Electronic Networks for International Research Collaboration:
Implications for Intellectual Property Protection in the Early Twenty-First Century

And that kind of exclusive ownership cuts directly against the grain of the technology in question. From the start, computers were seen as tools of collaboration, designed to facilitate brainstorming and data sharing.

—Henry Jenkins, Technology Review (March/April 2000)

Introduction

In the early twenty-first century, the convergence of digital computing and telecommunications is opening unprecedented opportunities for international research and education collaboration, sharing of data and information worldwide, and accelerated scientific and technological progress. Yet some scholars argue that overly expansive intellectual property rights that deny future creators and potential collaborators the raw material they need could limit the ability of the scientific, technical, and educational communities to capitalize on such opportunities.

      Some were prescient about shared knowledge in the network environment becoming embroiled in issues of intellectual property protection at the beginning of the 1990s. New electronic network technology already had potential to transform international research collaboration, and a legal regime was evolving for extension of intellectual property protection to the activities of network users. Furthermore, internationalization of research had spurred interest in intellectual property as a set of ground rules for distribution of rights in international research transactions. Digitization of information and growth of research networking had accelerated this trend and raised additional issues for science and technology communities in intellectual property protection. The question was raised, "How should the intellectual products of international research collaborations in the network environment be protected in the interests of scientific and technological progress?"

      Electronic networks already meant more than remote online access to large databases on mainframe computers. Distributed networks had already begun to lead to electronic equivalents for scientific publishing and to faster, less formal means of communicating intellectual advances. The result was an increasingly broad spectrum of scientific communication, ranging from informal electronic messaging to complex software environments for visualization technology in large-scale simulations. At the same time, the greatly expanded scope and connectivity of networks began to erode the barriers of time and distance for research collaboration on a global scale. James J. Duderstadt, president of the University of Michigan, envisioned "research collaboratories" as advanced, distributed research environments that would use multimedia information technology to relax the constraints of distance, time, and even reality on research collaboration. ¹

      However, there were growing tensions between collaborative practices electronic networks facilitated and traditional notions of individual property rights in intellectual production: In the informal and often unpredictable intellectual collaborations of networks, authorship often was unrecorded—not to say uncompensated. Electronic technology was playing a crucial role in promoting practices in which individual contributions to dynamic intellectual products were deemphasized and effectively merged. The potential of these networks emphasized iterative intellectual practices in which more than one person contributed cumulatively over time, collaboration between two or more people across nations, and sharing of collections of information and data on a global scale.

      The battle shaping up over the future of electronic networks was, on one side, between those who saw its potential for new creativity and productivity in international collaborative endeavors as a threat to individual proprietorship in information and who saw vigorous application and extension of traditional intellectual property protection to the network environment as the solution. On the other side were those who argued that the network environment might become a new intellectual commons that excessive legal control could stifle. What appeared to be needed were rules to protect intellectual products that featured collaboration and, in general, were more closely aligned with the characteristics of contemporary intellectual production in the network environment.

      Today, the still primitive applications of the early 1990s—the communication networks and information services such as Internet, Bitnet, and CompuServe, the more sophisticated hypertext applications that were just beginning to be developed—are widely available globally.

      Internet technologies are having a dramatic impact on research activities, including the creation of entirely new forms of research. The development of networking tools and Web servers stimulated the field of bioinformatics—the use of computers to analyze genes and proteins—to take off and become a key element of basic scientific research and new drug development.

      Science journals have gone online, and hyperlinks replace old-fashioned footnotes. ² Global digital libraries enable researchers anywhere in the world with a computer and a modem to tap into the entire database of the Library of Congress, the Bibliothèque de France, or the British Library.

      Today, there is a move toward collaboratively constructing knowledge using distributed workgroups supported by an electronic communication infrastructure. The exponential growth of knowledge has made it almost impossible for any organization to exist in isolation. Powerful computers and networks deliver educational services to anyone at any place and any time, no longer confining education to the campus or the academic schedule. Recently, almost thirty countries agreed to create a distributed global biodiversity information facility to link local digital databases that embrace the full range of biodiversity information including geographic, ecological, genetic, and molecular data. ³

      Today, the early 1990s battle over legal control of intellectual products in the network environment rages as full-fledged warfare. Few other issues before the 106th Congress generated so much lobbying as the question of whether digital databases—collections of facts such as telephone directories, weather reports, stock tables, and real estate listings—could be copied, repackaged, and distributed by anyone with an Internet connection. ⁴

      The House Judiciary Committee had approved a bill that would establish criminal penalties for the unauthorized use of material in databases. Opponents said the bill would allow companies with databases to control access to public domain facts. The question of database piracy is one instance in which Internet technology has outstripped traditional intellectual property laws, proponents said. Before there was an Internet, the issue was pretty well settled by copyright law. In March 1991, the US Supreme Court ruled in Feist Publications, Inc. v. Rural Telephone Service that the telephone white pages could not be copyrighted because the compilation did not involve selection and coordination; one could use the list without permission. ⁵

      Today, the ease with which information on the Internet can be retrieved and the size of the investments and potential profits blur the line between what is protected and what is not. Companies and interest groups have chosen sides on whether to adjust the level of protection upward depending on whether they primarily collect data that is put on the Internet (the stock exchanges, real estate brokers, Lexis-Nexis, eBay, American Medical Association) or use the data compiled by someone else (the Chamber of Commerce, Consumers Union, Yahoo, Schwab, research librarians).

