Search

Existing foundation models are trained on copyrighted material. Deploying these models can pose both legal and ethical risks when data creators fail to receive appropriate attribution or compensation. In the United States and several other countries, copyrighted content may be used to build foundation models without incurring liability due to the fair use doctrine. However, there is a caveat: If the model produces output that is similar to copyrighted data, particularly in scenarios that affect the market of that data, fair use may no longer apply to the output of the model. In this work, we emphasize that fair use is not guaranteed, and additional work may be necessary to keep model development and deployment squarely in the realm of fair use. First, we survey the potential risks of developing and deploying foundation models based on copyrighted content. We review relevant U.S. case law, drawing parallels to existing and potential applications for generating text, source code, and visual art. Experiments confirm that popular foundation models can generate content considerably similar to copyrighted material. Second, we discuss technical mitigations that can help foundation models stay in line with fair use. We argue that more research is needed to align mitigation strategies with the current state of the law. Lastly, we suggest that the law and technical mitigations should co-evolve. For example, coupled with other policy mechanisms, the law could more explicitly consider safe harbors when strong technical tools are used to mitigate infringement harms. This co-evolution may help strike a balance between intellectual property and innovation, which speaks to the original goal of fair use. But we emphasize that the strategies we describe here are not a panacea and more work is needed to develop policies that address the potential harms of foundation models.

It has long been recognized in Europe and elsewhere that standards-development organizations (SDOs) may adopt policies that require their participants to license patents essential to the SDO’s standards (standards-essential patents or SEPs) to manufacturers of standardized products (“implementers”) on a royalty-free (RF) basis.1 This requirement contrasts with SDO policies that permit SEP holders to charge implementers monetary patent royalties, sometimes on terms that are specified as “fair, reasonable and nondiscriminatory” (FRAND). As demonstrated by two decades of intensive litigation around the world, FRAND royalties have given rise to intractable disputes regarding the manner in which such royalties should be calculated and adjudicated. In contrast, standards distributed on an RF basis are comparatively free from litigation and the attendant transaction costs.2 Accordingly, numerous SDOs around the world have adopted RF licensing policies and many widely adopted standards, including Bluetooth, USB, IPv6, HTTP, HTML and XML, are distributed on an RF basis. This note briefly discusses the commercial considerations surrounding RF standards, the relationship between RF standards and open source software (OSS) and the SDO policy mechanisms – including “universal reciprocity” -- that enable RF licensing to succeed in the marketplace.

On 23 November 2017, the European Parliament, published a Working Document titled Three-Dimensional Printing, a Challenge in the fields of Intellectual Property Rights and Civil Liability. The publication went on to state that the European Commission ‘has made 3D printing one of the priority areas of technology’. The technology was also referred to in the European Commission’s recent reflection paper on ‘harnessing globalisation’ stating that 3D printing, amongst other emerging technologies, ‘will revolutionise how we produce, work, move and consume’ (emphasis added). In April 2018, the European Commission demonstrated its commitment to exploring the intellectual property implications of industrial three-dimensional printing by commissioning research in to this area, in a move to shape policy. In a further publication titled, The Next Production Revolution: Implications for Governments and Businesses, it was emphasised that ‘long-term thinking is essential’ for technologies such as 3D printing. The report went on to say that ‘in addition to addressing short-term challenges, leaders in business, education, unions and government must be ready to frame policies’ and reflect on ‘how policy priorities may need to evolve, in fields as diverse as the intellectual property system, competition and trade policies, and the distributional implications of future production’. As such, 3D printing, presents various challenges for the legal, ethical , medical and health and safety sectors. Amongst these various concerns, the report makes reference to the impact of 3D printing on intellectual property rights (IPRs). The imminent change and the need for responsive policy, particularly in the area of IPRs has been acknowledged and echoed in a number of countries, including the United Kingdom (UK), the United States of America (USA) and Australia – the jurisdictions which are the focus of this book. This introductory chapter is set out as follows. It commences with a brief introduction to 3D printing, additive manufacturing and the more recent phenomenon of the maker movement. Thereafter, it moves on to setting out the reasons for selecting UK, USA and Australia as a basis for consideration, before providing a brief synopsis of some of the intellectual property challenges addressed in the book. It concludes with an overview of the layout of the book.

In a 2006 short story, ‘Printcrime’, Cory Doctorow imagined a dystopian future of contraband 3D printers. In the work, police try to shut down a bootleg operation, which engaged in the 3D printing of intellectual property. In his 2009 novel Makers, Cory Doctorow explored the rise of the maker community, and its do-it-yourself ethic. In an interview about the novel, the author reflected: "There has never been a better time to be a maker because finding the people who know how to fix the thing that's broken has never been easier. Finding someone else who has done 80% of what you want to do, and sharing the things you have done with other people, has never been easier. A maker is someone who is of and in the 21st century." Rather prophetically, he discussed the prospect of intellectual property conflicts around 3D printing (particularly around copyright infringement and trademark infringement), and future controversies over 3D printing guns. In his 2015 short story, ‘The Man Who Sold the Moon’, Cory Doctorow imagined 3D printing in space. This body of creative work has been an important inspiration for the Maker Movement – but it has also shown a critical engagement with the law, ethics, and public policy associated with 3D printing and additive manufacturing. Inspired by such science fiction, there have since been a number of optimistic, utopian manifestos published on the topic of 3D printing and the rise of the Maker Movement. There has been high hopes that the emerging, disruptive technology will be part of a new industrial revolution. The founder and executive chairman of the World Economic Forum, Klaus Schwab, situates 3D printing within the framework of a fourth industrial revolution. He predicted: ‘As current size, cost and speed constraints are progressively overcome, 3D printing will become more pervasive to include integrated electronic components such as circuit boards and even human cells and organs.’ Schwab anticipated that there would be a ‘new generation of self-altering products capable of responding to environmental changes such as heat and humidity.’ Moreover, he expected that ‘this technology could be used in clothing or footwear, as well as in health-related products such as implants designed to adapt to the human body.’ Schwab placed 3D printing alongside autonomous vehicles, advanced robotics, and new material as physical manifestations of larger technological megatrends. In this context, this collection provides a sober, critical evaluation of the legal, ethical, and public policy issues in respect of intellectual property, innovation law, and 3D printing. Building upon Mark Lemley’s chapter, ‘IP in a World Without Scarcity,’ it considers the legal opportunities and challenges of the Maker Revolution. It provides both theoretical and empirical insights in respect of 3D printing, intellectual property, innovation, and regulation.

Traditional justifications for patents are all based on direct or indirect contribution to the creation of new products. Patents serve the social interest if they provide not just invention, but innovation the world would not otherwise have. Non-practicing entities (“NPEs”) as well as product-producing companies can sometimes provide such innovation, either directly, through working the patent or transferring technology to others who do, or indirectly, when others copy the patented innovation. The available evidence suggests, however, that patent licensing demands and lawsuits from NPEs are normally not cases that involve any of these activities. Some scholars have argued that patents can be valuable even without technology transfer because the ability to exclude others from the market may drive commercialization that would not otherwise occur. We demonstrate that even if various commercialization theories can sometimes justify patent protection, they cannot justify most NPE lawsuits or licensing demands.

