I’m pleased to announce that Alex Halderman, a second-year graduate student who works with me, now has a byline here on Freedom to Tinker. Alex works on computer security and infotech policy, and has done interesting research on topics such as compact disc copy protection and privacy-enhancing technology. He plans to attend the Grokster oral arguments tomorrow; I hope he will be able to give us a firsthand account of the arguments and the early-morning linewaiting/campout experience. I have also invited him to post on other topics as he sees fit.
A (True) Story for Grokster Eve
March 28, 2005 by
Recently I met a promising young computer scientist, whose name I will withhold for reasons that will soon be evident. He has developed a very interesting network software system that would be useful for a great many legitimate applications. I was impressed by his system and wondered why I hadn’t heard of it before.
The reason, it turns out, is that he isn’t sure he wants the public to find out about his research. He says this, even though his work would probably be of interest to many people, and could be useful to far more. The problem, he told me, is that if too many people find out what he has done and realize its value, some of them may start using it for illegal purposes. He doesn’t want that kind of trouble, so he is avoiding bringing his work to the attention of the broader public, publishing it in research venues where a small community of experts will see it, but avoiding any further disclosure.
It’s hard to blame him, given the unsettled state of secondary liability law. If some people start using his system illegally, will he be liable? Will he have to redesign his system to try (probably fruitlessly) to make illegal uses impossible? How many redesigns will be necessary? Will he have to face the same uncertainty that Bram Cohen, creator of BitTorrent, faces? He doesn’t want any of that
Coming: Mobile Phone Viruses
March 23, 2005 by
Clive Thompson at Slate has a scary-sounding new piece about cellphone viruses. As phones get smart – as they start running general-purpose operating systems and having complex software interfaces – they will tend to develop the kinds of software bugs that viruses can exploit. And as phones become more capable, virus-infected phones will be able to do more harm.
What will the viruses do after they break in? Thompson predicts that they’ll make expensive calls to overseas pay-services, running up the victim’s phone bill and transferring money to the pay-service owners, who presumably will be in cahoots with the virus authors. That might happen, but I don’t think it’s the most likely scenario.
The best bet, I think, is that cellphone viruses will look like PC viruses. In the PC world, many viruses are written for kicks, with no specific intent to cause harm (though harm often results when the virus spreads out of control). I would expect to see such mostly-harmless viruses in the cellphone world; and indeed that is what we apparently see with the CommWarrior virus described in the article. Other PC viruses aim to spy on the user, or to install a bot on the computer so that it can be commandeered later to send spam or launch denial of service attacks. All of this is likely in the cellphone world.
Will cellphones be able to resist viruses more effectively than PCs do? Thompson suspects they will:
Phone executives like to say that it’s easy for them to contain worms because their networks are gated communities. Verizon and Sprint can install antivirus software on their servers to automatically delete infected multimedia messages before they reach their victims.
…
The mobile-phone industry could solve the viral problem by developing an open-source, Linux-style cellular operating system. But that’s about as likely as Motorola and Nokia announcing that all your cell phone calls are going to be free.
I’m not as hopeful. Phone execs like to think of their networks as gated communities; but in the smartphone world all of the action is on the smartphone devices, not in the networks themselves, and the providers have less control over smartphone software than they think. Their communities may be gated, but the gates will have well-known holes (that’s how viruses will get in), and there will be plenty of third-party application software coming in and out. A smart device is only useful if it is configurable, and configurability is the enemy of the sort of regimented configuration control that they are invoking. Third-party services and applications provide tremendous value to users, but as users switch to such services the network providers lose control over users’ data.
The open-source argument is pretty weak too. An open-source operating system may have fewer security flaws (and even that is subject to debate) but the claim that it will have no known flaws, or nearly none, isn’t credible.
The more useful smartphones get, the more they will adopt a software structure like that of PCs, with all of the benefits and problems that come with such a structure – including viruses.
Apple Closes iTunes Store "Security Hole"
March 22, 2005 by
Last week, DVD-Jon and two colleagues released PyMusique, a tool for buying songs from Apple’s iTunes Music Store (iTMS) site. This got some people upset, because songs bought with PyMusique were not encumbered by any copy protection. Now Apple, predictably, has updated iTMS to make it incompatible with PyMusique.
The standard narrative about this goes as follows: (1) DVD-Jon and friends discover a security hole in iTMS. (2) The write PyMusique, which exploits the hole to get unprotected music. (3) Apple fixes the hole and iTMS is secure once again. The standard narrative misses the point entirely.
For starters, the security mechanisms of iTMS were, and are, well designed. A system that does what iTMS does will necessarily be unable to prevent unauthorized copying of music. That’s just a fact. Apple, to its credit, didn’t overinvest in fancy anti-copying technology that would be defeated anyway. Instead, Apple built a more modest and – here’s the key point – user-friendly system that gave users freedom to make legal use of music and provided speed bumps to steer consumer behavior, but didn’t pretend to stop determined infringers. There was no point in trying to stop determined infringers, because (a) there was nothing Apple could do to stop them from ripping iTMS content, and (b) all of the songs that might be ripped from iTMS were already available on the darknet anyway.
iTMS security is a bit like the lock on your screen door: it’s not very strong, but it doesn’t have to be, because the screen door around it is inherently vulnerable anyway. Putting an expensive lock on your screen door is a waste of money because it doesn’t make you any safer. Similarly with iTMS: spending more on copy protection would have been a waste, because it wouldn’t have reduced infringement.