      Undoubtedly, legal scholars say, publishers will see suboptimal profits from the production of databases if Congress fails to protect them against certain forms of piracy. However, if Congress combats this risk of market failure by enforcing strong monopolies in collections of data, such an approach may end up "balkanizing the information economy by recreating the medieval economic quandary in which products could not flow across countries or continents because too many feudal monopolists demanded payments every few miles down the road." ⁶

      This chapter considers the implications of broad and strong traditional intellectual property protection and raises broad protection issues that emphasize alternatives to vigorous extension of traditional principles in the network environment.

      The basic argument is that if taking full advantage of the potential of the network environment for collaborative research, cumulative innovation, and scientific and technological progress is the objective, intellectual property protection laws and the courts ought to be careful not to grant protection that is too broad in scope. The actual or potential harmful effects of protection can be mitigated if the law pays close attention to the extent of the author's or inventor's efforts and what they actually achieved, and to the broad nature of what was already known and in the public domain, and restricts scope accordingly. Intellectual property systems should promote liberal licensing to allow multiple players to follow initial inventors and authors and promote transformative or fair use defenses to give value-adding advances a chance over plaintiffs' intellectual products in infringement litigation.

      The framework for this chapter is based on the astute analysis of theory and historical development of intellectual property rights by legal scholar Peter Jaszi in his insightful articles on authorship or romantic creativity. ⁷ As he and others point out, this vision of romantic creativity expands our common-sense traditional intellectual property law and doctrines in the direction of greater protection. ⁸ The chapter also draws on evidence that in many cases, granting broad-scope protection stifles cumulative improvements in technology and that where advance has been rapid there always has been rivalry.

Rationale of Traditional Intellectual Property Protection

What do we know about current intellectual property systems that might help modernize these principles to conform to contemporary intellectual production in the network environment? Traditional theory of intellectual property protection is informed by a commonly perceived, seemingly basic contradiction of purpose that protection aims to promote access to works of authorship but confers on authors power to restrict or deny access to their works.

      The rationale, stated simply, is that without protection, the social benefits of research advance often exceed the private rewards; the intellectual property system is designed to narrow the difference through monopoly grants to authors and inventors. Although new inventions contribute enormously to social welfare, the creators and developers may not gain enough to cover their costs and compensate their risk if rival firms, that have not borne the costs of development, can rapidly copy the technology. In this event, fewer inventions might occur in competitive markets. Thus, monopoly pricing of new inventions for a limited term preserves incentives for firms to supply a continuing stream of new inventions on the market.

      Traditional analysis of intellectual property helps us to pigeonhole many of the overt controversies in intellectual property traceable to the basic tension between public benefits and private reward to authors and inventors to achieve those benefits. However, it does not go far enough to explain many of the conceptual challenges faced by traditional intellectual property law in a time of rapid change in modes of intellectual production.

Mythology of Traditional Protection

The Early 1990s

By the early 1990s, much more reflective analysis of traditional intellectual property concepts began to argue that only demythologizing traditional doctrine would make room for protection of contemporary intellectual practices. These critiques of intellectual property doctrine emphasized that traditional concepts had historical basis in individual authorship as a privileged category of intellectual activity that generated intellectual products of special social value and entitled practitioners (authors or inventors) to unique rewards and privileges.

      For specially privileged and gifted authors, all benefits from intellectual advances—some of which might have become available to competitors in the marketplace—were included in private rewards in the prices authors charged buyers for intellectual products. However, the reward system based on the author as an individual solely responsible for original work tends to mislead when applied to iterative, collaborative, and other kinds of multiauthor intellectual practices. Prices for their intellectual properties charged by authors may be too high because individual orchestration may not be the best production process for those intellectual products.

Expansion of Traditional Intellectual Property during the 1990s

Despite these admonitions, the strengthening and broadening of traditional intellectual property rights during the 1990s was remarkable. When Stanley Cohen and Herbert Boyer invented the gene-splicing technique in 1973, they resisted the suggestion of the Stanford University patent officer to patent. "My initial reaction ... was to question whether basic research of this type could or should be patented and to point out that our work had been dependent on a number of earlier discoveries by others," Cohen explained. ⁹ The cumulative nature of scientific knowledge meant that even the most outstanding discovery was heavily dependent on the work of many others, and to patent a discovery would be to lay claim to the freely given intellectual property of others.

      In 1980, however, patents became a primary legal instrument for research discoveries financed by the public sector and performed by U.S. universities. By 1997, the number of university patents issued had already doubled the early 1990s number, reaching more than 2,400 patents. ¹⁰ The increase far exceeded the rate of growth of total U.S. patenting.

