineering/CS professor with a very long and distinguished resume -- but who also has been deeply involved in a variety of public policy issues over the years, and tends to take a very careful, thoughtful approach to analyzing things. Over a decade ago, he was the CTO for the FCC for a brief period.
5. Orin Kerr. Straddles the worlds of technology, law and law enforcement quite effectively. While the former prosecutor sometimes is a bit too quick to justify law enforcement actions, he does seem to approach these types of questions with a very careful and critical eye.

At least that would be my initial top five list -- all of whom have technology experience, some legal experience, and while every one has had some connections to government, they are generally considered "outsiders", while still clearly having the credibility to be taken seriously by the federal government. There are, of course, plenty of others who I think would be good as well -- including folks like Julian Sanchez, Marcy Wheeler, Jim Harper, Jennifer Granick and others who have spent a lot of time digging deep into NSA claims and related privacy issues -- but they have less of the technology background necessary. There are plenty of other good names out there, so feel free to suggest some in the comments.

Filed Under: bruce schneier, chris soghoian, dave farber, ed felten, nsa, nsa surveillance, orin kerr, review board

Why The DOJ's Legal Defense That 'All' Info May Be 'Relevant' For The NSA Is Legally Hogwash

from the just-not-true dept

Last week, we noted the ridiculousness of the DOJ's attempted defense of the government's use of Section 215 of the Patriot Act to scoop up all phone records on every call, claiming a very twisted definition of "relevant" among other things. A variety of folks have been picking apart some of the claims in that analysis, and Orin Kerr has detailed some specific problems with the case law that the DOJ uses to back up its definition of "relevance." If this were an actual court document, it would be easy to counter, since it seems clear that the DOJ didn't use the most appropriate citations, twisted the citations it did use to mean more than they really do, and (most importantly) ignored much more relevant (and I mean that in the English sense, rather than the DOJ sense) citations and precedents. Kerr points out that the DOJ played down that the main precedents they use have specific conditions concerning necessity and context, but further notes how other cases appear to be much more germane to the topic:

A case that comes to mind is In re Grand Jury Subpoena Duces Tecum Dated Nov. 15, 1993, 846 F.Supp. 11 (S.D.N.Y. 1994) (Mukasey, J.). In that case, then-Judge (later Attorney General) Michael Mukasey quashed a grand jury subpoena that sought all computers and all electronic storage devices possessed by the target corporation. The subpoena failed Rule 17′s “relevance” standard because the computers contained so much irrelevant material intermingled with the data that was relevant to the investigation:

Government counsel have conceded on behalf of the grand jury that the subpoena demands irrelevant documents. Moreover, the government has acknowledged that a “key word” search of the information stored on the devices would reveal “which of the documents are likely to be relevant to the grand jury’s investigation.” Id. at 3. It follows that a subpoena demanding documents containing specified key words would identify relevant documents without requiring the production of irrelevant documents. To the extent the grand jury has reason to suspect that subpoenaed documents are being withheld, a court-appointed expert could search the hard drives and floppy disks.

“[B]ecause the subpoena at issue unnecessarily demands documents that are irrelevant to the grand jury inquiry,” Mukasey concluded, “it is unreasonably broad under Federal Rule of Criminal Procedure 17(c).”

Mukasey’s decision seems to cut against the Administration’s position. It blocks a subpoena for all the electronically-stored information because lots of the information to be obtained is not relevant even thought some of it is. Plus, the opinion is written by a recent Attorney General of the United States, which should give it extra prominence. And it’s a lot more significant a precedent than are Fourth Amendment cases involving warrants to search computers: Warrants do not apply the relevance doctrine while subpoenas do, so it seems odd to discuss the cases about computer warrants but not the cases on computer subpoenas. But the “white paper” doesn’t mention this case, so we don’t know if the Administration has a response to it– or if the FISC was ever alerted to it.

I'd suggest reading the entire thing, because it also delves deeply into why the cases that the government does cite don't really support its claims.

But, the real issue here is that this is the kind of thing you get when you have a secret process with a secret court and secret rulings -- with no adversarial hearings at all. It allows one side -- in this case the DOJ -- to make incredibly weak arguments based on distorting existing case law and completely ignoring much more relevant case law, and then getting away with it because no one's there to point out how misleading the defense is. No wonder the DOJ has fought for so many years to avoid having to reveal the legal interpretation of Section 215 that explains the secret interpretation for why 215 allows the scooping up of all phone records. Someone must realize that it would immediately be shown to be lacking. But, when you know that no one is seriously reviewing these things, and those who have the most interest in actually protecting our civil liberties are kept in the dark, you can get away with incredibly sloppy legal defenses. And it appears that's exactly what the DOJ did here.