BackgroundEjection fraction (EF) reserve has been found to be a useful adjunct for identifying high risk coronary artery disease in cardiac positron emission tomography (PET). We aimed to evaluate EF reserve obtained from technetium-99m sestamibi (Tc-99m) high-efficiency (HE) SPECT.MethodsFifty patients (mean age 69 years) undergoing regadenoson same-day rest (8-11 mCi)/stress (32-42 mCi) Tc-99m gated HE SPECT were enrolled. Stress imaging was started 1 minute after sequential intravenous regadenoson .4 mg and Tc-99m injections, and was composed of five 2 minutes supine gated acquisitions followed by two 4 minutes supine and upright images. Ischemic total perfusion deficit (ITPD) ≥5 % was considered as significant ischemia.ResultsSignificantly lower mean EF reserve was obtained in the 5th and 9th minute after regadenoson bolus in patients with significant ischemia vs patients without (5th minute: -4.2 ± 4.6% vs 1.3 ± 6.6%, P = .006; 9th minute: -2.7 ± 4.8% vs 2.0 ± 6.6%, P = .03).ConclusionsNegative EF reserve obtained between 5th and 9th minutes of regadenoson stress demonstrated best concordance with significant ischemia and may be a promising tool for detection of transient ischemic functional changes with Tc-99m HE-SPECT.

In recent years, the United States Supreme Court has focused increasing attention on two doctrines that provide immunity from antitrust liability for certain anticompetitive activity: the state action doctrine and the petitioning immunity doctrine (sometimes known as the Noerr-Pennington doctrine, after the two cases that established it). These doctrines have been the subject of seven Supreme Court decisions in as many years. In spite of (or perhaps because of) the Court’s numerous recent decisions, there remains a great deal of confusion about the source and the scope of these doctrines. This Article attempts to clarify both doctrines. The Supreme Court and a number of commentators contend that the antitrust immunity doctrines are the product of statutory interpretation of the antitrust laws themselves. The Court contends that petitioning and state action are “essentially dissimilar” to the types of business activity the antitrust laws were designed to regulate. This Article disagrees. Both petitioning and state action present precisely the sorts of problems with which the antitrust laws are concerned — exploitation of consumers through the charging of supracompetitive prices. To determine the source of antitrust immunity, the Court must look beyond the antitrust laws to the constitutional principles that are implicated by the doctrines. For the state action doctrine, the constitutional principle at stake is largely one of federalism, and the more general democratic principles embodied in the Court’s non-delegation jurisprudence. For the petitioning immunity doctrine, the First Amendment protection of speech and petitioning provides the relevant principles. After examining the source of the antitrust immunity doctrines, this Article considers the appropriate scope of those doctrines in light of the constitutional principles at issue.

In the last fifteen years, the primary means of legal protection for computer software has shifted from copyright to patent. We argue that one unanticipated effect of this trend may be to encourage software reuse. Traditionally, computer programmers have reinvented software components, coding programs from scratch each time a new one is desired rather than buying and reusing existing components. This process is inefficient, and is in stark contrast to the normal practice in other engineering disciplines. We argue that copyright law encourages reinvention and discourages the development of a market for tradeable software components because it allows competitors to appropriate the value of a new software invention without payment to the original developer of that invention, but forbids competitors from copying the computer code implementing that invention. The result is that competitors take inventions from others, but write their own code to implement those inventions. By contrast, patent law gives strong protection to inventions, forcing competitors to license the patent in order to make any product incorporating the idea. It is reasonable to expect that one effect of increasing reliance on patent law will be an increase in licenses of both the patented idea and the implementing code. This licensing should in turn pave the way for the trading and reuse of software components. Note: This paper was published in its final form at 49 Stan. L. Rev. 255 (1997).

The Supreme Courts decision in eBay v. MercExchange, which revolutionized the granting of injunctions in patent cases, has increasingly been applied to trademark cases as well. But courts applying eBay to trademark cases have ignored some fundamental differences between patent and trademark law, including the difficulty of quantifying injury to brand reputation, the rarity of damage awards in trademark cases, and the presence of a third party -- consumers -- with a strong interest in avoiding confusion. Some courts, particularly in the Ninth Circuit, have also turned eBay from a set of guidelines for case-by-case analysis of the equities of an injunction into a hard-and-fast set of requirements for injunctive relief. eBay should apply to trademark cases. But it must be applied with sensitivity to the very different contexts of patent and trademark. Doing so will make it more likely that trademark owners will obtain injunctions in "core" trademark infringement cases. Nonetheless, courts may still properly use eBay to deny injunctive relief in dilution and association cases where the evidence of harm to the mark owner or consumers is weak or nonexistent.

Virtually all the courts to consider non-literal infringement of software copyrights have lined up with the “narrow constructionists,” engaging in “analytic dissection” of computer programs in order to determine whether any copyrightable expression has actually been copied. Most commonly, this analytic dissection has taken the form of the “abstraction-filtration-comparison” test set forth in Computer Associates v. Altai. While there are still a few courts in which the “total concept and feel” approach remains the law, the approach is moribund: since Altai was decided, no court has endorsed the broader “total concept and feel” approach. Rather than ending, the debate over software copyright law appears to be shifting its focus. Having finally resolved the debate that has been plaguing software copyright law since its inception, courts are discovering to their chagrin, that deciding what test to apply actually tells you very little about how to apply that test. Despite the convergence of courts on Altais filtration approach, courts remain fundamentally conflicted in deciding how broadly to protect software copyright. Further, there remains a good deal of misunderstanding about what exactly it means to “abstract” and “filter” a computer program. I suggest a unified approach to evaluating non-literal infringement in software copyright cases. This approach focuses on exactly what is alleged to have been copied. It also acknowledges the increasing role of patent law in protecting computer software, and the role of other copyright concerns such as compatibility and fair use. The result of this unified approach is to provide relatively narrow copyright protection for computer programs in most cases of non-literal infringement.

Copyright owners have persuaded the courts that they should win cases in which a defendants use doesnt injure their market directly, but in which they could and would have charged a fee to grant permission for the use. Even assuming courts are right to have accepted this argument, it is unreasonable to then give the copyright owner not just the fee they would have charged but the power to prevent the use altogether or to collect damages far in excess of that fee. Licensing market cases are excellent choices for separating compensation and control, giving copyright owners the right to get paid without giving them control over transformative uses. Doing so is harder than simply denying injunctive relief, however. It requires us to rethink our definition of damages in copyright law with the aim of remedying injury rather than always seeking to deter infringement.

Michele Boldrin and David Levine offer a strong attack on intellectual property (IP), which they call “intellectual monopoly.” In their view, IP is not necessary to encourage invention or creation. Quite the contrary, they argue that we get innovation from competition, not monopoly. Further, because monopoly imposes well-recognized social costs, we are better off without it if it doesn’t in fact spur new innovation. Boldrin and Levine make a plausible case on their own terms. Nonetheless, I think their terms are misleading. IP rights are rarely if ever “intellectual monopolies.” Most patents, to say nothing of most copyrights, create no economic rents. What this means is that we can’t assume that IP rights generally impose deadweight losses on society. They cause deviation from atomistic, perfect competition, but they don’t cause monopoly pricing. With a small number of exceptions, therefore, they don’t cause the social harms Boldrin and Levine correctly associate with monopoly pricing.

Current debates over the autonomy of virtual worlds have an eeriesimilarity to discussions about the independence of cyberspace two decadesago. The history of the Internet offers some important lessons for how thelaw will affect virtual worlds, and how it should do so.