      The Cohen-Boyer patent is nonexclusively licensed to all takers at a low fee. But exclusive licenses on universities' broad basic patents may deter and complicate follow-on research. Research tools such as the Harvard mouse may lead to blockages on downstream research and commercial product development as upstream owners set up tollbooths that increase costs and inefficiencies for follow-on innovators. ¹¹

      Authorship was involved in rationalizing the extension of copyright protection to computer software in 1980. ¹² Copyright in a work protects only expression, leaving the idea free for reuse; the problem is that the distinction between idea and expression proves difficult to articulate in practice. The question of where and how to draw the line was at the center of debate over the proper scope of copyright protection for computer software, much of it revolving around the 1986 decision in Whelan Associates. v. Jaslow Dental Laboratory. ¹³ The Whelan court relied on decisions finding the infringement of one audiovisual work by another, where the two displayed the same total look and feel.

      The court assumed that intellectual production of software was similar enough to audiovisual works that software authors should have the same control over subsequent adoptions as audiovisual authors. However, software seems more an intellectual product that results from successive elaborations of an idea over time and less a discrete intellectual product in which a broad scope of protection is not a serious hindrance to elaborations by future authors. Advancing software technology often entailed the ability to use a number of already developed components and hence either the ability to negotiate a license or an environment where litigation was not a serious threat.

      A decade ago, the field of software functioned without patents. Early on, the courts, as in a 1972 Supreme Court ruling, compared software's logical steps to "mental processes" that not only could not be patented but had to be preserved in the public domain as the "basic tools of scientific and technological work." Over time, though, this position eroded. By the end of 2000, the Patent Office had granted almost 100,000 software patents. ¹⁴ The 1998 Federal Circuit ruling in State Street Bank & Trust Co. v. Signature Financial Group Inc. endorsed software patentability without qualification. ¹⁵

      This strengthening of intellectual property rights was especially prominent in the United States. But through 1994 negotiations regarding GATT, the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement), the United States was pushing on other countries its beliefs about strong, broad intellectual property protection. ¹⁶ Other countries were going along, not always simply as a reaction to the pressure but also because of an honest belief on the part of many parties that in the long run strong intellectual property protection would promote scientific and technological progress.

      In reaction to Feist, the U.S. Supreme Court ruling that copyright protection did not extend to databases that did not involve some "creative" selection and organization of data, the European Union in March 1996 enacted a database directive that granted special rights to the owner of a database, even if the database did not meet the standards of copyright protection under the directive. ¹⁷

      Europe and the United States sponsored initiatives in the international forums of the World Intellectual Property Organization (WIPO) also calling for creation of a new form of legal protection for the contents of databases aimed to protect database owners against unauthorized extractions, uses, and reuses of the whole or substantial parts of its contents. ¹⁸ The draft law never reached the conference's floor; it was deferred by the hostility of other national delegations and the vocal opposition of U.S. scientific organizations.

      Whether the U.S. Congress should give databases greater legal protection became a controversial issue.

The Late Twentieth Century

The potentially enormous negative impacts of traditional intellectual property law and, in particular, the image of the romantic author on property rights in research collaborations and cumulative improvements on original works did not go unnoticed as the decade unfolded. Shamans, Software, and Spleens: Law and the Construction of the Information Society, the widely regarded 1996 work by James Boyle, said that "we are driven to confer property rights in information on those who come closest to the image of the romantic author, those whose contributions to information production are most easily seen as original and transformative." ¹⁹ He argued that this had negative consequences; it led to having too many intellectual property rights, conferring them on the wrong people, and undervaluing the interests of both the sources of and the audiences for the information commodified.

      In an analysis parallel to authorship inquiries, Roberto Mazzoleni and Richard R. Nelson thoughtfully reviewed alternative theories about the functions of patents and their underlying assumptions. ²⁰ Traditional patent theory fits a discrete model of technical advance. The basic invention may be amenable to tailoring for different uses or customers; however, the invention does not point the way to wide-ranging subsequent technical advances.

      However, this characterization is inappropriate when technical advances are cumulative in that today's advance lays the basis for tomorrow's. In such chains of invention, the ability to operate the most advanced version is based on the ability to do things that were the subject of earlier inventions.

      Unless licensed easily and widely, such patents tend to limit the range of potential users who have access to all components of the technology. Also, there is no reason to assume that when blockages arise industries will always turn to the deadlock-breaking solutions, patent pooling, and cross-licensing.

      What follows briefly surveys cumulative intellectual products, multiauthor collaborations, and digital collections of information and data in networks for the effects of across-the-board strengthening of intellectual property rights and provides suggestions for intellectual property rules alternative to those of traditional, authorial protection.

Protection for Cumulative Intellectual Products

Protection for Internet-Based Bioinformatic Research Tools

Intellectual production that is the basis for traditional copyright protection focused on incentives for creating discrete intellectual products; it neglected to consider whether different incentives might be needed to advance intellectual production of the more cumulative kind that has come to typify intellectual software production.