And it's just all of our privacy that was trampled on in the process.

Filed Under: metadata, nsa, nsa surveillance, orin kerr, patriot act, privacy, relevance, relevant, section 215

Orin Kerr And Members Of The EFF Representing AT&T Hacker 'Weev' Pro Bono During His Appeal

from the and-hopefully,-head-off-further-damaging-CFAA-precedent dept

Andrew "Weev" Auernheimer is appealing his 41 month prison sentence (and its accompanying fine of $73,000). Many members of the security community have expressed concern with this ruling, especially in light of other CFAA cases. Auernheimer's exposure of AT&T's security hole doesn't really seem like the sort of thing that should be punished, at least not with multiple years in jail and a hefty fine. Then there's the unsettling feeling that the US prosecutors pushed hard for a prison sentence because they found Weev unlikable.

Fortunately for Weev (and others who have or will run afoul of the CFAA), Orin Kerr has stepped up to offer pro bono representation in Auernheimer's appeal (along with members of the EFF). Kerr, most recently spotted here going head-to-jackass with Rep. Gohmert over the legality of "destroying" a hacker's computer, has a very thorough post discussing his reasons for joining the fray. Basically, it boils down to this: nearly everything about the government's decision is wrong, which is problematic if this ruling is going to be used as precedent in future CFAA cases.

In the government’s view, visiting the URLs was an unauthorized access of AT&T’s website. But I think that’s wrong. At bottom, the conduct here was visiting a public website. As the Sixth Circuit stated in Pulte Homes, Inc. v. Laborers’ International Union Of North America, 648 F.3d 295 (6th Cir. 2011), everyone is authorized to visit an “unprotected website” that is “open to the public.” The fact that AT&T would not have wanted Spitler to visit those particular URLs doesn’t make visiting the public website and collecting the information a criminal unauthorized access. If you make information available to the public with the hope that only some people would bother to look, it’s not a crime for other people to see what you make available to them.

According to Kerr, undesirable access does not equal unauthorized access. The URLs were publicly available due to AT&T's own carelessness. What this actually looks like is the vindictive pursuit of an individual for publicly embarrassing the company. But it's not all on AT&T. The prosecutors themselves had to do a bit of creative sentencing to arrive at a "suitable" punishment for Weev's "hack."

Unauthorized access is ordinarily a misdemeanor. Why is this crime a felony? Here’s the government’s remarkable theory. All 50 states have state unauthorized access computer crime statutes similar to the federal unauthorized access statute. The government’s theory is that this overlap turns essentially all federal CFAA misdemeanors into federal felonies. They rely on 18 U.S.C. 1030(C)(2)(B)(ii), which states that a misdemeanor unauthorized access becomes a felony when it is “in furtherance of any criminal or tortious act in violation of the Constitution or laws of the United States or of any State.” The government argues that the existence of state unauthorized access crimes transform unauthorized access misdemeanor crimes into felonies: The overlap means that every federal unauthorized access crime is a federal crime “in furtherance of” the analogous state crime.

As Kerr states, this is nothing more than disingenuous double-counting being done for no other reason than to make the charges carry some weight. A misdemeanor results in a slap on the wrist, something that would hardly make AT&T happy. This isn't Kerr's (or the government's) first experience with hacking-related double-counting.

Back in 2011, Sarah Palin's email account was hacked and the Justice Department attempted to charge the hacker under two overlapping laws: "hacking into a computer" and "hacking an email account." This was overturned on appeal by the Fourth Circuit court, stating that the Justice Department's attempt to double dip a single action violated US principles on double jeopardy. This situation is more of the same, only with a convenient overlap of federal and state laws allowing prosecutors to ratchet up the charges from a misdemeanor to a full-blown felony.

In addition to these problems, Kerr also finds some jurisdictional issues at play. Even though none of the principals are located in New Jersey, the charges were brought in that state. The rationale? Some of the email addresses belonged to New Jersey residents. This paper-thin justification for filing charges in a pretty much unrelated state gives the appearance of prosecutorial venue shopping.

The most ridiculous aspect of the case is Kerr's final reason for stepping in: the sentence.

The largest part of Auernheimer’s sentence was due to an alleged $73,000 in loss suffered by AT&T. Under the provisions of the Sentencing Guidelines associated with 18 U.S.C. 1030, sentences are based primarily on the amount of loss caused by the crime. More dollar loss to the victim means more time in prison for the defendant.