Judges, lawyers, and scholars have long decried the high reversal rate district judges face in patent cases. Many have suggested greater district court specialization as a solution, and Congress in 2011 enacted legislation to promote such specialization. In this paper, we investigate the impact of a novel measure of experience – whether a district court judge has sat by designation on a Federal Circuit panel in a patent claim construction appeal – on the likelihood a district judge’s subsequent claim constructions are reversed. Before sitting by designation, judges who later do so actually have a slightly higher claim construction reversal rate than judges who never do so. After sitting by designation, the reversal rate of district court judges on subsequent claim construction appeals decreases by 50 percent. This decrease is not fully explained by other measures of experience, including the number of prior patent cases or years on the bench. Nor is it fully explained by the timing of the appeal, the particular district court judge or various other characteristics of the patents, the parties and the litigation. Our results suggest a simple way to reduce the reversal rate in patent and perhaps other sorts of cases. However, our evidence suggests this increased agreement is due to increased Federal Circuit trust in the decisions of individual judges who have sat by designation and not increased district judge understanding of claim construction.

The case for patent enforcement driving innovation is far from clear. Butpatents may serve myriad other purposes besides excluding competitors. Weshould focus more of our attention on how patents are used outsidetraditional litigation and licensing.

The new twenty-year patent term has engendered considerable controversy. A number of inventors have complained that the new patent term may reduce patent protection because some applications spend a number of years in “prosecution” before the PTO. Under the old law, delay in processing an application did not hurt the patentee, but under the new rule, each day spent in prosecuting the patent is a day of protection lost. The purpose of this study is to evaluate in a neutral and systematic way the likely effects of the new law. I conclude that on average, and for most industries, the new law gives more protection to patentees than the old law. However, there is some question as to whether the biotechnology industry will receive less protection under the new law. I conclude that there is no significant relationship between the length of time a patent spends in prosecution and whether it will be judged valid in the courts. Finally, the study produces some interesting data about the problem of “submarine patents.”

The Economic Espionage Act of 1996 was intended to address both the general need for a federal criminal deterrent against trade secret theft as well as the apparent threat from foreign state-sponsored industrial espionage. This article examines the background of this new law, provides critical analysis of its most important terms, describes the process involved in a reference of trade secret theft to the federal authorities, and suggests practical strategies for businesses to take advantage of this resource as well as to avoid exposure to liability for the mishandling of information belonging to others.

Economists and legal scholars have debated the reasons people adopt open source software, and accordingly whether and to what extent the open source model can scale, replacing proprietary rights as a primary means of production. In this study, we use the release by a biotechnology company of similar software under both proprietary and open source licenses to investigate who uses open source software and why. We find that academic users are somewhat more likely to adopt open source software than private firms. We find only modest differences in the willingness of open source users to modify or improve existing programs. And we find that users of open source software often make business decisions that seem indifferent to the norms of open source distribution. Our findings cast some doubt on the penetration of the open source ethos beyond traditional software markets.

In this short piece we take a first look at the Supreme Court's decisionsin Halo and Stryker, identifying some of the questions that remainunanswered in patent treble damages, including the role of willfulness, therole of opinion letters, and the role of the jury.

The patent system seems in the midst of truly dramatic change. The last twenty years have seen the rise of a new business model – the patent troll – that grew to become a majority of all patent lawsuits. They have seen a significant expansion in the number of patents granted and a fundamental change in the industries in which those patents are filed. They have seen the passage of the most important legislative reform in the last sixty years, a law that reoriented legal challenges to patents away from courts and toward the Patent and Trademark Office (PTO). And they have seen remarkable changes in nearly every important legal doctrine, from patent eligibility to obviousness to infringement to remedies. These changes have prompted alarm in a number of quarters. From the 1990s to the 2000s, as the number of patents and patent troll suits skyrocketed, technology companies and academics worried about the “crisis” in the patent system – a crisis of overprotection that might interfere with rather than promote innovation. By 2015, as patent reform took effect and the Supreme Court undid many of the Federal Circuit’s expansions of patent rights, it was patent owners who were speaking of a crisis in the patent system – a crisis of underprotection that might leave innovators without adequate protection. Depending on one’s perspective, then, the sky seems to have been falling on the patent system for some time. Despite the undeniable significance of these changes in both directions, something curious has happened to the fundamental characteristics of the patent ecosystem during this period: very little. Whether we look at the number of patent applications filed, the number of patents issued, the number of lawsuits filed, the patentee win rate in those lawsuits, or the market for patent licenses, the data show very little evidence that patent owners and challengers are behaving differently because of changes in the law. The patent system, then, seems surprisingly resilient to changes in the law. This is a puzzle. In this article, I document this phenomenon and give some thought to why the fundamental characteristics of the patent system seem resistant to even major changes in patent law and procedure. The results pose some profound questions not only for efforts at patent reform but for the role of the patent system in society as a whole.

In a string of recent opinions, the Supreme Court has made it harder for consumers to avoid arbitration clauses, even when businesses strategically insert provisions in them that effectively prevent consumers from being able to bring any claim in any forum. In American Express Co. v. Italian Colors Restaurant, an antitrust case, the Court held that class-action waivers embedded in mandatory arbitration clauses were enforceable even when they had the effect of making it economically irrational for the victims of antitrust violations to pursue their claims. Courts have long considered antitrust claims to be too complex and too important to trust to private arbitrators. By the 1980s, the Supreme Court permitted federal statutory rights, including antitrust claims, to be arbitrated so long as the plaintiffs could effectively vindicate their rights in the alternative forum. In 2013, the Supreme Court in Italian Colors fundamentally weakened the Effective Vindication Doctrine when it held that arbitration clauses that precluded class actions and classwide arbitration were enforceable even when they effectively prohibited all individual plaintiffs from bringing a case. Arbitration differs from litigation in ways that harm the interests of consumer antitrust plaintiffs. For example, arbitration limits discovery and has no meaningful appeals process. Furthermore, defendants use the terms in arbitration clauses to prevent class actions and to undercut the pro-plaintiff features of antitrust law, including mandatory treble damages, meaningful injunctive relief, recovery of attorneys’ fees, and a lengthy statute of limitations. With the Court’s undermining of the Effective Vindication Doctrine in Italian Colors, defendants’ efforts to dismantle these pro-plaintiff components of antitrust law may prove more successful in the future. The problems associated with antitrust arbitration are magnified in concentrated markets. Supporters of enforcing arbitration clauses assume that they these contractual provisions are the result of an informed, voluntary bargain. But when a market is dominated by a single supplier or a small group of firms, consumers often find it impossible to purchase a necessary product while retaining the right to sue, especially since arbitration clauses are generally embedded in contracts of adhesion. This means that in the markets most likely to be affected by antitrust violations, consumers are least likely to be able to avoid mandatory arbitration clauses. Furthermore, when mergers result in concentrated markets, they can increase the problems explored in Part Two. Antitrust authorities can address the problem of proliferating arbitration clauses. When evaluating mergers, officials at the Federal Trade Commission and the Antitrust Division of the Department of Justice can threaten to challenge the merger unless the merging parties agree to specified conditions, such as the divestiture of certain assets. Because those mergers that pose the greatest risk of anticompetitive effects also magnify the problems associated with mandatory arbitration clauses, antitrust officials would be wise to condition merger approval on the merging parties’ agreement to not require arbitration of antitrust claims.