      Today, bioinformaticians churn out software tools to find genes and other significant features in DNA sequences and to compare DNA, RNA, and protein sequences. ²¹ The explosion has been triggered not only by the supply of data spewing from the genome projects but also by demand from biologists with Internet access. More than 500 different bioinformatics tools are available over the Internet, and their number is growing constantly as software developers working in the life sciences update and produce useful new applications. ²²

      Many of the new data-mining research tools are cumulative improvements over techniques in machine learning and artificial intelligence. The hottest research tool, known as hidden Markov models, springs directly from statistics and linguistics. But granting broad protection to statistical software programs may preclude development of future data-mining algorithms without engaging in infringement or making sometimes complex licensing arrangements with owners. ²³

      The issues of patent protection of computer algorithms useful in bioinformatics applications are fast developing alongside copyright issues. Although most of the attention regarding the 1998 State Street decision (in which Federal Circuit once again confirmed the qualification of such computer programs as patentable subject matter) has revolved around e-commerce business methods, implications for bioinformatics are equally striking. Several categories of bioinformatics-related software applications came to the center stage as subject matter for which patent protection appeared feasible: the database architecture for genetic information; algorithms specially designed to search, manipulate, and manage data meaningfully; and user-friendly interfaces to facilitate information requests and provide readily understandable results.

      Such patent positions cause concern because bioinformatics-related software applications may be viewed as upstream research tools on which owners may stack tollbooths on the road to subsequent downstream research and commercial product development. The problem is likely to become increasingly serious in computer software, legal scholars say, where the practical limit on claim breadth seems to be only the imagination of the claim drafter. ²⁴ An example is a recent patent claiming the use of computer-implemented artificial neural networks for identifying binding motifs of polypeptides; the patent is based on one example of the technique. ²⁵ When the U.S. Supreme Court last explored this issue in 1966 in Brenner v. Manson, it rejected a broad claim. Such a patent, the Court said, may confer power to block off whole areas of scientific development without compensating the public.

      In dealing with this issue, legal scholars say, it is crucial to balance incentives to initial innovators against incentives to follow-on innovators. Experience suggests that this balance is weighted too much in favor of the initial innovator, and the utility doctrine, for example, which was at issue in Brenner v. Manson, should be used to restrict the patenting of very fundamental concepts. ²⁶

      Fortunately, the general trend appears to be toward liberal licensing of these research tools. For example, eBioinformatics, Inc. enables researchers worldwide to take advantage of bioinformatics software applications over the Internet, using a pay-as-you-go access model. Also, the evolution of patent protection may reflect that related to other database browsers: The dynamics between Netscape and Microsoft, involving methods of viewing Internet materials, or between Lexis-Nexis and the West Group involving ways of accessing legal databases, may provide useful paradigms for development in the bioinformatics arena.

Transformative or Fair Use Defenses in the Network Environment

Fair use analysis also mitigates the problem in cumulative improvements caused by broad protection. The doctrine was created in the 1840s in Folsom v. March. ²⁷ Transformation of the taken material was necessary to make it fair use, the Court found, and this is the basis for the rule that a fair use must be a transformative use. Copyright would protect the first author against thieves but not against those whose investment of their borrowings from the initial source produced a higher net yield.

      The unanimous 1994 Supreme Court decision in Campbell v. Acuff-Rose Music Inc. held that a derivative work's purpose and character should be judged primarily by whether and to what extent it is transformative, that is, the extent to which the derivative work adds something new, with a further purpose or different character, that alters the preexisting work with new expression, meaning, or message. ²⁸ Viewed in this light, a defendant's use of the intellectual product will constitute infringement or free riding on plaintiff's efforts only when the defendant's use of the information is not transformative. ²⁹

      Although fair use analysis applies specifically to copyright protection, patent law limits infringement through a parallel inquiry: The Supreme Court long ago ruled that if a defendant so transformed a device that the claims of the patent, literally construed, ceased to represent her invention, she is not an infringer on "reverse equivalents" grounds. ³⁰ Once a court completes assessment of the protected product, it would consider the advance in the accused product. Analysis along these lines might have reduced the blocking effect of patents in certain fields.

      However, legal scholars say that fair or transformative use provisions can be undermined in the network environment if applications such as digital libraries—which promote sharing information—require users to authenticate themselves and then track and regulate what the users do with the information. ³¹ Fair use defenses have been deliberately constructed as gray areas of the law so that the legality of an act may depend on multiple factors that judges will weigh together if litigation occurs. The public interest in copyright law does not go away just because the work is in digital form.

      But when the rock band Metallica filed suit against Napster Inc., whose popular MP3-sharing software lets users exchange music files over the Internet, Napster sought refuge from charges of copyright violations under the Digital Millennium Copyright Act (DMCA). ³² The DMCA says Internet service providers (ISPs) are not liable for piracy committed by users on their services. ³³ Some called for an amendment to the act that would crack down on emerging Internet technologies such as Napster. ³⁴ They held that services such as Napster should be forced to gather identifiable user data such as addresses and should be required to block users who violate the law.

      Others were convinced that Napster is lawful to begin with and, more importantly, that any efforts to shut it down would prove futile. Napster will be replaced by a host of decentralized Internet-mounted software technologies or networks. These perform Napster-like functions without needing a Web-based hub site. They offer no easy target for a suit because there is no Web site to shut down.

