Articles by Tim Lee

Timothy B. Lee (Contributor, 2004-2009) is an adjunct scholar at the Cato Institute. He is currently a PhD student and a member of the Center for Information Technology Policy at Princeton University. He contributes regularly to a variety of online publications, including Ars Technica, Techdirt, Cato @ Liberty, and The Angry Blog. He has been a Mac bigot since 1984, a Unix, vi, and Perl bigot since 1998, and a sworn enemy of HTML-formatted email for as long as certain companies have thought that was a good idea. You can reach him by email at leex1008@umn.edu.


The RIAA’s New Clothes

by on October 10, 2005

I’ve got a new article up about the recording industry’s short-sighted strategy to online music downloads. I point out that digital rights management technologies don’t prevent piracy, but they do treat their customers like criminals and give people like Steve Jobs control over their customers.

I think it would be great if it became conventional wisdom that DRM technology is the perpetual motion machine of the 21st century. DRM is fundamentally contrary to the way computers work, because there’s no such thing as an uncopyable bit. You can write software to obfuscate your data, thus making copying more cumbersome, but that just makes cracking it more time-consuming, not fundamentally more difficult. Every few years, technology companies promise a new generation of copy-protection that will actually work. And each generation, they fail miserably.

The sooner the folks at the RIAA and MPAA realize that, the sooner they’ll stop hassling their paying consumers with arbitrary and pointless restrictions that penalize their customers while doing nothing to stop pirates.

SHARE:

Confused reasoning about the DMCA

by on October 6, 2005

Over at CNET, Adam’s colleague Patrick Ross has an incredibly confused defense of the Digital Millenium Copyright Act:

When content producers know that they can experiment with various protection approaches, they’re more comfortable entering the online market. Also, investors are more inclined to fund such efforts. Imagine a world of unlimited digital content, packaged with a range of TPM at varying prices. In that world, consumers can purchase exactly the amount of use they need and not pay for more. But if HR-1201 [the Digital Media Consumers Rights Act, introduced by Rep. Boucher] becomes law, every consumer could legally hack any TPM by claiming fair use, and as fair use isn’t codified, there would be as many definitions of it as there are consumers. Consumers would be legally sanctioned to break their contracts with the content provider. No sane business operator enters a contract in which one party has the right to disregard its terms at will, but that’s what HR-1201 permits. That hated TPM would disappear from the market, as there’s no reason to employ a lock if everyone has a legal right to the key. But as TPM leaves, so do the digital offerings that come with it.

The first sentence of the last paragraph is a simple, unambiguous falsehood. The Boucher bill specifies “that it is not a violation of [the DMCA] to circumvent a technological measure in connection with gaining access to or using a work if the circumvention does not result in an infringement of the copyright in the work.”

You tell me how that abrogates any contracts.

This isn’t just nitpicking. TPM schemes can place any kind of restriction they want on their cusomers, including restrictions imposed retroactively and without notifying their customers. (Apple, for example, unilaterally reduced the number of copies of its songs existing customers were allowed to burn to CD after the songs had already been sold) There is no reason whatsoever to assume that a TPM scheme constitutes a “contract” between a company and its customers. Prohibiting circumvention of TPM schemes doesn’t aid contract enforcement, because a TPM isn’t a contract.

On the other hand, the Boucher bill would not in any way prevent the enforcement of actual contracts. Fair use is not a defense against breach of contract, and Boucher’s legislation wouldn’t change that. So if a user signs an agreement promising not to break a TPM scheme, and then breaks it anyway, the company that designed the TPM would have every right to sue the customer.

There’s also absolutely no reason to think that content wouldn’t be made available for download without TPMs. Every CD is effectively TPM-free. I could take any CD I own, “rip” the songs on it, and upload them to a P2P network. Yet the recording industry still sells CDs. Why would we think online downloading would be any different?

The folks at PFF desperately want to portray the DMCA as a “free market” approach to copyright problems. But the shoe just doesn’t fit. TPM systems are not contracts, and circumventing them, as such, is not theft. Simply stated, the DMCA is a government-imposed restriction on the design of technological devices. It’s fine to argue that such restrictions are needed to curtail piracy, but such restrictions have nothing to do with freedom of contract.

