Chapter 7

are not like physical property rights. In analyzing the DMCA, where do we turn for analogies? To physical property, violence, and theft. The cases analyzing the DMCA are full of analogies to trespass, to breaking and entering, to burglars' tools, and to safecrackers. Private property carries a lot of baggage with it, but we know it well--it is the place we naturally turn for insight. Even I, in order to point out some of the difficulties with those analogies, had to turn to farmers and barbed wire and public rights-of-way along highways. There is nothing wrong with analogies. They help us understand things that are new by comparing them to things we think we understand better. Analogies are only bad when they ignore the key difference between the two things being analyzed. That is what happens here. 83

Jefferson reminded us that intellectual property rights are clearly artifacts of state creation, monopolies whose internal limitations in scope, duration, and so on are just as important as the rights themselves. Jefferson doubts whether even property rights over land can be understood as natural and absolute--copyrights and patents, which cover subject matter that can be infinitely reproduced without diminishing its substance, clearly cannot. They frequently involve a claim to control purchasers' behavior with respect to some aspect of an artifact after it has been sold to them in the marketplace, making simpleminded analogies to "breaking and entering" inappropriate--the extent of the property in question is precisely the issue in dispute. (When Johansen was tried in Norway under the national computer crime law, the court laconically observed that he had bought the DVDs, and one cannot break into one's own property--effectively turning the analogy on its head.) Jefferson starts from the baseline that monopoly is the exception and freedom is the rule--any limitations on that freedom have to be justified. That is why he always discusses the right and the limitations on the right as an inseparable pair. One cannot discuss them in isolation. 84

Kaplan and Newman are fine, thoughtful judges. They do not altogether ignore those points. But look how the analysis is set up. At several points in the discussion, there seems to be the assumption that copyright owners have entitlements to total control as of right and that fair use is a mere lucky loophole which, because it can be negated by the happenstance of whether one can get physical access, can hardly have major First Amendment status. They keep pointing out that physical control and tangible property rights frequently allow copyright holders to make fair use impracticable. "And so what?" Jefferson might have responded. This is a classic non sequitur. The question is whether the Congress has the power to add a new right of access- denial to the intellectual property monopoly it is constructing, undermining--as to some works and some fair uses--the balance that the law sets up. The citizen is not pleading for a new right of access, trumping all physical restraint and tangible property rights. The citizen is claiming that Congress has no power to give exclusive rights to restrain copying of digital content while simultaneously taking away the citizen's existing right to get access to that content for the purposes of fair use--at least in those cases where access is physically possible and violates no other property right, real or intellectual. 85

The Constitution does not require the United States to break into President Nixon's desk to get me his tapes, buy me a tape recorder, or give me a right to 18.5 minutes on the broadcast airwaves to play them. But if I can get access to the tapes legally, it does forbid the government from giving President Nixon the power to put a red dot on those tapes and thus claim an intellectual property right to stop me playing them on TV or digitizing them to make the sounds clearer. The restraints imposed by physical happenstance and tangible property rights are different from those imposed by copyright--a congressionally created monopoly over expression. We cannot assume because one is constitutionally acceptable that the others are too. Jefferson understood that, and his analysis can help us even in a constitutional conflict over a technology he could hardly have dreamt of. (Though perhaps with Jefferson, this is a bad bet.) 86

The same point comes up in a different way when the court disconnects the fair use discussion from the exclusive rights discussion. The question is not "Do I have a constitutionally protected right of physical access to a preferred version of a movie, so as to make my task easy?" That gets the court caught up in questions of when a majority of movies will only be available on DVD, or how poor a substitute the analog version would be, or how many fair uses will require actually cutting a digital fence. But all of these inquiries miss the point. The question is "Can Congress hand out the exclusive rights of copyright over digital works if it does not accompany those rights with the suite of limitations that the court has repeatedly said "saves" copyright from violating the First Amendment?" The proportion of digital works to the total number of works produced in other formats is irrelevant. As to these works, the rule is unconstitutional. But what about the number or proportion of types of fair uses affected? That is more relevant but still not dispositive in the way Kaplan and Newman imagined. True, not every trivial statutory modification of fair use makes copyright unconstitutional. But this is not a trivial modification: over an entire class of works, copyright owners are given a legal power to deprive users of their privilege to gain otherwise lawful access for the purposes of fair use. If you give the digital filmmaker the exclusive rights of copyright but forbid the film professor from going through the otherwise lawful process of parodying or quoting, that rule is unconstitutional, no matter how many other fair uses are unaffected. If the copyright law were amended to forbid journalists playing, on a Friday, excerpts of legally acquired red-dotted tapes made by presidents whose last name begins with N, it would still be unconstitutional. 87