      But what about the larger question? Companies already offer digital technologies that can supply "wraps," in which copyrighted music or movie or video files can be sealed at the behest of their copyright holders. Because they can continue to wrap the copyrighted files even after payment has been made, they can limit the number of times the consumer can open the file, they can dictate the brands or genres of machine or computer that will be permitted to play the work, and they can limit or bar the consumer from copying the work, either in whole or in part. ³⁵

      Armed with this sort of unprecedented technological control over their works, authors and publishers will no longer need to permit consumers to use their works in ways that, until now, consumers have taken for granted and copyright law has declined to protect, deeming them legitimate, noninfringing fair uses of copyrighted material.

      The DVD case pitted the movie industry against Internet journalist Eric Corley, who posted a descrambling code, called DeCSS (descramble content scrambling system) on his Web site, 2600.com, the self-proclaimed "hacker journal." DeCSS descrambles the encryption system used by the film industry to keep people from making copies of DVD movies. ³⁶ The prosecution argued that DeCSS violates the DMCA, which prohibits the circumvention of technologies used to protect copyrighted materials. The defense insisted that DeCSS, which was written as part of an open-source project to allow the Linux operating system to play DVDs, merely provides consumers with the rights to fair use (such as taping a CD so that you can play it in your car) they have traditionally enjoyed.

      The movie industry won; the U.S. District Court ordered a permanent injunction against the hacker Web site. ³⁷ The Court's comments made a thoughtful contribution to the debate over these issues. However, that debate may no longer be about the role of government after the digital revolution has rendered intellectual property rights unenforceable. Rather, it may be about the role of government now that the digital revolution threatens to render intellectual property rights absolute.

      Digital library builders clearly will need to keep track of emerging legal issues likely to affect their libraries of digital works and take some responsibility for how these issues are resolved. But laws vary from country to country; "international copyright," in the sense of a uniform law binding all nation states, does not exist. Rather, at present we have a system of interlocking national copyrights, woven together by the principle of national treatment. ³⁸ Some legal scholars ask whether this would require reengineering of certain parts of the information infrastructure to zone cyberspace to conform to national boundaries so that national policies can be preserved in cyberspace.

      Also involving claims of fair use, raising questions of originality and of whether pointing to another's work can be a work in itself, hyperlinking and related suits involving "unauthorized" use of Web sites have proven legally complex. ³⁹ Legal problems are sure to arise for scholars as scientific authors increasingly cite references by hyperlinking "without permission" to other Web sites. ⁴⁰

      Online auction company eBay worried that another company would take its listing for, say, a set of Wedgwood china and the listings on all the other auction sites for a set of the same china and provide a new list for comparative shoppers. eBay said that the prices and availability of its items change minute by minute. The competing Web site could have misinformation. For these reasons, eBay sued a company called Bidder's Edge, and a U.S. District Court issued a preliminary injunction preventing Bidder's Edge from using eBay's data. ⁴¹

      Some legal scholars were concerned that the ruling could affect hyperlinking and more generally the free flow of information over the Internet. ⁴² It might mean that everybody has a right to pick and choose who comes to their Web site and for what purposes. More troubling, the decision may make the whole concept of search engines tenuous. Under one interpretation of the Court's ruling, anyone who is not authorized to come to a site could be deemed a trespasser after the fact. Thus, anyone who wants to block access to a search engine can do so or can give preferential treatment to a search engine.

Licensing Digital Works

A fair use defense did not work in the seminal ruling in which a New York district court ruled against MP3.com's personalized service in response to a recording industry copyright violation suit. ⁴³ "The complex marvels of cyberspatial communication may create difficult legal issues," the judge said, "but not in this case." ⁴⁴ My.MP3.com subscribers can pop a physical CD into their disc drives and have digital copies instantly appear in their music collections. MP3.com gets the digital copy from its database of 80,000 predigitized CDs, many of which belong to the major music labels. The record companies alleged that copying and distributing their CDs was unlawful.

      The Court disagreed with MP3.com's contention that copying the CDs from its database is permissible under the fair use doctrine. The Court also dismissed MP3.com's claim that giving the consumer a digital version of a CD was simply a "functional equivalent" of storing user CDs for members.

      Soon afterward, Warner Music Group and BMG, two Big Five labels, settled their copyright infringement claims against MP3.com. ⁴⁵ They issued the music download site a license to use their copyrights for its My.MP3.com service, which requires MP3.com to pay the labels each time a copyrighted song is uploaded to the system and each time a song is streamed. The deal sets a dramatic precedent for the future of music and other digital works. MP3.com said license fees would be recouped through advertising and additional CD sales rather than through a subscription fee to users.

