Now, in one sense, these lawsuits over trivial collections of hyperlinks and headlines might be seen as irrelevant. They may indicate we are handing out rights unnecessarily—did we really need a legal monopoly, and court involvement, to get someone to compile hyperlinks on a Web page? But it is hard to see social harm. As with the patents over "sealed crustless" peanut butter sandwiches or "methods of swinging on a swing," we may shake our heads at the stupidity of the system, but if the problems consist only of trivial creations, at least we are not likely to grieve because some vital piece of information was locked up. But we should not be so quick to declare such examples irrelevant. They tend to show that the system for drawing the boundaries of the right is broken—and that is of general concern, even if the issue at hand is not. 30
Finally, is the database right encouraging scientific innovation or hurting it? Here the evidence is merely suggestive. Scientists have claimed that the European database right, together with the perverse failure of European governments to take advantage of the limited scientific research exceptions allowed by the directive, have made it much harder to aggregate data, to replicate studies, and to judge published articles. In fact, academic scientific bodies have been among the strongest critics of database protection. But negative evidence, by its nature, is hard to produce; "show me the science that did not get done!" Certainly, both U.S. science and commerce have benefited extraordinarily from the openness of U.S. data policy. I will deal with this issue in the next part of this chapter. 31
If the United States does not give intellectual property protection to raw data, to facts, how is it that the database industry has managed to thrive here and to do better than in Europe, which has extremely strong protection? The economists described in Chapter 1 would surely tell us that this is a potential "public goods" problem. If it is hard to exclude others from the resource—it is cheap and easy to copy—and if the use of the resource is not "rival"—if I don't use up your facts by consulting them—then we ought to see the kind of dystopia economists predict. What would that consist of? First it might result in underproduction. Databases with a social value higher than their cost of creation would not get made because the creator could not get an adequate return on investment. In some cases it might even lead to the reverse—overproduction, where each party creates the database for itself. We get a social overinvestment to produce the resource because there is no legal right to exclude others from it. If you gave the first creator an intellectual property right over the data, they could sell to subsequent users at a price lower than their own cost to create the database. Everyone would win. But the United States did not give the intellectual property right and yet its database industry is flourishing. There are lots of commercial database providers and many different kinds of databases. How can this be? Is the economic model wrong? 32
The answer to that is no, the model is not wrong. It is, however, incomplete and all too often applied in sweeping ways without acknowledging that its basic assumptions may not hold in a particular case. That sounds vague. Let me give a concrete example. Westlaw is one of the two leading legal database providers and, as I mentioned before, one of the key proponents of creating intellectual property rights over unoriginal databases. (There is considerable question whether such a law would be constitutional in the United States, but I will pass over that argument for the moment.) Westlaw's "problem" is that much of the material that it provides to its subscribers is not covered by copyright. Under Section 105 of the U.S. Copyright Act, works of the federal government cannot be copyrighted. They pass immediately into the public domain. Thus all the federal court decisions, from district courts all the way up to the Supreme Court, all the federal statutes, the infinite complexity of the Federal Register, all this is free from copyright. This might seem logical for government-created work, for which the taxpayer has already paid, but as I will explain in the next section of the chapter, not every country adopts such a policy. 33
West, another Thomson subsidiary that owns Westlaw, publishes the standard case reporter series. When lawyers or judges refer to a particular opinion, or quote a passage within an opinion, they will almost always use the page number of the West edition. After all, if no one else can find the cases or statutes or paragraphs of an opinion that you are referring to, legal argument is all but impossible. (This might seem like a great idea to you. I beg to differ.) As electronic versions of legal materials became more prevalent, West began getting more competition. Its competitors did two things that West found unforgivable. First, they frequently copied the text of the cases from West's electronic services, or CD-ROMs, rather than retyping them themselves. Since the cases were works of the federal government, this was perfectly legal provided the competitors did not include West's own material, such as summaries of the cases written by its employees or its key number system for finding related issues. Second, the competitors would include, within their electronic editions, the page numbers to West's editions. Since lawyers need to cite the precise words or arguments they are referring to, providing the raw opinion alone would have been all but useless. Because West's page numbers were one of the standard ways to cite case opinions, competitors would indicate where the page breaks on the printed page would have been, just as West did in its own databases. 34
West's reaction to all of this was exactly like Apple's reaction in the story I told in Chapter 5 about the iPod or like Rural's reaction to the copying of its phone directory. This was theft! They were freeloading on West's hard work! West had mixed its sweat with these cites, and so should be able to exclude other people from them! Since it could not claim copyright over the cases, West claimed copyright over the order in which they were arranged, saying that when its competitors provided its page numbers for citation purposes, they were infringing that copyright. 35
In the end, West lost its legal battles to claim copyright over the arrangement of the collections of cases and the sequence in which they were presented. The Court held that, as with the phone directory, the order in which the cases were arranged lacked the minimum originality required to sustain a copyright claim.4 At this stage, according to the standard public goods story, West's business should have collapsed. Unable to exclude competitors from much of the raw material of its databases, West would be undercut by competitors. More importantly, from the point of view of intellectual property policy, its fate would deter potential investors in other databases—databases that we would lose without even knowing they could have been possible. Except that is not the way it turned out. West has continued to thrive. Indeed, its profits have been quite remarkable. How can this be? 36
The West story shows us three ways in which we can leap too quickly from the abstract claim that some information goods are public goods—nonexcludable and nonrival—to the claim that this particular information good has those attributes. The reality is much more complex. Type www.westlaw.com into your Internet browser. That will take you to the home page of West's excellent legal research service. Now, I have a password to that site. You probably do not. Without a password, you cannot get access to West's site at all. To the average consumer, the password acts as a physical or technical barrier, making the good "excludable"—that is, making it possible to exclude someone from it without invoking intellectual property rights. But what about competitors? They could buy access and use that access to download vast quantities of the material that is unprotected by copyright. Or could they? Again, West can erect a variety of barriers, ranging from technical limits on how much can be downloaded to contractual restrictions on what those who purchase its service can do ("No copying every federal case," for example). 37
Let's say the competitor somehow manages to get around all this. Let's say it somehow avoids copying the material that West does have a copyright over—such as the headnotes and case synopses. The competitor launches their competing site at lower prices amidst much fanfare. Do I immediately and faithlessly desert West for a lower-priced competitor? Not at all. First of all, there are lots of useful things in the West database that are covered by copyright—law review articles and certain treatises, for example. The competitor frequently cannot copy those without coming to the same sort of agreements that West has with the copyright holders. For much legal research, that secondary material is as important as the cases. If West has both, and the competitor only one, I will stick with West. Second, West's service is very well designed. (It is only their copyright policies I dislike, not the product.) If a judge cites a law review article in a case, West will helpfully provide a hyperlink to the precise section of the article she is referring to. I can click on it and in a second see what the substance of the argument is. The reverse is true if a law review article cites a statute or a case. Cases have "flags" on them indicating whether they have been overruled or cited approvingly in subsequent decisions. In other words, faced with the competitive pressure of those who would commoditize their service and provide it at lower cost, West has done what any smart company would: added features and competed by offering a superior service. Often it has done so by "tying" its uncopyrightable data structures to its huge library of copyrighted legal material. 38
The company that challenged Westlaw in court was called Hyperlaw. It won triumphantly. The courts declared that federal cases and the page numbers in the West volumes were in the public domain. That decision came in 1998 and Westlaw has lobbied hard since then to reverse it by statute, to create some version of the Database Directive in the United States. To date, they have failed. The victor, Hyperlaw, has since gone out of business. Westlaw has not. 39