The legal implementation of this conclusion would be simple. It would be unconstitutional to punish an individual for gaining access in order to make a fair use. However, if they cut down the digital fence to make illicit copies, both the cutting and the copying would be illegal. But what about the prohibition of trafficking in digital wire cutters, technologies such as DeCSS? There the constitutional question is harder. I would argue that the First Amendment requires an interpretation of the antitrafficking provisions that comes closer to the ruling in the Sony case. If Mr. Johansen did indeed make DeCSS to play DVDs on his Linux computer, and if that were indeed a substantial noninfringing use, then it cannot be illegal for him to develop the technology. But I accept that this is a harder line to draw constitutionally. About my first conclusion, though, I think the argument is both strong and clear. 88

Ironically, there is some support for my claim and it comes from an even higher, if not uniformly more thoughtful, set of judges than Newman and Kaplan. In the depressing case of Eldred v. Ashcroft, the Supreme Court upheld retrospective copyright term extensions against a variety of constitutional challenges. (Full disclosure: I assisted in the preparation of an amicus curiae brief in the case.) One of those challenges was based on the First Amendment. The fairly reasonable claim was that Congress could not retroactively lock up an entire twenty-year swathe of culture that had already been produced. Such a law would be all restraint of expression, performance, republication, adaption, and so on, with no incentive benefits. The Court was unconvinced. But it did say: 89

To the extent such assertions raise First Amendment concerns, copyright's built-in free speech safeguards are generally adequate to address them. We recognize that the D.C. Circuit spoke too broadly when it declared copyrights "categorically immune from challenges under the First Amendment." . . . But when, as in this case, Congress has not altered the traditional contours of copyright protection, further First Amendment scrutiny is unnecessary.14 90

The DMCA, of course, does exactly this. As to digital works it alters the "traditional contours of copyright protection" in a way that affects "copyright's built-in free speech safeguards." That is what the Farmers' Tale was all about. Perhaps one day, in a case not involving a Norwegian teenager, a hacker magazine run by a long-haired editor with an Orwellian nom de plume, and an obscure technology that is accused of posing apocalyptic threats to the American film industry, that point will come out more clearly. 91

But the issue of speech regulation is only half of the story. Intellectual property rights over digital technologies affect not only speech, but the framework of competition and markets as well, as the next example makes clear. 92

The Apple of Forbidden Knowledge: The DMCA and Competition 93

You could tell it was a bizarre feud by the statement Apple issued, one strangely at odds with the Californian Zen-chic the company normally projects. "We are stunned that RealNetworks has adopted the tactics and ethics of a hacker to break into the iPod, and we are investigating the implications of their actions under the DMCA and other laws."15 94

What vile thing had RealNetworks done? They had developed a program called Harmony that would allow iPod owners to buy songs from Real's Music Store and play them on their own iPods. That's it. So why all the outrage? It turns out that like the story of DeCSS, this little controversy has a lot to teach us about the landscape of intellectual property disputes, about the mental topography of the high-tech economy. But where the DeCSS case was a war of metaphors around the boundaries of freedom of expression, the iPod story is about ways in which intellectual property marks the limits of competition. 95

Apple iPods can be used to store all kinds of material, from word processing documents to MP3 files. If you want to use these popular digital music players to download copy-protected music, though, you have only one source: Apple's iTunes service, which offers songs at 99 cents a pop in the United States, 79 pence in the United Kingdom. If you try to download copy-protected material from any other service, the iPod will refuse to play it. Or at least, that had been the case until Real managed to make their Harmony service compatible. 96