SHARE:

More on Google Print

by on September 25, 2005 · 6 comments

I’m certainly not going to claim that Google is going to win in court, since fair use determinations are notoriously hard to predict. But I think that on the merits, their case is a lot stronger than Jerry gives them credit for.

It’s not clear to me why Jerry doesn’t consider a parody a derivative work. By definition, a parody takes a work, retains some elements of it, drops some others, and transforms it into a new work. We could imagine an alternative universe in which publishers sold “parody rights” the same way they now sell film rights or sequel rights.

UMG v. mp3.com was a badly reasoned, short-sighted decision, and I think there’s a very good chance that it would have been overturned on appeal. If the decision is right that “Any allegedly positive impact of defendant’s activities on plaintiffs’ prior market in no way frees defendant to usurp a further market that directly derives from reproduction of the plaintiffs’ copyrighted works,” then it’s hard to see how any fair use defense could succeed. After all, one could imagine a world in which Hollywood won the Sony Betamax case and licensed Hollywood-approved VCRs that charged the consumer a fee every time a TV show was recorded. Or, if the music industry had won the Diamond case, they could have licensed MP3 players that send the appropriate publisher a penny every time a song was played. Likewise, if Kelly v. Arriba Soft had come out in favor of Mr. Kelly, perhaps owners of photographs could have licensed “search engine rights” to their photographs.

All fair uses involve reproducing copyrighted content in order to create derivative works. By definition, that undermines the ability of the copyright holder to sell derivative works of the same kind. If that were the standard, no use would withstand fair use scrutiny. That’s why courts have generally focused on the market for the original product, not the market for hypothetical derivative works.

Personally, I think the factor that will weigh most heavily in Google’s favor is factor 1, the purpose and character of the use. As the Supreme Court put it, that factor asks whether a use “merely supersedes the objects of the original creation, or whether and to what extent it is transformative.” MP3.com was arguably superseding the market for ordinary CDs. Google Print is clearly not superseding the market for books. And it’s every bit as transformative as Arriba Soft’s software was.

SHARE: 6 comments

Innovation under Attack

by on September 22, 2005

Publishers have sued to stop Google Print, a search engine for books, on the theory that it’s an infringement of copyright to make digital copies of copyrighted books, even if they never show those copies in their entirety to anyone.

The publishers’ position is anti-innovation in a very fundamental way. In the analog world, there’s a clear distinction between “using” a copyrighted work (say, reading a book) and “copying” it (say, using a photocopier). Copyright law says that you’re allowed to use a book you legally own, but generally speaking, you can’t make copies, at least not in a commercial product.

But the “physics” of the digital world are different. Every “use” of content involves the creation of a copy of that data. When you read this web page, dozens of copies of the document were created as it was passed across the Internet. If making a digital copy is a copyright infringement, that means that no one can use their copyrighted content on digital systems without the explicit permission of the copyright owner.

Fortunately, that’s not how the courts have ruled in the past. In 1984, the Supreme Court held that it was a fair use to make personal copies of TV shows for the purpose of “time shifting.” In the 1999 Diamond decision, the Ninth Circuit held that “space shifting”–making copies of music for listening on an MP3 player is a fair use. And in 2002, the Ninth Circuit held that displaying thumbnails of copyrighted images is a fair use. In each case, the court appreciated that new technological realities made the copying involved in these uses fundamentally different than the copying prohibited by traditional copyright law.

Unfortunately, judges have not always been so clear-sighted. In the 2000 MP3.com case, a stubbornly literalist district judge held that storing copies of CDs on MP3.com’s servers for future transmission to customers (all of whom had shown they were legal owners of the CDs) was not a fair use. Unfortunately, that case was settled before it could be appealed.

The courts need to clearly say that the mere act of making a digital copy is not a violation of copyright. What matters is how those copies are used. Fortunately, I think the folks at Google understand what’s at stake, and they know that the future of their business may depend on this issue. They are in the business of organizing the world’s information, most of which is owned by other people. If they have to get permission from each individual copyright holder, many of the innovative things they’d like to do with that information will become logistically impossible. So I’m crossing my fingers and hoping that Larry and Sergei fight this thing all the way to the Supreme Court.