How can we fix the PTO, allowing examiners to effectively distinguishbetween patentable and unpatentable inventions, without slowing the processto a crawl or wasting a bunch of money? This essay reviews the recentliterature and considers a number of proposals and their problems. Itconcludes that we can make the PTO better, but are unlikely to solve theproblem of bad patents.

This paper summarizes the most important patent cases in the past 12months, from May 2015 through April 2016.

The Supreme Court and the Federal Circuit have repeatedly emphasized the public interest in testing the validity of patents, weeding out patents that should not have been issued. But there is one important group of people the law systematically prevents from challenging bad patents. Curiously, it is the very group patent law is supposed to support: inventors themselves. The century-old doctrine of assignor estoppel precludes inventors who file patent applications from later challenging the validity or enforceability of the patents they receive. The stated rationale for assignor estoppel is that it would be unfair to allow the inventor to benefit from obtaining a patent and later change her tune and attack the patent when it benefits her to do so. The Supreme Court has traditionally disfavored the doctrine, reading it narrowly. But the Federal Circuit has expanded the doctrine in a variety of dimensions, and applied it even when the benefit to the inventor is illusory. Further, the doctrine misunderstands the role of inventor-employees in the modern world. More important, the expansive modern form of assignor estoppel interferes substantially with employee mobility. Inventors as a class are put under burdens that we apply to no other employee. If they start a company, or even go to work for an existing company in the same field, they will not be able to defend a patent suit from their old employer. The result is a sort of partial noncompete clause, one imposed without even the fiction of agreement and one that binds anyone the inventor comes in contact with after leaving the job. Abundant evidence suggests that noncompetes in general retard innovation and economic growth, and several states prohibit them outright, while all others limit them. But assignor estoppel is a federal law doctrine that overrides those state choices. It is time to rethink the doctrine of assignor estoppel. I describe the doctrine, its rationale, and how it has expanded dramatically in the past twenty-five years. I argue that the doctrine is out of touch with the realities of both modern inventing and modern patent law, and that it interferes with both the invalidation of bad patents and the goal of employee mobility. Should the Supreme Court take up the doctrine, it is unlikely to survive in its current form. Rather, it should—and will—return to its much more limited roots.

Much of the policy debate over the patent system has focused on the perceived problems with non-practicing entities (NPEs), also called patent trolls. Drawing on a comprehensive data set we built of every patent lawsuit filed in 2008 and 2009 that resulted in a ruling on the merits, we find that the situation is rather more complicated than simply operating companies vs. NPEs. While operating companies fare better in litigation than NPEs overall, breaking NPEs into different categories reveals more complexity. Patent Assertion Entities (PAEs) in particular win very few cases. Further, once we remove certain pharmaceutical cases from the mix, no patent plaintiff fares very well. That is particularly true of software, computer, and electronics patents.

In a string of recent opinions, the Supreme Court has made it harder for consumers to avoid arbitration clauses, even when businesses strategically insert provisions in them that effectively prevent consumers from being able to bring any claim in any forum. In American Express Co. v. Italian Colors Restaurant, an antitrust case, the Court held that class-action waivers embedded in mandatory arbitration clauses were enforceable even when they had the effect of making it economically irrational for the victims of antitrust violations to pursue their claims. Courts have long considered antitrust claims to be too complex and too important to trust to private arbitrators. By the 1980s, the Supreme Court permitted federal statutory rights, including antitrust claims, to be arbitrated so long as the plaintiffs could effectively vindicate their rights in the alternative forum. In 2013, the Supreme Court in Italian Colors fundamentally weakened the Effective Vindication Doctrine when it held that arbitration clauses that precluded class actions and classwide arbitration were enforceable even when they effectively prohibited all individual plaintiffs from bringing a case. Arbitration differs from litigation in ways that harm the interests of consumer antitrust plaintiffs. For example, arbitration limits discovery and has no meaningful appeals process. Furthermore, defendants use the terms in arbitration clauses to prevent class actions and to undercut the pro-plaintiff features of antitrust law, including mandatory treble damages, meaningful injunctive relief, recovery of attorneys’ fees, and a lengthy statute of limitations. With the Court’s undermining of the Effective Vindication Doctrine in Italian Colors, defendants’ efforts to dismantle these pro-plaintiff components of antitrust law may prove more successful in the future. The problems associated with antitrust arbitration are magnified in concentrated markets. Supporters of enforcing arbitration clauses assume that they these contractual provisions are the result of an informed, voluntary bargain. But when a market is dominated by a single supplier or a small group of firms, consumers often find it impossible to purchase a necessary product while retaining the right to sue, especially since arbitration clauses are generally embedded in contracts of adhesion. This means that in the markets most likely to be affected by antitrust violations, consumers are least likely to be able to avoid mandatory arbitration clauses. Furthermore, when mergers result in concentrated markets, they can increase the problems explored in Part Two. Antitrust authorities can address the problem of proliferating arbitration clauses. When evaluating mergers, officials at the Federal Trade Commission and the Antitrust Division of the Department of Justice can threaten to challenge the merger unless the merging parties agree to specified conditions, such as the divestiture of certain assets. Because those mergers that pose the greatest risk of anticompetitive effects also magnify the problems associated with mandatory arbitration clauses, antitrust officials would be wise to condition merger approval on the merging parties’ agreement to not require arbitration of antitrust claims.

In 2011, Congress enacted the America Invents Act (AIA), the most substantial overhaul of the patent system in the past sixty years. The most significant change in the AIA was the move from a first to invent regime to a first inventor to file regime. The goal of the move to first to file, besides harmonization, is to encourage inventors to move with alacrity to share their invention with the world. There is an ambiguity in the AIA, however, that threatens that disclosure objective. Some commentators have argued that Congress intended to fundamentally change the rules of prior art in a way that would encourage secrecy rather than disclosure. Under this interpretation of the new law, an inventor can use its process in secret for commercial purposes, potentially forever, and still file a patent on that invention at some point in the future. Far from encouraging disclosure, on this interpretation the effect of the AIA is to encourage secrecy and delay in patenting. Curiously, the argument is that Congress signaled its intent to make this fairly radical change by re-enacting language that had been in the Patent Act for the last 140 years: the words "public use." Because two of these commentators, Bob Armitage and Joe Matal, were involved in the drafting of the AIA, this argument has carried substantial weight, and the PTO in 2013 adopted regulations that read the term "public use" in the AIA as meaning something completely different than it had for the century before 2011. In this paper, I make two points. First, as a matter of statutory interpretation it is unlikely that Congress intended to make such a change, not only because they readopted existing statutory language but because other parts of the statute make no sense under such an interpretation. Second, reading the AIA as making such a change would be unwise as a policy matter, not only because it would encourage secrecy but because it would undermine confidence that other terms reenacted in the AIA have the same meaning they have accrued in decades of common law.

The patent statutes were wisely drafted with an expansive vision of patentable subject matter. Efforts to graft judicially created limitations onto that expansive scope in the past have proven fruitless and indeed counterproductive. In deciding Bilski v. Doll, the Supreme Court should not impose a requirement that patentable inventions require a machine or the physical transformation of some material. It should instead maintain the rule that patents are available for "anything under the sun made by man," including discoveries of ideas, laws of nature, or natural phenomena, so long as they are implemented in a practical application. In short, the test should be as it has been: where an idea is claimed as applied, it is eligible for patentability, but if it is claimed merely in the abstract it is not.