Real's actions meant that consumers had two sources of copy- protected music for their iPods. Presumably all the virtues of competition, including improved variety and lowered prices, would follow. The iPod owners would be happy. But Apple was not. The first lesson of the story is how strangely people use the metaphors of tangible property in new-economy disputes. How exactly had Real "broken into" the iPod? It had not broken into my iPod, which is after all my iPod. If I want to use Real's service to download music to my own device, where's the breaking and entering? 97

What Real had done was make the iPod "interoperable" with another format. If Boyle's word processing program can convert Microsoft Word files into Boyle's format, allowing Word users to switch programs, am I "breaking into Word"? Well, Microsoft might think so, but most of us do not. So leaving aside the legal claim for a moment, where is the ethical foul? 98

Apple was saying (and apparently believed) that Real had broken into something different from my iPod or your iPod. They had broken into the idea of an iPod. (I imagine a small, platonic white rectangle, presumably imbued with the spirit of Steve Jobs.) Their true sin was trying to understand the iPod so that they could make it do things that Apple did not want it to do. As an ethical matter, does figuring out how things work, in order to compete with the original manufacturer, count as breaking and entering? In the strange netherworld between hardware and software, device and product, the answer is often a morally heartfelt "yes!" I would stress "morally heartfelt." It is true manufacturers want to make lots of money and would rather not have competitors. Bob Young of Red Hat claims "every business person wakes up in the morning and says 'how can I become a monopolist?' " Beyond that, though, innovators actually come to believe that they have the moral right to control the uses of their goods after they are sold. This isn't your iPod, it's Apple's iPod. 99

Yet even if they believe this, we don't have to agree. In the material world, when a razor manufacturer claims that a generic razor blade maker is "stealing my customers" by making compatible blades, we simply laugh. The "hacking" there consists of looking at the razor and manufacturing a blade that will fit. To say this is somehow immoral seems laughable. Is the conclusion changed when the information about compatibility is inscribed in binary code and silicon circuits, rather than the molded plastic of a razor cartridge? What if ensuring the "fit" between the two products is not a matter of making sure the new blades snugly connect to the razor but of making sure the software embedded in my generic product sends the right code to the original product in order to be recognized? Our moral intuitions are a little less confident here. All kinds of bad policy can flourish in that area of moral uncertainty. 100

This leads us to the law. Surely Apple's suggestion that the DMCA might prohibit what Real had done is as baseless as their moral argument? In the United States, the answer is "probably," at least if the courts continue in the direction they are currently taking, but it is a closer call than you would think. Internationally, the answer is even less certain. That is where the iPod war provides its second new-economy lesson. Think for a moment about the way that the law shapes the business choices in this dispute. 101

In a competitive market, Apple would choose whether to make the iPod an open platform, able to work with everyone's music service, or to try to keep it closed, hoping to extract more money by using consumers' loyalty to the hardware to drive them to the tied music service. If they attempted to keep it closed, competitors would try to make compatible products, acting like the manufacturers of generic razor blades or printer cartridges. 102

The war would be fought out on the hardware (and software) level, with the manufacturer of the platform constantly seeking to make the competing products incompatible, to bad-mouth their quality, and to use fear, uncertainty, and doubt to stop consumers from switching. (Apple's actual words were: "When we update our iPod software from time to time, it is highly likely that Real's Harmony technology will cease to work with current and future iPods.") Meanwhile the competitors would race to untangle the knots as fast as the platform manufacturer could tie them. If the consumers got irritated enough they could give up their sunk costs and switch to another product altogether. 103

All of this seems fine, even if it represents the kind of socially wasteful arms race that led critics of capitalism to prophesy its inevitable doom. Competition is good and competition will often require interoperability. But what do we mean by competition? Is it competition if I assassinate your employees or poison the food in your restaurant? If I trespass on your land in order to sell a competing product? If I break into your safe to steal your trade secrets, use my monopoly position in the market to impose resale price agreements, or violate your patent? It is the law that draws the line between competition and theft, between virtuous competitive imitation and illicit "piracy." 104

Sometimes we need to give innovators property rights that allow them to prevent second-comers from free riding on their efforts. We have to do so because it is necessary to encourage future innovation. On the other hand, sometimes we not only allow the second-comer to free ride, we positively encourage it, believing that this is an integral part of competition and that there are adequate incentives to encourage innovation without the state stepping in. Intellectual property policy, indeed a large part of the policy behind all property rights, is about drawing the line between the two situations. Too far in one direction and innovation suffers because potential investors realize good ideas will immediately be copied. Too far in the other direction and monopolies hurt both competition and future innovation. 105