SHARE:

Retarding the Progress of Science and the Useful Arts

by on September 8, 2005 · 8 comments

This is three weeks old, but I missed it when it happened, so better late than never:

Google has put its digital library project on hold until November, citing concerns from copyright holders and publishers. The search company unveiled Google Print in October 2004 as a way for publishers to make their books accessible over the Internet. The company also quickly introduced a comparable program to allow libraries to scan in their collections, index them, and make them web-accessible as well.

The rest of the article has a bunch of details on the concessions Google is making to the publishers to keep them happy. Maybe I’m missing something, but it looks to me like Google is bending over backwards to accomodate the publishers, and the publishers are stiff-arming them. As I argued back in July, I think Google has a perfectly plausible case that Google Print is a fair use under copyright law, and that they have every right to do what they’re doing without consulting the publishers and without giving them a cut of the ad revenue. Yet even after Google has offered to share the revenues with the publishers and easily opt out of the program, the publishers have refused to budge.

I think it’s hard to overstate how big a deal this is. We’re all used to using search engines to search content on the web. It would be amazing if the same functionality were to exist with every book in the library. Until now, the barrier has been technology. But now, when such a search engine is becoming technologically feasible, it looks like it’s going to be thwarted by the lawyers.

Why shouldn’t Goole just negotiate with the publishers and get their permission? If you want to see how that will end up, just look at Nexis and Factiva. Nexis is an extremely useful and powerful search engine, but it’s expensive, it’s proprietary, and it’s not especially user-friendly. And if you want to search for articles that appeared in the Wall Street Journal, you can’t do that, because the Dow Jones company decided they wanted to create their own proprietary search service. It’s a royal pain in the ass, and it will probably never change, because anyone who tried to create a full-text search of Wall Street Journal content– even if they made you buy the actual articles from the WSJ web page– would get sued out of existence.

We’re at a crossroads. If Google (or someone else) stands up to the publishing industry and wins in court, we’ll end up with a future in which book searches are like web searches. If Google caves, and no one else dares get into a lawsuit with the publishers, then book searching is going to look like Nexis and Factiva– proprietary, expensive, and fractured among different mutually exclusive services.

The saddest part, I think, is that hardly anyone is even paying attention. If Google Print gets strangled by the publishing industry, 99% of consumers will never even realize what they’ve lost.

SHARE: 8 comments

“Unbundling”: Good Riddance

by on August 6, 2005

The FCC has finally put an end to the dubious 1990s experiment with “unbundling” DSL services. Or, to put it a little bluntly, the FCC has decided it will no longer expropriate the infrastructure of the Baby Bells to be used at government-mandated prices by their competitors.

Various liberal commentators, such as Matt Yglesias have painted unbundling as a noble Clinton-era experiment that was allowed to wither on the vine under the Bush administration. Had the Republicans continued the Democrats’ vigorous efforts on behalf of competition, the theory goes, we would now have a healthy, competitive broadband marketplace.

Matt points out that Southeast Asia enjoys better broadband service than the United States and suggests that Japan and South Korea pursued “open access” policies like those the FCC is now abandoning, while the United States has dropped the ball.

But that line of argument doesn’t make very much sense. If you’re Verizon, and you know that you will be required to “share” any new infrastructure you build with your competitors, you are unlikely to spend very much money upgrading your infrastructure. And if you’re one of those competitors, you have little reason to build competing infrastructure when you have guaranteed “access” to Verizon’s infrastructure at government-mandated prices. Hence, although you might have some “competition” in downstream services, “open access” policies are going to retard the build-out of new infrastructure in the “local loop.”

You can see this dynamic in Matt’s own backyard. A couple of years ago, Verizon began rolling out its fiber-to-the-premises service in select cities including Northern Virginia. They did so only after receiving assurances from the FCC that any new fiber infrastructure would not be subject to unbundling rules. You can also see it at work in the cable market: today, some cable companies are offering 6 Mbit services. When I had a cable modem 3 years ago, the best I could get is 1.5 Mbit.

Now, Matt points out that other countries have even faster service. He says we’re falling behind. I have to admit I haven’t studied the telecom markets in Japan and South Korea in any detail, so I can’t say what other differences might explain the discrepency. A couple of things come to mind. One is that, obviously, higher-density cities will have an easier time rolling out new services. Another is that, because Japan and South Korea industrialized fairly recently, the wires in the ground are likely to be newer than those in the United States, making rollout of faster services more practical. It’s also possible that cultural factors come into play. Japan and South Korea are intensely literate and gadget-happy societies. South Korea is obsessed with video games the way we’re obsessed with football. It’s possible (although I’m by no means claiming that I know enough to say it’s true) that Japan and South Korea have more broadband users because they have more nerdy people.