This Supreme Court amicus brief, filed in Federal Trade Commission v. Watson, explains why exclusion-payment settlements, by which brand-name drug companies pay generic firms to delay entering the market, contravene the policies of patent law, antitrust law, and the Hatch-Waxman Act. It addresses five points. First, the settlements are not consistent with the Hatch-Waxman Act, Congress’s framework for balancing patent and antitrust law in the pharmaceutical industry, which encouraged generics to challenge patents. Second, the settlements are anticompetitive, serving as a form of market division, which is the practical result when brands pay generics to drop challenges to weak patents and delay entering the market instead. Third, the mere fact of a patent cannot justify the payments. The Patent Office frequently issues invalid patents, and the patents at the heart of these settlements present concern, often covering not the drug’s active ingredient but narrower aspects like the formulation or method of use that are less innovative and bear more potential for anticompetitive mischief. Patent policy encourages challenges to weak patents, and the procedural presumption of validity does not justify the settlements. Fourth, exclusion payments are not needed to settle cases in the public interest; history has shown that brands and generics can reach settlements without them. Fifth, the most appropriate antitrust framework employs a “quick look” rule-of-reason analysis that treats exclusion payments as presumptively unlawful. Such a framework recognizes the potentially severe anticompetitive effects of exclusion-payment settlements while permitting the settling parties to introduce possible procompetitive justifications, if any, for their agreement.

Antitrust law promotes competition in the service of economic efficiency. Government regulation may or may not promote either competition or efficiency, depending on both the goals of the agency and the effects of industry "capture." Antitrust courts have long included regulated industries within their purview, working to ensure that regulated industries could not use the limits that regulation imposes on the normal competitive process to achieve anticompetitive ends. Doing so makes sense; an antitrust law that ignored anticompetitive behavior in any regulated industry would be a law full of holes. The role of antitrust in policing regulated industries appears to be changing, however. A cluster of Supreme Court decisions in the past decade have fundamentally altered the relationship between antitrust and regulation, placing antitrust law in a subordinate relationship that, some have argued, requires it to defer not just to regulatory decisions but perhaps even to the silence of regulatory agencies in their areas of expertise. Absolute antitrust deference to regulatory agencies makes little sense as a matter either of economics or experience. Economic theory teaches that antitrust courts are better equipped than regulators to assure efficient outcomes in many circumstances. Public choice theory - and long experience - suggests that agencies that start out trying to limit problematic behavior by industries often end up condoning that behavior and even insulating those industries from market forces. And as history has shown, relying on regulatory oversight alone without the backdrop of antitrust law would leave both temporal and substantive gaps in enforcement, which unscrupulous competitors could exploit to the clear detriment of consumers. The mere existence of a competition-conscious regulatory structure cannot guarantee against abuses of that structure, or against exclusionary behavior that falls just beyond its jurisdiction. Indeed - and perhaps ironically - the very regulatory structure that exists to promote competition can create gaming opportunities for competitors bent on achieving anti-competitive goals. Such "regulatory gaming" undermines both the regulatory system itself and the longstanding complementary relationship between regulatory and antitrust law. We argue that the risk of regulatory gaming provides an important example of the need for ongoing antitrust oversight of regulated industries. We define regulatory gaming as private behavior that harnesses pro-competitive or neutral regulations and uses them for exclusionary purposes. We identify three possible instances of regulatory gaming: (1) product-hopping, in which the branded company makes repeated changes in drug formulation to prevent generic substitution, rather than to improve the efficacy of the drug product; (2) manipulation of government standard-setting organizations; and (3) claims of price squeezes by partially regulated industries. Our goal in this paper is not to persuade the reader that these particular examples of regulatory gaming do or do not violate the antitrust laws. Rather, our point is that whether or not particular acts of regulatory gaming harm competition is and should be an antitrust question, not merely one that involves interpreting statutes or agency regulations. Some level of antitrust enforcement - with appropriate deference to firm decisions about product design and affirmative regulatory decisions that affect market conditions - provides a necessary check on behavior, such as product hopping, that has no purpose but to exclude competition.

Traditional justifications for patents are all based on direct or indirect contribution to the creation of new products. Patents serve the social interest if they provide not just invention, but innovation the world would not otherwise have. Non-practicing entities (NPEs) as well as product-producing companies can sometimes provide such innovation, either directly, through working the patent or transfer of technology to others who do, or indirectly, when others copy the patented innovation. The available evidence suggests, however, that patent licensing demands and lawsuits from NPEs are normally not cases that involve any of these activities. Some scholars have argued that patents can be valuable even without technology transfer because the ability to exclude others from the market may drive commercialization that would not otherwise occur. We demonstrate that even if various commercialization theories can sometimes justify patent protection, they cannot justify most NPE lawsuits or licensing demands.

Patentees overwhelmingly lose their cases, despite a seeming host of procedural advantages. The same is not true of other IP plaintiffs. Why? In this article, I suggest that the explanation lies in the "fractioning" of patent law into smaller and smaller issues. Claim construction after Markman is the clearest example, but there are others. We no longer decide in a holistic manner what a patent claim covers. Instead, we decide what each word of a claim covers. Because there are more and more such issues, and the patentee must win each of them, patentees face a form of multiple jeopardy. It is ironic that patent claims, developed to broaden and strengthen the patent right, have instead become obstacles to the patentees success.

Patent law turns the attorney-client privilege on its head. Patent law punishes willful infringers by subjecting them to treble damages. An odd set of legal rules stemming from patent laws effort to determine what constitutes willful infringement effectively requires companies confronted with a patent first to obtain a written opinion of counsel and then to disclose that opinion in court. To do that, the accused infringer will have to waive its attorney-client privilege in virtually every case. Even worse, the law puts the question whether an accused infringer will have to waive privilege in the hands of the patent owner, who can send a carefully crafted letter putting a potential defendant on notice of the patent. A patent owner thus triggers the obligation to obtain a written opinion of counsel without actually threatening to sue anyone. In turn, accused infringers who are aware of these rules respond to such letters by obtaining a sort of pseudo-legal advice that both they and their attorneys recognize to be a construct. Both plaintiffs and defendants are playing a costly game. The rules of this game have perverse consequences for patent law. Some of these consequences affect litigation - lawyers and clients who know that the lawyers advice will be disclosed to the other side will behave differently, withholding information and candid advice from each other. But other consequences extend beyond litigation. They infect pre-litigation advice, essentially making it impossible for a competent lawyer to advise a client that a competitors patent should be avoided. The rules of the willful infringement game set traps for the unwary, who may not realize the consequences of failing to obtain the necessary written opinion of counsel. They interfere with a clients ability to choose counsel. And they discourage engineers and companies from reading a competitors patents at all, thereby undermining the disclosure function that is at the foundation of the patent system. One possibility is to abolish the willfulness rule entirely. We ultimately reject this approach because we worry that ordinary patent damages alone will be insufficient to deter infringement optimally in many cases. Another possibility is to abolish the rule that requiring disclosure of opinions of counsel. While a good idea, this option would not solve the problems created by the willfulness game, because many defendants will still need to rely on the opinion of counsel in order to disprove willfulness. Instead, we think the better approach to willful infringement is first to redefine it as adopting a technology with knowledge that it was derived from the patentee, and second to adjust the premium charged for it. Many of the problems with the willfulness rules stem from the fact that willfulness is an ongoing inquiry. The ongoing nature of the inquiry adversely affects a defendant that develops or adopts a technology in good faith but later learns it is infringing a patent. Changing the focus of the inquiry to the time of adoption is consistent with the ordinary understanding of willfulness outside of patent law and would help end the willfulness game. An independent developer could never be a willful infringer, and thus would not need either to obtain or disclose in court a written opinion of counsel merely because it later learned of a patent. By contrast, an accused infringer would need advice of counsel if it was aware of a patent and affirmatively sought to design around the patent. Such an accused infringer therefore would have to waive privilege. But since only the accused infringers intent at the time of adoption would matter, the scope of privilege waiver would be limited to communications at the time of adoption, and would not infect the advice given by litigation counsel.