Imagine you are the first person to invest in getting the public to eat burritos for breakfast, or to place a petrol station at a certain crossroads, or to clip papers together with a folded bit of wire. In each case we give you some property rights. The fast-food vendor may own a trademarked phrase or jingle that the public learns to associate with his product. Since the patent office issued a patent for the sealed and crimped "peanut butter and jelly" sandwich I described at the beginning of the book, even a patent is not out of the question if your disgusting concoction is sufficiently novel and nonobvious. But we should not allow you to have a patent over all burritos, or burritos for breakfast, still less over the idea of fast food. As for the paper clip maker, there might be a trademark over the particular paper clip, but the idea of folding wire to secure paper stays in the public domain. The owner of the petrol station gets physical ownership of the land, but cannot stop a second-comer from setting up shop across the road, even if the first-comer's labor, capital, and effort proved that the location is a good one. We positively encourage follow-on imitation in those cases. 106

Now how about the case in point? What does Apple get in the way of property rights? Think back to my description of the intellectual property system in Chapter 1. They can get patents over those aspects of the iPod--both hardware and software--that are sufficiently innovative. Patents are what we use to protect inventions. They also get a copyright over the various pieces of software involved. That protects them only against someone who copies their code, not someone who writes new software to do the same thing. Copyrights are what we use to protect original expression. They get rights under trademark law over the name and perhaps parts of the design of the product--maybe the distinctive look of the iPod--though that is a bit more complex. All of these rights, plus being the first to break into the market in a big way, the brilliance of the design, and the tight integration between the hardware and the service, produce a formidable competitive advantage. The iPod is a very good product. 107

Now if a competitor infringes any of Apple's rights, for example by making a literal copy of the code, using their trademark in a way the law does not allow, or infringing on one of their patents, then Apple can shut them down and extract hefty damages. Quite right, too. But should they be able to prevent someone from making an interoperable product, provided they do not violate any of these existing rights in the process? Laws like the DMCA make that question more complicated. 108

Nowadays, there is software in many, many more products than you would imagine. Your watch, your phone, your printer, your thermostat, your garage door opener, your refrigerator, your microwave, your television--the odds are that if you bought them in the last ten years, they have some software component. In the 1970s the courts and Congress had concluded that software could be copyrighted as original expression, like a song or a novel, as well as being patented when it was novel, nonobvious, and useful. Frequently, different aspects of the same program will be covered by copyright and by patent. But software is a machine made of words, the machine of the digital age. That fact already causes some problems for our competition policy. Will the exceptions and limitations designed to deal with a copyright over a novel work adequately when they are applied to Microsoft Windows? That issue was already unclear. With the DMCA, we have added another crucial problem. Where there is copyrighted software there can be digital fences around it. If the copyright owner can forbid people to cut these fences to gain access to the software, then it can effectively enlarge its monopoly, capture tied services, and prohibit generic competition. 109

It was just this line of thought that led some other companies to do more than merely make threatening noises about the DMCA. Lexmark makes printers. But it also makes lots of money off the replacement ink or toner cartridges for those printers. In some cases, in fact, that is where printer companies make the majority of their profits. As a result, they are not exactly keen on generic replacements. Chamberlain makes garage door opener systems. But they also sell replacements for the controllers--the little devices that you use to trigger the door. Lawyers from both of those firms looked at the DMCA and saw a chance to do something most companies would love to do; to make generic competition illegal. Lexmark designed their printer program so that it would not accept a toner cartridge unless it received the correct "checksum" or validation number. So far, this looks no different from the razor manufacturer trying to make it difficult to manufacture a compatible replacement blade. Generic competitors now had to embed chips in their printer cartridges which would produce the correct code, otherwise they would not work in Lexmark printers. 110