AT&T claims it incurred costs of $73,000 due to Auernheimer's actions. But it claimed no loss to its computers, it suffered no downtime and lost no data. The only assertion of loss comes via AT&T's efforts to notify customers of the data breach.

First, AT&T notified its customers by e-mail. That was free, leading to a “cost” so far of zero. But then AT&T decided to follow-up the e-mail notification with paper letter notification, and the postage and paper costs amounted to about $73,000.

That's right. Auernheimer has to repay AT&T for envelopes and stamps with $73,000 of his own money -- and 3-1/2 years of his life. As Kerr points out, AT&T cannot reasonably pin this notification expense on Auernheimer as these costs are not "directly attributable" to the defendant's access of its supposedly off-limits URLs. Furthermore, Kerr says these costs are not "reasonable," considering AT&T's electronic notice to its customers was largely successful. In essence, Weev is doing time because he raided AT&T's petty cash box by proxy. Hopefully, this appeal will overturn this misguided sentence and prevent the CFAA from becoming an even worse law, thanks to the precedent set by this decision.

Filed Under: andrew auernheimer, cfaa, eff, orin kerr, weev

Experts Scratching Their Heads At House Judiciary's Awful CFAA Reform Proposal

from the why-would-they-do-this? dept

On Monday, we broke the news of the House Judiciary Committee circulating a terrible bill that would make the Computer Fraud and Abuse Act (CFAA) much worse, rather than better. It would expand definitions and make it even easier for the Justice Department to go after people for harmless activity. In fact, even the part we originally thought might fix one of the worst parts of the CFAA actually makes it worse.

Now that the bill has been out a few days, various experts on the CFAA are scratching their heads about why the House Judiciary Committee is even bothering with this draft bill. As Orin Kerr notes, this seems to be a basic rehash of the DOJ's attempt 2 years ago to expand the CFAA. He suggests (and we agree) that the Judiciary Committee stop taking DOJ language from 2011 and start dealing in the present, and deal with the very real problems with the CFAA, and not just with a DOJ who wants more power.

They’re looking for feedback, so here is mine: Stop taking DOJ’s language from back in 2011 and packaging it as something new. Based on a quick read, it seems that the amendments for 1030 in the new draft are mostly copied from a bill that Senator Leahy offered (with substantial input from DOJ, as I understand it) back in November 2011. I criticized that language here. The new circulating draft also adopts the sentencing enhancements (minus mandatories) and the proposed 1030a that DOJ advocated in May 2011. I criticized that initial DOJ language here. (There’s also a breach notification provision in the new language, but I haven’t followed that issue closely; I don’t know if that proposal is also based on old language.)

[....] This language is really, really broad. If I read it correctly, the language would make it a felony to lie about your age on an online dating profile if you intended to contact someone online and ask them personal questions. It would make it a felony crime for anyone to violate the TOS on a government website. It would also make it a federal felony crime to violate TOS in the course of committing a very minor state misdemeanor. If there is a genuine argument for federal felony liability in these circumstances, I hope readers will enlighten me: I cannot understand what they are.

Of course, when we brought up similar examples in our original post, people said we were overreacting. Hmm. Meanwhile Paul Rosenzweig, the former Deputy Assistant Secretary for Policy at Homeland Security is similarly stumped by the direction of the reform.

My quick review and reaction to this bill is that it seems to answer most of what the Department of Justice wants with very little for the internet online community in return. Most notably the bill would make violations of the CFAA predicate acts for a RICO criminal charge — what this means is that if you engage in just two instances of violating the CFAA, then you are engaged in a pattern of racketeering, with substantial criminal penalties and .. .since the criminal definitions translate directly to civil liability .. a very significant possibility of a “bet the company” civil suit. Not a move designed to foster innovation, I think.

Hopefully, the House Judiciary Committee goes back to the drawing board on this, and takes a closer look at things like Aaron's Law, which is being developed to cut back on the excesses of the CFAA, rather than expand them.

Filed Under: cfaa, cfaa reform, doj, house judiciary committee, orin kerr, paul rosenzweig

Rep. Gohmert Wants A Law That Allows Victims To Destroy The Computers Of People Who Hacked Them

from the do-these-people-even-listen-to-themselves? dept

Last week, we had talked about some concerns about how various cybersecurity provisions would allow those hit by malicious hackers to "hack back" or, as some call it, engage in an "active defense." There were significant concerns about this, but as Marvin Ammori briefly mentioned in last week's favorites post, Rep. Louis Gohmert seems to not only think hacking back is a good idea, but that it should be explicitly allowed under the CFAA (Computer Fraud and Abuse Act). You can see his explicit statements to this effect below during last week's House Judiciary Committee hearing on the CFAA. It appears he heard a story about someone installing some malware on a hacker's computer to get a photograph of them, and has decided "that's a good thing, that helps you get at the bad guys," without ever thinking of the very, very long list of dangerous consequences of allowing such things: In case the video embed is not working above, I created a short highlight that just covers the ~5 minute exchange involving Gohmert.