Rapid advances in digital and life sciences technology continue to spur the evolution of intellectual property law. As professors and practitioners in this field know all too well, Congress and the courts continue to develop intellectual property law and jurisprudence at a rapid pace. For that reason, we have significantly augmented and revised "Intellectual Property in the New Technological Age. The 2016 Edition reflects the following principal developments: ● Trade Secrets: Congress passed the Defend Trade Secrets Act of 2016, one of the most momentous changes in the history of trade secret protection. The new law opens up the federal courts to trade secret cases, provides for ex parte seizures of misappropriated trade secrets in “extraordinary circumstances,” and establishes immunity for whistleblowers. ● Patents: The past several years have witnessed some of the most significant developments in U.S. patent history — from the establishment of the new administrative review proceedings at the Patent Office to important shifts in patent-eligibility, claim indefiniteness, and enhanced damages at the Supreme Court and means-plus-function claim interpretation and infringement doctrine at the Federal Circuit. We have restructured the patent chapter to illuminate these areas. We have also significantly expanded coverage of design patents in response to the growing importance of this form of protection. ● Copyrights: The Supreme Court issued important decisions addressing the public performance right and the first sale doctrine. The past few years also witnessed important developments in the Online Service Provider safe harbor, fair use, and state protection for pre-1972 sound recordings. We have also integrated the digital copyright materials into a unified treatment of copyright law and substantially revamped the fair use section to reflect the broadening landscape of this important doctrine. ● Trademarks: We have integrated important cases on federal registrability of disparaging marks, merchandising rights, likelihood of confusion on the Internet, and remedies. ● Other State Protections: We have updated material on the right of publicity, an active and growing area. We have also reorganized the chapter and focused it on IP regimes.

In Edwards v. Arthur Andersen, the California Supreme Court reaffirmed thestate's strong policy against noncompetition agreements, rejecting theNinth Circuit's "narrow restraint" exception. We explain what the Courtdid, why California's policy makes sense, and what the opinion will meanfor employers, for the high-tech industry, and for trade secret law.

On Feb 12, 2013, the PTO held a roundtable about software patents at Stanford. Software patents have received a lot of attention and we dont believe it is undue: software patents are behind a disproportionate share of patent litigations -- more specifically, over half (55%) of all patent defendants and 82% of PAE ("patent troll") defendants are there because of a software patent, applying the Graham-Vishnubhakat definition to data provided by RPX Corporation. In this presentation, we more rigorously apply 35 USC 112(f) in accordance with the proposal Mark Lemley outlines in his WIRED oped "Lets Go Back to Claiming the Problem Not the Solution" to 30 patents - 10 PAE and 20 control patents, provided by Patent Freedom. We find that 1) PAE patents are overwhelmingly functionally claimed (100%), but non-PAE patents are also functionally claimed (50%), 2) 40%-50% of the PAE patents were only supported by the highest levels of abstraction, but that 3) not all code is created equal - low-level code over generic elements does not necessarily promote technical progress. We provide 5 examples to support these findings, and discuss their implications.

In his influential paper “Contracting Into Liability Rules: Intellectual Property Rights and Collective Rights Organizations,” Rob Merges makes the case that intellectual property (IP) owners vested with property entitlements can and do contract away their right to an injunction when it is efficient for them to do so. The result was to cement for many the superiority of property over liability rules, since Merges demonstrated a seemingly critical asymmetry between the two. Merges’s evidence suggested that if a judge or a legislature gets the damages calculation wrong, we are stuck with an inefficient liability rule, but that we weren’t similarly stuck with an inefficiently-allocated property rule. The evidence Merges brought to bear in his path-breaking article is extremely important. But it is incomplete. True, parties can contract around inefficient property rules in IP cases. But as I show in this paper, they can - and do - contract around inefficient liability rules as well. The result does not prove the superiority of liability rules over property rules, but it does undermine a major premise that has been used to support the claim that IP rights must be protected by property rules.

What are patentees entitled to if they don't get an injunction? Courts havetended to award reasonable royalties, but there is considerable – andunnecessary confusion about who awards those royalties and how they arecalculated. In this paper, I endeavor to answer those questions.

We don’t want a "Mother, may I?" regulatory regime for innovation. A regulatory regime that requires permission to enter the market or develop a new product is a problem for innovation because it relies on the government, not the innovator, to decide the course of innovation. Giving private entities "Mother, may I?" control over entry is no better; while market competition is efficient, private decisions by a single actor controlling a market generally arent. Unfortunately, IP law increasingly looks like a regulatory regime, not a common law mechanism for internalizing social costs. The Copyright Act has more and more regulatory characteristics, and copyright owners increasingly seek to put the government in the position of having to approve new media technologies. Patent law too seems to impose entry barriers, not just in the pharmaceutical and biotechnology industries, where government regulation expressly prevents entry, but also in the costs patent enforcement imposes on start-ups in the software industry. The turn towards regulatory IP does not bode well for innovation and competition.

The Internet Corporation for Assigned Names and Numbers (ICANN) is a private non-profit company which, pursuant to contracts with the US government, acts as the de facto regulator for DNS policy. ICANN decides what TLDs will be made available to users, and which registrars will be permitted to offer those TLDs for sale. In this article we focus on a hitherto-neglected implication of ICANNs assertion that it is a private rather than a public actor: its potential liability under the U.S. antitrust laws, and the liability of those who transact with it. ICANN argues that it is not as closely tied to the government as NSI and IANA were in the days before ICANN was created. If this is correct, it seems likely that ICANN will not benefit from the antitrust immunity those actors enjoyed. Some of ICANNs regulatory actions may restrain competition, e.g. its requirement that applicants for new gTLDs demonstrate that their proposals would not enable competitive (alternate) roots and ICANNs preventing certain types of non-price competition among registrars (requiring the UDRP). ICANNs rule adoption process might be characterized as anticompetitive collusion by existing registrars, who are likely not be subject to the Noerr-Pennington lobbying exemption. Whether ICANN has in fact violated the antitrust laws depends on whether it is an antitrust state actor, whether the DNS is an essential facility, and on whether it can shelter under precedents that protect standard-setting bodies. If (as seems likely) a private ICANN and those who petition it are subject to antitrust law, everyone involved in the process needs to review their conduct with an eye towards legal liability. ICANN should act very differently with respect to both the UDRP and the competitive roots if it is to avoid restraining trade.