Static Control Components is a North Carolina company that manufactures chips whose main function is to send the correct code to the printer program. With this chip implanted in them, generic cartridges would work in Lexmark printers. Lexmark's response could have been to change their program, rendering the chip obsolete, just as Apple could change the iTunes software to lock out Real Music's Rhapsody. Doing so would have been quite within their rights. Indeed it is a standard part of the interoperability wars. Instead, Lexmark sued Static Controls, claiming, among other things, a violation of the DMCA.16 Like Apple in the press release I quoted earlier, Lexmark clearly saw this as a kind of digital breaking and entering. This was their printer, their printer program, their market for replacement cartridges. Static was just helping a bunch of cheats camouflage their generic cartridges as authentic Lexmark cartridges. Translated into the legal language of the DMCA the claim is a little different, but still recognizable. Static was "trafficking" in a device that allowed the "circumvention of a technical protection measure" used to prevent "access to a copyrighted work"--namely the computer program inside the printer. That is behavior that the DMCA forbids. 111

The garage door company, Chamberlain--who also claimed to be concerned about the security of their garage doors--made a similar argument. In order to get the garage door to open, the generic replacement opener had to provide the right code to the program in the actual motor system. That program is copyrighted. The code controls "access" to it. Suddenly, the manufacturers of generic printer cartridges and garage door openers start to look rather like Jon Johansen. 112

Surely the courts did not accept this argument? Bizarrely enough, some of them did--at least at first. But perhaps it was not so bizarre. The DMCA was indeed a radical new law. It did shift the boundaries of power between intellectual property owners and others. And intellectual property rights are always about restraining competition, defining what is legitimate and what is not--that is what they do. There was a respectable argument that these devices did in fact violate the DMCA. In fact, it was respectable enough to convince a federal judge. The district court judge in the Lexmark case concluded that Lexmark was likely to win on both the DMCA claim and on a more traditional copyright claim and issued an injunction against Static Control. In Skylink, the case involving garage door openers, by contrast, the district court held that the universal garage door opener did not violate the DMCA. Both cases were appealed and both appeals courts sided with the generic manufacturers, saying that the DMCA did not prohibit this kind of access--merely making a computer program work the way it was supposed to. 113

The U.S. Court of Appeals for the Federal Circuit (CAFC) heard the Skylink appeal. In a remarkably far-reaching decision, the court effectively took many of the positions that Mr. Corley's lawyers had argued for in the DeCSS case, but they did so not to protect speech, but to protect competition. In fact, they implied that taking Chamberlain's side in the case would silently overrule the antitrust statutes. They also interpreted the new right created by the DMCA so as to add an implicit limitation. In their construction, merely gaining access is not illegal; only gaining access for the purpose of violating the copyright holders' rights violates the statute. The Reimerdes court had been willing to accept that the new access right allows a copyright holder to prohibit "fair uses as well as foul." When Chamberlain made the same argument as to their garage door opener program, the CAFC was incredulous. 114

Such an entitlement [as the one Chamberlain claims] would go far beyond the idea that the DMCA allows copyright owner to prohibit "fair uses . . . as well as foul." Reimerdes, 111 F. Supp. 2d at 304. Chamberlain's proposed construction would allow copyright owners to prohibit exclusively fair uses even in the absence of any feared foul use. It would, therefore, allow any copyright owner, through a combination of contractual terms and technological measures, to repeal the fair use doctrine with respect to an individual copyrighted work--or even selected copies of that copyrighted work.17 115

There are multiple ironies here. The CAFC rarely meets an intellectual property right it does not like. It has presided over a twenty-year expansion of American patent law that many scholars find indefensible. But when (for dubious jurisdictional reasons) it sorties beyond its traditional ambit of patent law, it is stunned by the potential expansiveness of the DMCA. Then there is the comparison with the Reimerdes case. How interesting that the First Amendment and concerns about free expression have comparatively little bite when applied to the DMCA, but antitrust and concerns about competition require that we curtail it. After all, the heart of Mr. Johansen's argument was that he had to write the DeCSS program in order to play his own DVDs on his own computer--to get access to his own DVDs, just as the purchaser of a replacement garage door control is getting access to the program that operates his own garage door. Indeed, Mr. Johansen's criticism of CSS was that it allowed the movie companies, "through a combination of contractual terms and technological measures, to repeal the fair use doctrine with respect to an individual copyrighted work." Mr. Corley echoed those claims. 116