Here's the basic transcript. The really crazy part is where Gohmert says he doesn't care as long as the hack back is "destroying that hacker's computer."

Rep. Gohmert: It's my understanding that under 18 USC 1030 that it is a criminal violation of law to do anything that helps take control of another computer, even for a moment. Is that your understanding?

Orin Kerr: It depends exactly what you mean by "taking control." If "taking control" includes gaining access to the computer, assuming a network your not supposed to take control of, then yes, that would clearly be prohibited by the statute.

Rep. Gohmert: For example, my understanding is that there was a recent example where someone had inserted malware on their own computer, such that when their computer was hacked and the data downloaded, it took the malware into the hacker's computer, such that when it was activated, it allowed the person whose computer was hacked to get a picture of the person looking at the screen. So they had the person who did the hacking, and actually did damage to all the data in the computer. Now, some of us would think 'that's terrific, that helps you get at the bad guys.' But my understanding is that since that allowed the hackee to momentarily take over the computer and destroy information in that computer and to see who was using that computer, then actually that person would have been in violation of 18 USC 1030. So I'm wondering if one of the potential helps or solutions for us would be to amend 18 USC 1030 to make an exception such that if the malware or software that allows someone to take over a computer is taking over a hacker's computer, that it's not a violation. Perhaps it would be like for what we do for assaultive offenses, you have a self-defense. If this is a part of a self-defense protection system, then it would be a defense that you violated 1030. Anybody see any problems with helping people by amending our criminal code to allow such exceptions or have any suggestions along these lines?

Orin Kerr: Mr. Gohmert, that's a great question that is very much debated in computer security circles. Because, from what I hear there is a lot of this "hacking back" as they refer to it. But at least under current law, it is mostly illegal to do that.... The real difficulty is in the details. In what circumstances do you allow someone to counterhack, how broadly are they allowed to counterhack, how far can they go? The difficulty, I think, is that once you open that door as a matter of law, it's something that can be difficult to cabin. So I think if there is such an exception, it should be quite a narrow one to avoid it from becoming the sort of exception that swallows the rule.

Rep. Gohmert: Well, I'm not sure that I would care if it destroyed a hacker's computer completely. As long as it was confined to that hacker. Are you saying we need to afford the hacker protection so we don't hurt him too bad?

Orin Kerr: (brief confounded look on his face) Uh... no. The difficulty is that you don't know who the hacker is. So it might be that you think the hacker is one person, but their routing communications... Let's say, you think you're being hacked by a French company, or even a company in the United States...

Rep. Gohmert: Oh and it might be the United States Government! And we don't want to hurt them if they're snooping on our people. Is that...?

Orin Kerr: No.

Rep. Gohmert: I don't understand why you're wanting to be protective of the hacker.

Orin Kerr: The difficulty is first, identifying who is the hacker. You don't know when someone's intruding into your network who's behind it. So all you'll know is that there's an IP address that seems to go back to a specific computer. But you won't know who it is who's behind the attack. That's the difficulty.

First off, kudos to Orin Kerr for keeping a (mostly) straight face through that exchange. There are many amazing things about this particular exchange, but the fact that Rep. Gohmert is one of the people in charge of how the CFAA gets reformed, and doesn't understand these very basic concepts, is immensely troubling. Among the headsmackers in that exchange: the idea that hackers are bad -- and not just partially bad, but apparently obviously and totally bad, like out of a movie. Also: that they're somehow easy to identify and that a freebie on hackbacks wouldn't be abused in amazing ways. Further, as Kerr pretty clearly points out that you can't automatically track back and (without saying so directly, but clearly implying) that hackers likely would shield their identity or fake someone else's identity, Gohmert still doesn't get it and somehow thinks that Kerr is saying we don't want to allow hackbacks on US government snooping (which, again, Gohmert seems to have no problem with). Yikes. Please do not let people like this near laws that have anything to do with computers. To me, this level of misunderstanding is worse than the whole "series of tubes" garbage from a few years back by Senator Stevens.

I'm sorry, but it seems that if you can't understand that there isn't some magic list that says "these hackers are bad, and therefore we should destroy their computers," I don't think you should have any role in making laws around this topic.

Filed Under: cfaa, destroying computers, hackers, hacking, louis gohmert, orin kerr