      Although some prominent music acts such as Metallica and Dr. Dre sided with the labels against the free download site, rock star and actress Courtney Love said, "Stealing an artist's music without paying for it is absolute piracy. And I'm talking about major-label recording contracts, not Napster." Love is among the musicians raising the revolutionary notion that the Internet may offer them a better shake than the labels, which they accuse of imposing burdensome contracts. ⁴⁶

      Scientists are also proposing that current practices by publishers holding copyright and in some cases limiting other distribution of journal articles be reexamined. They believe that scientific authors—especially of works based on government-supported research—should give prospective publishers a nonexclusive license to use their work in a value-added publication in traditional or electronic form. But the author should retain the right to distribute informally, such as through a Web server for direct interaction with peers. In many instances, they say, publishers enforce tighter controls over Internet copyright, dissemination, and pricing than exist in the traditional print world, and this conflicts with the new environment for communicating about science. ⁴⁷

Collaborative Workers

Intellectual Workers

The Ideology of Authorship   Early eighteenth-century British laws that first identified authors as bearers of portable legal rights in intellectual productions were never intended to benefit working authors. Authorship was used as a maneuver to justify alienating the rights of intellectual workers to their productions and entitling publishers to benefit from commercial exploitation of these products.

      Before 1710, authors typically sold their manuscripts for lump sum payments; afterwards authors parted with rights in their works on the same basis or (with increasing frequency) exchanged those rights for a promise of royalties at contractually fixed rates. Authorship continues to be deployed to support a regime that disassociates intellectual workers from a legal interest in their productions: the work-for-hire doctrine. This rule awards ownership of works produced within the scope of employment to the employer. ⁴⁸

      The Supreme Court's 1989 pronouncement on the subject, Community for Creative NonViolence v. Reid makes the point that in work-for-hire cases, the crucial inquiry is into the hiring party's "control over the manner and means" by which the intellectual product is accomplished. ⁴⁹ In this decision, the Court identified circumstances supporting the conclusion that the commissioned artist was the author of his own sculpture. However, the outcome should only emphasize that the convention is one that often deprives intellectual workers of authorship.

      Courts also continue to marginalize works executed by mere "hired artisans," thus undercutting the claims of these intellectual products to legitimacy as independent products of authorship. The "true" artist is one who had escaped "division of labor" and united reason and imagination in a harmonious alliance that represents the whole person. An example of this tendency is provided by the 1992 court decision in Rogers v. Koons. ⁵⁰

      Before comparing the substantial similarity of the two works in question to determine infringement, the Court compared the differences in working methods of the two artists: Rogers was a complete artist who has a studio where he makes his living in creating, publishing, and otherwise making use of his rights in his photographic work. Koons's production of intellectual works was characterized by extreme division of labor: Certain European studios execute his porcelain works, other studios make his mirror pieces, and a small studio in Italy carves the wood sculptures. In effect, the case for infringement was strengthened by an emphasis on the difference in working methods of the people who created the intellectual products.

Faculty Intellectual Property Rights to Online Courses    Traditionally, there was no discussion between faculty and universities of intellectual property protection of teaching because teaching traditionally was nonreplicable: It could not extend itself outside what might be called the single performance.

      Today, because of the rapid growth of Web-based distance education programs, these rights are becoming a hot issue, and institutions have been working to spell out who owns and controls the materials that appear online. ⁵¹ Some universities are sharing the spoils of ownership with faculty members—as an incentive for them to create online courses—and others are not.

      If software is viewed as traditional scholarly publishing, then professors own the work. ⁵² However, the university has ownership claims if the program is seen as an invention, or a work for hire. Copyright laws hold that an employer owns a creative work that falls within the bounds of the worker's employment, such as a program that helps a professor grade papers, legal experts say. In addition, universities could claim ownership if a professor creates the software as an assigned project or uses university resources, including grants awarded by the school.

      No clear blueprint has yet emerged. Some faculty unions are beginning to look to other fields, such as the movie and recording industry, for ideas. The unions in the entertainment industry have been fighting for the rights to be paid for repeated performances for years, and they are trying to learn how they can work with that. In such an approach, a faculty member would own the rights to online instructional materials and could sell access to various online colleges. In fact, the day when professors make deals as rock stars and athletes do may not be far off.

      But many professors are not comfortable with a world of free-agent academics. One of the dangers they see in modeling contracts and agreements off other industries is the insensitivity to the kind of language used in higher education. The word employee is not a popular term for a university professor. "I would really like to think that I'm a scholar or an educator," some faculty say. Institutions that are keeping their hands on potential royalties acknowledge that the ownership guidelines may have deterred some potential online instructors and that unions probably will raise the issue at future contract talks. It will become more contentious as time passes, many conclude.

Multiauthor Collaboration

Disaggregating Multiauthor Collaboration    Traditional protection fails with the most obvious form of multiauthor creativity: intellectual collaborations. Although copyright law has a category for works created by several authors working together on a preconcerted basis, the consequences that flow from the categorization as a work as one of joint authorship reflect the individualistic bias of traditional protection. In effect, a joint work has several individual authors. Each joint author must possess and should retain the legal prerogatives associated with individual authorship. Far from acknowledging the extent to which participation in a collaboration entails surrender of individual prerogative, law assumes the continued dominance of individual authorship.