In any event, if your goal is to spur investments in new infrastructure, the first step must be to insure that the company that invests capital reaps the profits from that investment. It’s hard to see how “open access” rules could possibly accomplish that. Once the dust has settled from the well-deserved death of “unbundling,” we should have a thorough debate about how best to lower barriers to entry to the broadband market, so that companies can more easily build infrastructure (especially wireless infrastructure) to compete with the incumbents. But the debate must start with the principle that the government should respect the rights of companies who invest in infrastructure to profit from their investments, rather than “unbundling” them and giving “access” to other companies who have not bothered to make such investments.

SHARE:

A La Carte Nonsense

by on July 26, 2005 · 6 comments

I’m not too surprised to see populist interest groups on the right and left jump on an intellectually incoherent but crowd-pleasing proposal like “a la carte” cable, but Matt Yglesias should know better.

What everyone seems to miss in this debate is that a cable channel isn’t like a banana. If every grocery store somehow forced you to buy a banana with every orange, the banana-orange bundle would be more expensive than a banana or an orange alone, and a lot of bananas and oranges would end up in the garbage.

But a cable channel is a non-rivalrous good. The marginal cost of providing it to another consumer is zero. The goal of the cable company is to recover it’s rather large fixed costs in equipment, programming, etc. It will price its products so that it is able to recover those costs along with a profit margin.

To simplify things, let’s imagine that a cable company has only two channels, Spike TV and Women’s Entertainment, and only two kinds of customers, men and women. Men value STV at $10 and WE at $4. Women value WE at $10 and STV at $4. The cable company might bundle the channels together and charge $12 for the bundle. Each consumer would be getting $14 of TV for $12.

Now, people like Yglesias seem to assume that in an a la carte world make each channel would cost, say, $6. In that case, men would buy only STV, women would buy only WE, and consumers would save a bunch of money.

But that’s absurd. The cable company would lose half of its revenue in that scenario, and would be unlikely to even be covering its fixed costs. More likely, it would set the price for each channel at $10. The cable company would still be losing a lot of revenue, but that might be enough to keep it in business.

But notice that both the consumer and the cable company loses in this scenario. Before, the cable company was getting $12/subscriber, now it’s getting $10. The male consumer, meanwhile, went from getting $14 of TV for $12 to getting $10 of TV for $10. There might be a show he likes on WE, but not that he likes enough to pay twice as much for his cable bill.

Bundling increases consumer welfare by distributing low-marginal-cost goods to wider audiences. A la carte cable wouldn’t save consumers money. It would simply reduce the number of channels on their TVs. Buying twice as many cable channels isn’t like buying twice as many bananas.

It might be objected that the cable company does pay a per-viewer fee to the studio for those channels. But that’s just the same phenomenon one step removed. How do the studios price their channels when selling them to the cable company? Their marginal costs are also close to zero, so the same bundling argument above applies to them. If they gave their customers the option of buying channels a la carte, they’d have to dramatically raise their per-subscriber rates to cover their fixed costs. Consumers would be the loser–paying about the same for a much smaller variety of channels.

SHARE: 6 comments

Hands Off My Cell Phone

by on July 16, 2005

I disagree with James on cell phone bans. First of all, as one of his commenters point out, cell phones are not the only distracting thing in the car. They’re probably not even the most distracting. People eat, yell at their kids, change the station on the radio, apply make-up, and do all manner of other distracting things in the car. It’s not at all clear to me why we should single out cell phones for special treatment.

Secondly, context matters. If I’m zipping along in the left lane of an almost-empty freeway, being on my cell phone poses pretty minimal risk of accident. Likewise, if I’m in a residential area cruising along at 5 MPH (say, I’m almost to a friend’s house and calling the friend for directions) my chances of getting in an accident are likewise pretty low. And anyway, the damage will be minimal if I hit something at 5 MPH. So no, you shouldn’t be on the phone while changing lanes in rush hour traffic. But not all cell phone use in cars is bad.