We award patents to inventors because we hope to encourage new ideas. For this reason, the fundamental requirement for getting a patent is that you have invented something new. It is curious, then, that patent law itself purports to pay no attention to which aspects of a patentee’s invention are in fact new. A patented invention is legally defined by its claims – written definitions of the invention. And those written definitions virtually never call out what it is that is new about the patentee’s invention. Even if the parties do identify the novel element of an invention, the law purports not to care. Long-standing patent law doctrine has decried any focus on the “point of novelty” of an invention. The United States Court of Appeals for the Federal Circuit evaluates the claim as a whole, not just the piece of the claim that the patentee actually added to the storehouse of knowledge. As the court frequently puts it, “there is no legally recognizable . . . ‘gist’ or ‘heart’ of the invention.” It turns out, however, to be hard to sustain a rule that a law concerned with novelty will pay no attention to the point of novelty. And so point-of-novelty issues crop up in a number of different doctrines in patent law, from figuring out who counts as an inventor to whether the inventor has disclosed the “best mode” of practicing the invention to when the sale of a product exhausts the patentee’s rights in the patent. Courts are inconsistent in whether and how they consider the point of novelty in these doctrines and more. But when the Federal Circuit presented with a question in point-of-novelty terms, it most often falls back on the mantra that there is no point of novelty to an invention, even if it means discarding long-standing precedent. It’s time to rethink the no-point-of-novelty doctrine in patent law. I argue that ignoring what is novel about patentee’s invention makes little sense as an across-the-board matter, and leads to a variety of harmful consequences. While refusing to focus on the point of novelty serves some valuable purposes, there are other ways to achieve those ends. And in the end, a patent regime that pays attention to what the patentee actually invented, not what the patent lawyer wrote down, is more likely to achieve the goal of promoting innovation.

Obviousness is the ultimate condition of patentability. The obviousness requirement - that inventions must, to qualify for a patent, be not simply new but sufficiently different that they would not have been obvious to the ordinarily skilled scientist - is in dispute in almost every case, and it is responsible for invalidating more patents than any other patent rule. It is also perhaps the most vexing doctrine to apply, in significant part because the ultimate question of obviousness has an I know it when I see it quality that is hard to break down into objective elements. That hasnt stopped the Federal Circuit from trying to find those objective elements. In the last quarter-century, the court has created a variety of rules designed to cabin the obviousness inquiry: an invention cant be obvious unless there is a teaching, suggestion, or motivation to combine prior art elements or modify existing technology; an invention cant be obvious merely because it is obvious to try; and so forth. In KSR v. Teleflex, the Supreme Court rejected the use of rigid rules to decide obviousness cases. In its place, the Court offered not a new test, but a constellation of factors designed to discern whether the person having ordinary skill in the art (the PHOSITA) would likely think to make the patented invention. In short, the Court sought to take a realist approach to obviousness - to make the obviousness determination less of a legal construct and to put more weight on the factual determination of what scientists would actually think and do about a particular invention. As a general principle, this realist focus is a laudable one. The too-rigid application of rules designed to prevent hindsight bias had led to a number or results that defied common sense, including the outcome of KSR itself in the Federal Circuit. But the realist approach has some (dare we say it) nonobvious implications for evidence and procedure, both in the Patent and Trademark Office (PTO) and in the courts. The greater focus on the characteristics of individual cases suggests a need for evidence and factual determinations, but the legal and structural framework under which obviousness is tested makes it difficult to make and review those determinations. The realist approach is also incomplete, because both the knowledge of the PHOSITA and the way the court approaches so-called secondary considerations of nonobviousness depend critically on the counterfactual assumption that the PHOSITA, while ordinarily skilled, is perfectly informed about the prior art. If we are to take a realist approach to obviousness, we should make it a consistent approach, so the ultimate obviousness determination actually reflects what scientists in the field would actually think. So far, despite KSR, it does not. The result of taking the realist approach seriously may be - to the surprise of many - a law of obviousness that is in some respects more favorable to patentability than the standard it displaced.

Intellectual property (IP) law doctrines fall into three basic categories: validity, infringement and defenses. Virtually every significant legal doctrine in IP is either about whether the plaintiff has a valid IP right that the law will recognize – validity – about whether what the defendant did violates that right – infringement – or about whether the defendant is somehow privileged to violate that right-defenses. IP regimes tend to enforce a more or less strict separation between these three legal doctrines. They apply different burdens of proof and persuasion to infringement and validity. In many cases they ask different actors to decide one doctrine but not the other. The U.S. Patent and Trademark Office, for example, decides questions of patent and trademark validity but not questions of infringement. Even in court, resolution of one issue is often allocated to a judge while the jury decides a different issue. And even where none of that is true, the nature of IP law is to categorize an argument in order to apply the proper rules for that argument. The result of this separation is that parties treat IP rights “like a nose of wax, which may be turned and twisted in any direction.” When infringement is at issue, IP owners tout the breadth of their rights, while accused infringers seek to cabin them within narrow bounds. When it comes to validity, however, the parties reverse their position, with IP owners emphasizing the narrowness of their rights in order to avoid having those rights held invalid and accused infringers arguing the reverse. Because of the separation between validity, infringement, and defenses, it is often possible for a party to successfully argue that an IP right means one thing in one context and something very different in another. And courts won’t necessarily detect the problem because they are thinking of only the precise legal issue before them. The result is a number of IP doctrines that simply make no sense to an outsider. In patent law, for instance, it is accepted law that there is no “practicing the prior art” defense. In other words, one can be held liable for doing precisely what others had legally done before, even though a patent isn’t supposed to cover things people have already done. In design patent law, one can be held liable for making a design that an “ordinary observer” would find too similar to a patented design, even though the things that make the two look similar – say, the roundness of the wheels on my car – are not things the patentee is entitled to own. In copyright, once a court has concluded that someone has actually copied from the plaintiff, a song will sometimes be deemed infringing because of its similarity to a prior song, even if the similarity is overwhelmingly attributable to unprotectable standard components of the genre. And in trademark, a party can be deemed infringing because its products look to similar to the plaintiffs’ mark and therefore make confusion likely, even if that confusion is likely caused by non-source-designating features of the design. The culprit is simple, but fundamental: IP regimes largely lack an integrated procedure for deciding the proper extent of an IP right. The proper scope of an IP right is not a matter of natural right or immutable definition. Rather, it is a function of the purposes of the IP regime. But without some way of assessing how broad an IP right is that considers validity, infringement, and defenses together, courts will always be prone to make mistakes in applying any one of the doctrines. In this article, we suggest that IP regimes need a process for determining the scope of an IP right. Scope is not merely validity, and it is not merely infringement. Rather, it is the range of things the IP right lawfully protects against competition. IP rights that claim too broad a scope tend to be invalid, either because they tread on the rights of those who came before or because they cover things that the law has made a decision not to allow anyone to own. IP rights with narrower scope are valid, but the narrowness of that scope should be reflected in the determination of what actions do and do not infringe that right. And whatever the doctrinal label, we should not allow an IP owner to capture something that is not within the legitimate scope of her right. Nor should it follow from the fact that some uses are outside the lawful scope of an IP owner’s right that the IP right itself is invalid and cannot be asserted against anyone. Only by evaluating scope in a single, integrated proceeding can courts avoid the nose of wax problem that has grown endemic in IP law. Scope is, quite simply, the fundamental question that underlies everything else in IP law, but which courts rarely think about expressly.