      Also, in many instances law refuses to acknowledge existence of joint authorship. The 1976 Copyright Act substantially narrowed the range of circumstances in which what might be called a collaboration in the lay sense is recognized as a joint work in the legal sense. Before 1976, courts held that joint authorship need not be the result of face-to-face collaboration, as when lyrics are added to an existing tune. By contrast, the 1976 Copyright Act defines joint authorship so as to require the intention at the time of the writing is done that the parts be combined into an integrated whole.

      Temporality aside, the requirement for joint works that the contributions be intended to be merged into inseparable or interdependent parts of the whole is restrictive, although it is not new. The insistence of the law on disaggregating collaborations rather than protecting them as joint works is apparent in recent decisions concerning whether the contribution of each joint author must be protectable, copyrightable work.

      Most courts have answered in the affirmative, as revealed in the 1990 decision Childress v. Taylor. ⁵³ Childress, a stage performer, conceived of the idea of a play based on the life of Moms Mabley, a legendary black entertainer. After assembling extensive documentation on Mabley's life, she persuaded Taylor to prepare a script incorporating her ideas and research.

      Childress allegedly misappropriated the contents of the script for her own play, and Taylor brought suit. The Court ruled for Taylor as "sole author" of the contested work on grounds that Childress's ideas and research did not represent a copyrightable contribution to the disputed work. In 1991, the Court of Appeals revisited Childress v. Taylor and conceded that copyright policy does not dictate a clear answer to whether each contribution to a joint work must be copyrightable. It is difficult to see why the contributions of all joint authors need be copyrightable. An individual creates a copyrightable form of expression; the resulting work is no less a valuable result of the creative process simply because the idea and the expression came from two different people.

      The Court of Appeals came as close as anything in law to acknowledging the complexities of collaboration and the reimagination of collaboration by individualistic assumptions of authorship. However, rather than confronting the reality of prevailing practice and revising protection to take account of collaborative production, the Court retrenched into insisting on copyrightable contributions for all "joint" authors.

      In the wake of the 1997 Thomson v. Larson decision and the Second Circuit Court of Appeals' affirmation of that decision, which denied dramaturg Lynn Thomson's claim to joint authorship status with playwright Jonathan Larson and her request for an interest in the royalties, questions about the Childress test for joint authorship were being raised. ⁵⁴ One conclusion was that the provisions of the Copyright Act concerning joint authorship should be read broadly enough to support and promote collaboration between artists. To deny an artist protection under the act by not examining a collaboration from all angles, including how the collaborators work together, is to risk denying an artist an economic incentive to continue to collaborate. ⁵⁵

Coauthorship in Biology    Today, joint authorship tensions arising between researchers who devote themselves to sequencing genomes and those who interpret their results are creating a demand for new approaches to collaboration and coauthorship in biology. ⁵⁶ It has become a well-established tradition among the publicly funded genome-sequencing community to place sequence data on public and freely accessible databases as the sequences are generated. There is nothing to stop others from using those data to do good science.

      Do people putting data and analysis freely on the Web prejudice their chances of publication? The journal Nature believes that genomics databases, like preprint servers and conferences, represent a form of intracommunity networking from which all researchers benefit. Nature does not count them as prior publications. If an exciting result is picked up from such a source by the media, that does not necessarily disqualify a paper from consideration as long as researchers have not preempted peer review or publication by encouraging prior publicity.

      However, the problem, from the point of view of those doing the sequencing, occurs when they are continuing their sequencing while others, perhaps better placed to annotate the sequence, are free to use it to publish biologically useful information. What rights of first publication do the sequencers then have? As the discoverers of the sequence, they surely deserve some credit in the subsequent elucidation of function, credit that should extend beyond a simple reference to the database Web site. Yet once someone else has annotated their sequence data, the sequencers' own ambitions in that direction have effectively been preempted. One could argue that those involved in sequencing cannot have both prompt openness and self-protection and that, once the data are publicly available, it is open season.

      Others might propose the opposite extreme: that the sequencers have the right to coauthorship on papers that spring directly from their efforts. That seems to amount to an extension of publishing practice in biology to recognize the fact that distinct but interdependent roles—in effect, collaboration—are played by those who produce basic information and those who accomplish interpretation.

      Such relationships are already common in the physical sciences; for example, in astronomy the developer of a new detector is sometimes, by agreement, included on initial papers that emerge from the application of that device. In high-energy physics, huge collaborations are necessary, and everyone gets due credit by a listing of the hundreds involved. Although these particular examples may not translate directly to biology, they suggest a way forward.

Database Protection

Copyright Protection of Databases

The Supreme Court held by a unanimous vote in Feist that to pass the constitutional threshold, copyrightable works must possess "some minimal degree of creativity." (See also "Introduction" and "Expansion of Traditional Intellectual Property during the 1990s" earlier in this chapter.) Feist, a firm that specialized in publishing regional telephone directories, was refused a license to use Rural's directory, which covered a small part of the larger geographic area to be encompassed in Feist's projected directory. Feist went ahead and copied Rural's alphabetically organized listings, and Rural sued for copyright infringement. In producing the directory, the Court held, Rural had expended sufficient efforts to make the directory useful but insufficient creativity to make it original. As far as the opinion reveals, the Court was moved solely by its adherence to a vision of authorship originality. The opinion wears its values on its sleeve; its rhetoric proceeds from unreconstructed faith in the gospel of romantic authorship.