Rather than owning up to its savvy engineering decision not to overinvest in fruitless copy protection, Apple apparently feels compelled to pretend publicly that iTMS is “secure” in the sense that heroic effort is required to illegally redistribute content bought from iTMS. That’s obviously untrue, but Apple is unwilling to admit that in public. (The famous reality distortion field plays a role here.)
So DVD-Jon and friends came along and released software that let people buy music that wasn’t wrapped in the usual weak iTMS copy-protection mechanisms. It was always possible to get such music, by buying it via the normal methods and then stripping off the copy-protection in one of several well-known ways. So PyMusique didn’t prove anything that we didn’t already know; but it didn’t really harm Apple or anybody else either.
Still, Apple apparently wanted to maintain the pretext of iTMS security, so it updated iTMS to make it incompatible with PyMusique. It’s still possible to make a new version of PyMusique that lets people buy music from iTMS and end up with that music in uncopyprotected form; but at least Apple can give the impression of policing its security perimeter.
We haven’t seen the end of this charade. Expect more iTMS “bugs” and more “fixes” from Apple.
UPDATE (7:50 PM): As predicted, DVD-Jon has reverse-engineered Apple’s fix and says he can now reenable PyMusique. That was quick!
Cornell Researchers on P2P Quality Control
March 21, 2005 by
Kevin Walsh and Emin Gün Sirer, of Cornell University, have a new paper on Credence, a system for detecting unwanted files in P2P networks. It’s a kind of reputation system for files, designed to detect in advance that certain files are not what they claim to be. One use of this technology is to detect spoofed files inserted into P2P nets by copyright owners.
Credence is really a reputation system for files. Users cast votes, which are simple thumbs-up or thumbs-down verdicts on particular files, saying whether a file is what it claims to be. Every vote is digitally signed by the user who cast it, so that recipients can verify the authenticity of votes they are given. If you’re not sure whether a file is genuine, you can ask people to send you votes, and then you can combine the votes in a special way to yield a quality score for the file.
P2P systems are open, and they generally don’t register their users (or at least they don’t prevent fraudulent or repeated registrations) so users cannot reliably be identified by their real names. Instead, users make up pseudonyms for themselves. Suppose Alice wants to join the system. She makes a pseudonym for herself, by generating a cryptographic key-pair, consisting of a private key and a public key. Alice keeps the private key secret, and uses it to put digital signatures on the votes she casts. Along with each vote, she distributes a copy of the public key, which anybody can use to verify the digital signature. The public key also serves as a “name” for Alice’s pseudonym. The key attribute of this system is that if Bob wants to forge a vote, that is, to create a vote that appears to have come from Alice’s pseudonym, he must somehow determine Alice’s private key, which is essentially impossible if Alice does her cryptography correctly. In short, the cryptography ensures that anybody can make a pseudonym, but only the creator of a pseudonym can cast votes on its behalf.
This only solves half of the problem, because an adversary can create as many pseudonyms as he likes, and have them cast false votes (i.e, votes in favor of the validity of files that are actually invalid, or against the validity of files that are actually valid). So you can’t just add up all of the votes you receive; you need some way to tell whose votes to trust and whose to ignore. Here Credence uses a simple rule – trust people who tend to vote the same way that you do. Suppose Alice knows that files A, B, and C are valid, and that files X, Y, and Z are not valid. If some pseudonym “Bob” has votes in favor of A, B, and C, and against X, Y, and Z, then Alice concludes that “Bob” tends to vote accurately. If another pseudonym “Charlie” votes the opposite way on those six files, then Alice concludes that votes from “Charlie” tend to be the opposite of the truth. So if she sees some new file that “Bob” says is valid and “Charlie” says is invalid, Alice will conclude that the file is valid. Each party’s vote on the new file gets a weight, equal to the correlation between that party’s votes and Alice’s votes on other files. (The paper hints at further mechanisms that assign trust to people whose votes correlate with those of other people Alice trusts.)
This scheme presents would-be adversaries with a dilemma. If Alice’s votes are truthful, then if you want to mislead Alice about one file, you have to earn her trust by telling her the truth about other files. You can tell occasional lies, but on the whole you have to be a truth-teller. (You can achieve the same effect by lying about almost everything, and telling the truth about just one file. Then Alice will conclude that you are a habitual liar, and will count your votes with negative weight, giving credence to the opposite of what you say. Again, you have to provide Alice with many useful-to-her votes in order to trick her once.)
It looks like this method will work, if it can be implemented efficiently in a real network. The real question, I think, is whether it will scale up to enormous P2P networks containing huge numbers of files. Here I have serious doubts. The paper’s authors don’t claim that users have to know about all of the votes cast in the system. But they’re not entirely clear on how individual users can efficiently get the votes they need to make good decisions, if the network is very large. In the long run, I don’t think this scaling problem is insurmountable; but more research is required to solve it.