The process of claim construction - determining the meaning of patent claims - is the most important part of patent litigation. There are a number of rules or canons that courts can use in applying the interpretive sources to reach an understanding of what patent claims mean. The canon that has arguably had the most significant impact on claim construction is the doctrine of claim differentiation. The claim differentiation doctrine in its broadest reading provides that no two claims in the same patent should be interpreted to cover the same thing. As a general matter, applying the doctrine of claim differentiation results in broader constructions of patent claims, because it is most commonly used to prevent defendants from limiting a broad genus claim to the range of embodiments actually disclosed or more explicitly recited in other claims. Sometimes this is the right result, because defendants are improperly seeking to limit broader genus claims to the preferred embodiments disclosed in the specification. But at other times it leads to problematic results. In this article, I conduct an empirical review of claim differentiation decisions in the Federal Circuit and in the district courts, and I suggest limiting principles that can be used to distinguish good uses of the doctrine from bad.

The United States Patent and Trademark Office is tasked with the job of reading patent applications and determining which ones qualify for patent protection. It is a Herculean task, and the Patent Office pursues it subject to enormous informational and budgetary constraints. Nonetheless, under current law, courts are bound to defer to the Patent Offices decisions regarding patent validity. In this Article, we argue for reform. Deference to previous decision-makers is appropriate in instances where those previous decisions have a high likelihood of accuracy, and the patent system should endeavor to create processes that fit this mold. But granting significant deference to the initial process of patent review is indefensible and counter-productive. Patents should be vulnerable to challenge until and unless they are significantly evaluated in an information-rich environment. At that point, they will have earned and therefore should be accorded a presumption of validity. Such an approach would better serve the patents systems long-run incentive goals, and it would give patent applicants better incentives to file for genuine inventions but leave their more obvious and incremental accomplishments outside the patent systems purview. Here, we therefore suggest the creation of a two-tier system of patent validity, with patents that are subject to intensive scrutiny accorded a strong presumption of validity, while untested patents are left to be evaluated more fully in court.

We argued in our paper, "Patent Hold-Up and Royalty Stacking," that the threat to obtain a permanent injunction greatly enhances the patent holders negotiating power, leading to royalty rates that exceed a benchmark level based on the value of the patented technology and the strength of the patent. John Golden, in his extensive comment on our paper, claims: "Lemley and Shapiro err when they claim to have proven that patentees whose inventions are only one component of a larger product are systematically overcompensated." However, the error is Goldens not ours. When patentees systematically capture value they did not create from others who did create it, they are being overcompensated by any reasonable measure, including the standard economic models on which we relied. In Part II, we briefly respond to his criticism of our empirical study of court-awarded reasonable royalties. Finally, Golden also claims that our recommendation to reduce the availability of permanent injunctions to patent holders who have claims to reasonable royalties but not lost profits remedy "threatens to distort markets for innovation." We strongly disagree. It is patent holdup, which skews damages in ways more favorable to reasonable royalties, that distorts markets for innovation. A rule such as the one we propose, in which damages are calibrated to compensate patentees for their loss, is sensible public policy.

Trade secret law is a puzzle. Courts and scholars have struggled for over a century to figure out why we protect trade secrets. The puzzle is not in understanding what trade secret law covers; there seems to be widespread agreement on the basic contours of the law. Nor is the problem that people object to the effects of the law. Rather, the puzzle is a theoretical one: no one can seem to agree where trade secret law comes from or how to fit it into the broader framework of legal doctrine. Courts, lawyers, scholars, and treatise writers argue over whether trade secrets are a creature of contract, of tort, of property, or even of criminal law. None of these different justifications have proven entirely persuasive. Worse, they have contributed to inconsistent treatment of the basic elements of a trade secret cause of action, and uncertainty as to the relationship between trade secret laws and other causes of action. Robert Bone has gone so far as to suggest that this theoretical incoherence suggests that there is no need for trade secret law as a separate doctrine at all. In this article, I suggest that trade secrets can be justified as a form, not of traditional property, but of intellectual property (IP). The incentive justification for encouraging new inventions is straightforward. Granting legal protection for those new inventions not only encourages their creation, but enables an inventor to sell her idea. And while we have other laws that encourage inventions, notably patent law, trade secrecy offers some significant advantages for inventors over patent protection. It seems odd, though, for the law to encourage secrets, or to encourage only those inventions that are kept secret. I argue that, paradoxically, trade secret law is actually designed to encourage disclosure, not secrecy. Without legal protection, companies in certain industries would invest too much in keeping secrets. Trade secret law develops as a substitute for the physical and contractual restrictions those companies would otherwise impose in an effort to prevent a competitor from acquiring their information. The puzzle then becomes why the law would require secrecy as an element of the cause of action if its goal is to reduce secrecy. I argue that the secrecy requirement serves a channeling function. Only the developers of some kinds of inventions have the option to over-invest in physical secrecy in the absence of legal protection. For products that are inherently self-disclosing (the wheel, say, or the paper clip), trying to keep the idea secret is a lost cause. We dont need trade secret law to encourage disclosure of inherently self-disclosing products - inventors of such products will get patent protection or nothing. But if trade secret law prevented the use of ideas whether or not they were secret, the result would be less, not more, diffusion of valuable information. The secrecy requirement therefore serves a gatekeeper function, ensuring that the law encourages disclosure of information that would otherwise be kept secret, while channeling inventors of self-disclosing products to the patent system. My argument has a number of implications for trade secret policy. First, the theory works only if we treat trade secrets as an IP right, requiring proof of secrecy as an element of protection. If we give the protection to things that are public, we defeat the purpose and give windfalls to people who may not be inventors (what we might call "trade secret trolls"). Courts that think of trade secret law as a common law tort rather than an IP right are apt to overlook the secrecy requirement in their zeal to reach "bad actors." Second, an IP theory of trade secrets also encourages preemption of "unjust enrichment" theories and other common-law ways courts are tempted to give private parties legal control over information in the public domain. Thus, an IP theory of trade secrets is in part a "negative" one: the value of trade secret law lies in part in defining the boundaries of the cause of action and preempting others that might reach too far. Finally, treating trade secrets as IP rights helps secure their place in the pantheon of legal protection for inventions. The traditional conception of the tradeoff between patents and trade secrets views the disclosure function of the patent system as one of its great advantages over trade secret law. And indeed the law operates in various ways to encourage inventors to choose patent over trade secret protection where both are possible. But for certain types of inventions we may actually get more useful "disclosure" at less cost from trade secret than from patent law.

In the Festo decision, the Federal Circuit significantly changed the scope of the doctrine of equivalents in patent law. The doctrine of prosecution history estoppel precludes a patent owner from claiming during litigation to own ground given up during patent prosecution. Under the old rule, called the "flexible bar," estoppel was based on a multi-factor test and would apply only if the patentee had no choice but to amend its claims in the way it did. Festo replaced the flexible bar with an "absolute bar," under which virtually any amendment to a patent precludes resort to the doctrine of equivalents for that claim element. The Supreme Court is now considering whether the flexible or absolute bar is the right rule. We believe there is middle ground in this debate that has been ignored with both parties. Whether prosecution history estoppel applies to any given amendment should depend on the reasonably foreseeable effect of that amendment. Normally, patentees will understand that they are surrendering coverage by amending their patent claims, and so a rule precluding them from reclaiming that ground makes sense. But in some cases the absolute bar will produce unexpected and unintended results. We argue that the application of estoppel should turn on whether the effect of a change would be foreseeable to a reasonable patentee at the time of the amendment. This "foreseeable" bar better balances the competing policies of strong protection for pioneering inventors and notice to improvers who wish to design around a patent.