Of course, the situations are not identical. The key limitation in Skylink is that the court saw no threat of "foul use." The Reimerdes court could see little else. On the other hand, the rulings are not easily reconciled. The Skylink court cannot imagine that Congress would want to give the copyright holder a new "property" right to prevent access unconnected to any underlying copyright violation. 117

As we have seen, Congress chose to create new causes of action for circumvention and for trafficking in circumvention devices. Congress did not choose to create new property rights. . . . Were we to interpret Congress's words in a way that eliminated all balance and granted copyright owners carte blanche authority to preclude all use, Congressional intent would remain unrealized. 118

Yet, arguably, that is exactly what the Reimerdes decision does, precisely because it focuses on enabling access alone, not access for the purpose of violating one of the rights of the copyright holder. The Reimerdes court saw a violation of the law just in cutting the wire or making a wire cutter. The Skylink court focused on whether the person cutting the wire was going to trespass once the cutting was done. In effect, the two courts disagree on which of the options offered to the legislature in the Farmers' Tale was actually enacted by Congress. Which court is correct? The Skylink decision strikes me as sensible. It also makes the statute constitutionally much more defensible--something that the Skylink court does not consider. But in the process, it has to rewrite the DMCA substantially. One should not presume that it will be this interpretation that will triumph. 119

SUMMING UP: EXAGGERATIONS, HALF-TRUTHS, AND BIPOLAR DISORDERS IN TECHNOLOGY POLICY 120

Let me return to the question with which I began the chapter. For many critics of contemporary intellectual property law, the DMCA is the very embodiment of all that is wrong. (I still cherish a friend's account of British protesters outside the American Embassy in London singing "D-M-C-A" to the tune of the Village People's "YMCA" and holding up signs calling for the law's repeal--to the great confusion of the diplomatic personnel.) The critics conjure up a digital apocalypse--a world of perfect control achieved through legally backed digital fences, in which both speech and competition suffer, and where citizens lose privacy, the privilege of fair use, and the right to criticize popular culture rather than simply consume it. In their view, the legal disaster is only exacerbated by bumbling judges who do not understand the technology and who are easily fooled by the doom-laden rhetoric of the content companies. The DMCA's supporters, on the other hand, think criticisms of the DMCA are overblown, that the dark tales of digital control are either paranoid delusions or tendentious exaggerations, and that far from being excessive, the DMCA's provisions are not sufficient to control an epidemic of illicit copying. More draconian intervention is needed. As for fair use, as I pointed out before, many of the DMCA's supporters do not think fair use is that important economically or culturally speaking. At best it is a "loophole" that copyright owners should have the right to close; certainly not an affirmative right of the public or a reserved limitation on the original property grant from the state. 121

Who is right? Obviously, I disagree profoundly with the DMCA's supporters. I wrote this book partly to explain--using Jefferson and Macaulay and the Sony case--what was wrong with their logic. It would be both convenient and predictable for me to claim that the DMCA is the intellectual property incarnation of the Antichrist. But it would not be true. In fact, I would not even put the DMCA in the top three of bad intellectual property initiatives worldwide. And many of the fears conjured up about it are indeed overblown. 122

Of course, the critics have a point. The DMCA is a very badly drafted law. As I have tried to show here, its key provisions were probably unnecessary and are, in my view, unconstitutional. If coupled with a number of other legal "innovations" favored by the content industry, the DMCA could play a very destructive role. In general, in fact, the Farmers' Tale is fairly accurate in describing both the origins of and the threats posed by the DMCA. Yet the single largest of those threats--the idea that the DMCA could be used to fence off large portions of the public domain and to make the fair use provisions of the Copyright Act essentially irrelevant--is still largely a threat rather than a reality. In some cases, fair use rights are curtailed. But for most citizens and for the majority of media, the DMCA has had relatively little effect. Digital rights management (DRM) certainly exists; indeed it is all around us. You can see that every time you try to play a DVD bought in another part of the world, open an Adobe eBook, or copy a song you have downloaded from iTunes. But so far, the world of legally backed digital rights management has not brought about the worst of the dystopian consequences that some people, including me, feared might result. 123