Thirdly, is that really the best use of police resources? Even if the study is right, and cell phone use is killing people, it’s not at all clear that a ban would do much to deter cell phone use. It’s not very easy to tell who’s using a cell phone from outside, and there aren’t nearly enough police officers to enforce a ban effectively. A lot of people will just ignore the ban, on the (reasonable) assumption that they’re unlikely to be caught. I mean, really, has mandatory seat belt laws increased seatbelt use?

Finally, the study found that handsfree phones are just as distracting as normal phones. I don’t find this surprising at all. DC has a cell phone ban, and so I tried to use my hands free kit as often as possible. When my phone rings, I have to fish my phone out of my pocket, fish the handsfree kit out of the ash tray, plug the receiver into the phone, put the reciever into my ear (sometimes it falls out and I have to do it again) and then find a place to set the phone for the duration of the call. Since my cell phone calls are usually quite short, I think I’m a lot less dangerous having a phone to my ear for 30 seconds than spending 15 second fiddling around with things on my lap while the phone is ringing.

Bottom line: the police have better things to worry about.

SHARE:

Wires Are So 20th Century

by on July 11, 2005

There are officially more cell phones than land lines in the United States. I ditched my last landline in 2003 and I haven’t looked back.

If there was ever a philosophical argument for regulating telephone service as a “natural monopoly,” there certainly isn’t any more. These days, the Baby Bells are just four competitors in the vibrant market for telephone service. Their phones just happen to be the anachronistic ones with cords still attached to them.

(Hat tip: Ezra)

SHARE:

I’m an Intruding Hacker

by on July 7, 2005 · 6 comments

Alex Leary of the St. Petersburg times is on the story of an “intruder” who “hacked” into a wide-open WiFi and “stole” Internet service. Fortunately, the good guys caught him in the act, and now he’s facing hard time.

Mr. Leary, needs to take a deep breath and calm down.

Here’s something he might ponder once he’s got his blood pressure under control: on a recent trip to the Midwest, I probably “stole” Internet access from a dozen open access points around town. When I needed to check my email or look up an address, I’d drive down the street until I found myself within range of an unsecured wireless network. Then I logged in, checked my email, and logged off.

It’s almost certain that the owner didn’t even know I was there. The amount of bandwidth I “stole” was trivial, probably worth a fraction of a penny. I didn’t try to hack into anyone’s computer or snoop their private data. So am I a criminal?

ISPs will point out that such access technically violates their Terms of Service. Which is true. Comcast, for example, prohibits its broadband users from “making available to anyone outside the Premises the ability to use the Service (i.e. wi-fi, or other methods of networking).” But the TOS is an agreement with the guy who owns the access point, not with the guy who’s barrowing it. If Comcast has a beef with how its customers are using their service, they should take that up with the customer, not the guy driving by on the street. More to the point, this is the sort of thing that’s best dealt with with benign neglect. That provision in the TOS is to prevent a whole apartment building from sharing one broadband connection, depriving the ISP of revenue. But allowing me to use a WiFi network for 30 seconds isn’t going to make me any less likely to sign up for home broadband.

But I guess people have a tendency to get freaked out about things they don’t understand. When everything about a computer network is a mystery to you, I imagine it’s frightening to think that random strangers might have access. But whether or not you think it’s ethical to log into an unsecured wireless network, it’s certainly not a huge deal. Trading kiddie porn and stealing credit card numbers are a big deal, but one can do that with any Internet connection. There’s nothing special about WiFi in that respect.

Moreover, it takes all of 2 minutes, and virtually no technical savvy, to set a password for your access point. The procedure depends on which one you’ve got, so consult your manual, but most likely it involves opening your browser, typing in a number like “10.0.0.1” for the address, and clicking a “change password” button. If you don’t want people sharing your network–and more power to you if you don’t–that will deter 99.9% of the people who might try to log in. And I’d certainly be open to the idea that the remaining .1% should be liable for criminal sanctions.

A more tech-savvy reporter likely would have made fun of the bumbling flatfoots who think checking your email on a wide-open computer network is a felony. Of course, a more tech-savvy policeman wouldn’t have made an arrest in the first place.

SHARE: 6 comments

← Previous Entries

Next Entries →