In many cases, citizens simply reject digital rights management. They will not buy products that use it. Attempts to introduce it into music CDs, for example, have been a resounding failure. In other cases, DRM has not been used in ways that the critics feared. There are genuine scandals, of course--cryptography research has been chilled, the DMCA has been turned to anticompetitive ends, and so on. It is also troubling to see federal judges issuing injunctions not only against banned material but also against those who link to the banned material. Somehow the blithe reassurance that this is consistent with the First Amendment fails to comfort one. But many of the evils prophesied for the DMCA remain as just that: prophecies. 124

There are also entries on the positive side of the ledger. The "safe-harbors" that the DMCA gave Internet service providers and search services have been a vital and positive force in the development of the Internet. It may even be true that in some cases, such as iTunes, the DMCA did what its backers claimed it would--encourage new provision of digital content by reassuring the record companies that they could put their music online surrounded by legally backed digital rights management. (Notably, however, the trend is now going the other way. Companies are coming to realize that many consumers prefer, and will pay more for, unprotected MP3 files.) 125

Of course, depending on your view of the music industry, that might seem like a mixed blessing. One might also wonder if the same consumer benefits might have been produced with a much less restrictive law. But with the exception of a few important areas--such as cryptography research, where its effects are reported to be severe--I would have to say that the criticisms focus too much on the DMCA, to the exclusion of the rest of the intellectual property landscape. Yes, the DMCA offers enormous potential for abuse, particularly in conjunction with some other developments in intellectual property that I will discuss later, but much of the abuse has not yet happened. Yet even if it never did happen, the DMCA has important lessons to teach us. 126

In this section I have tried to show how legal rules--particularly intellectual property rules--define the boundaries of legitimate competition. We used to assume that this was principally the function of patent and trademark law, less so of copyright. Of course, copyright would affect competition in publishing and in the TV and movie industries, but it hardly seemed central to competition policy in general. But once courts and legislatures accepted that software is copyrightable, that assessment changed. The levers and cogs of the machines of the modern economy are forged out of ones and zeros instead of steel and brass. In that situation, copyright is central to the competition policy of a high-tech economy. 127

As the Apple case shows, our moral intuitions about competition are going to be cloudier in the world of digital content and cyberspace. The same is true of the law. Even in the material world it can be hard to draw the line between the legitimate and ruthless pursuit of commercial advantage and various forms of unfair competition, antitrust violations, and so on. But in the immaterial world, the boundaries are even harder to draw. Is this the digital equivalent of trespass or legitimate passage on a public road that runs through your property? As I pointed out earlier, the constant analogies to physical property are likely to conceal as much as they reveal. Is this virtuous competitive imitation or illicit copying? We have strong, and by no means coherent, moral and legal intuitions about the answers to such questions. And our legal structure often gives us the raw material to make a very good case for both sides of the argument. 128

Into this already troubled situation, with a set of rules designed for original expression in novels and poems being applied to machines made of computer code, we add the DMCA and its new rights of uncertain extent. Copyright had a well- developed set of exceptions to deal with anticompetitive behavior. Where the existing exceptions did not function, courts tended to turn to fair use as the universal method for patching the system up--the duct tape of the copyright system. Without an evolving idea of fair use, copyright would overshoot its bounds as it was applied to new technologies and new economic conditions. Indeed that was the point of the Sony Axiom. The DMCA threw this system into disarray, into a war of competing metaphors. 129

The Skylink court sees monopolists being handed carte blanche to abolish the restraints on their monopolies. Competition policy demands that we construe the DMCA narrowly. The Reimerdes court sees a virus masquerading as speech, a digital pandemic that must be stopped at all costs by a draconian program of electronic public health. Each proceeds to construe the statute around the reality they have created. It is by no means certain which metaphor will win the day, still less which resolution will triumph in other countries that have passed versions of the DMCA. International attitudes toward speech, competition, and the necessary exceptions in a copyright system vary widely. Yet backed by the story of the Internet Threat, the content companies are already saying that we need to go further both nationally and internationally--introducing more technology mandates, requiring computers to have hardware that will only play approved copyrighted versions, allowing content companies to hack into private computers in search of material they think is theirs, and so on. Remember the suggestion from the beginning of the chapter, that all cars be assumed to be getaway vehicles for the felonious filchers of vegetables, and thus that they should be fitted with radio beacons, have the size of their cargo space reduced, and so on? The Farmers' Tale continues to evolve.