      As a matter of information policy, the Court may have been right to cut back on available grounds of legal protection for data compilations. There is a point, legal scholars say, where too many property rights owned by too many parties creates a legal "smog," or an "anticommons." ⁵⁷ When this situation is reached, rethinking and rebundling of property rights may be necessary.

      Observed in this light, Feist might be seen in part as an attempt to avoid a "tragedy of the anticommons," an attempt to make sure that the threshold for copyright ownership was not set so low that all sorts of infringement claims would arise, resulting in the underuse of facts and data from public domain sources.

      Whether the U.S. Congress should give databases greater legal protection became a controversial issue. The concern of many opponents was that extending copyright protection to compilations that do not reflect original or creative work would lead to inappropriate and intrusive limits on the flow of scientific information or raise the costs of using such information to a level that would impede scientific progress. ⁵⁸ Later scientists and engineers could not combine data legitimately accessed from one commercial database with data extracted from other databases to make a complex new database for addressing hard problems without obtaining additional licenses and permissions. This remains perhaps the single most critical problem for scientific and technical research, legal scholars say. ⁵⁹

Electronic Journals as Databases

Electronic databases of DNA sequences, protein structures, or galaxy images permeate the Web, but the Web also opens the way to weaving electronic journals and scientific libraries into a single interconnected database. Already a hyperlink can take a subscriber from one article to related articles in the same journal, other journals, and resources such as electronic databases. By offering authors' raw data or the software used in the analysis, some of the journals even allow readers to double-check an author's work.

      As journals become increasingly interconnected, researchers will find themselves hyperlinking from one cited or related article to the next, regardless of who the original publisher is. If the browsing is done in a single publisher's database, it could be covered by a single subscription to the database. But if the hyperlinks connect articles or databases of different publishers, then it will result in "buying by the glass rather than the bottle." ⁶⁰ One can imagine a variety of licensing schemes in which multiple publishers or holdouts might exercise their ownership and place limits on what researchers can do in acquiring and using the hyperlinked journals and databases.

Cross-Licensing Schemes

But particularly when intellectual products are multicomponent systems, when broad protection on components led to blockages, they were resolved, in some cases, by the development of (sometimes elaborate) cross-licensing schemes. The staggering complexity of the scholarly journal landscape may motivate publishers to attempt to minimize litigation and other transactional costs by voluntarily entering into cross-licensing arrangements that are tantamount to scholarly publishing pools of intellectual properties.

      In fact, there are signs that cross-licensing schemes are emerging in scientific publishing to solve some problems. Scientists and publishers increasingly view titles as merely part of hyperlinked content databases made up of constellations of journals. User profit and nonprofit organizations unveiled a scheme in 1999 designed to, in effect, cross-link journal articles through their reference lists, making it easy for researchers to locate and obtain the text of a referenced article through the Internet. ⁶¹ The publishers had consulted their legal advisers and concluded that they would not run afoul of antitrust laws, which prohibit collusion between competitors, as long as they do not try to exclude anyone. The lead organizers would work with the International Digital Object Identifier (DOI) Foundation to devise "tags" that can be used to find and track the journal articles. "We are trying to get away from single journals and think more in terms of developing knowledge environments that integrate a number of relevant sources," said Oxford University Press. ⁶²

Conclusion

In the early 1990s, the potential of electronic networks attracted attention in a variety of national and international forums. These discussions emphasized types of technology that enabled new modes of production for intellectual works that were beyond automation and portended a transformation where authors became far more collaborative and more iterative modes of intellectual production for an electronic environment where more discrete blocks of knowledge existed before.

      This chapter has sought to demonstrate how the persistence of the notion of authorship—individual originality—continues today and makes it difficult for any new legal synthesis that would focus on iterative works or collaborative modes of intellectual production. The law continues to prove ungenerous to multiauthor intellectual products, and such works become marginalized or invisible within the prevailing ideological framework of authorship.

      Thoughtful action to reshape traditional protection would advance the transformational potential of today's digital technologies and electronic networks. Initiatives in this area are particularly appropriate in intellectual communities in light of the growing urgency of sound policy and the clear benefits to be generated for scientific and technological progress by international collaboration in electronic networks. Will the intellectual property system be likely to address these issues head on? It continues to be frustrating for intellectual workers to realize that the system rarely does that. But still these concerns must be addressed for the early twenty-first century because "the wrong decisions today could lessen the vitality of our research enterprise, weaken the national system of innovation, and compromise our future technological superiority, which all depend on maintaining an appropriate balance between upstream and downstream uses of data and factual information." ⁶³

Notes

      1. James J. Duderstadt, "Plenary Address: An Information Highway to the Future," National NET '91 Conference, Washington, D.C. (March 27, 1991).

      2. Jenny Lyn Bader, "Forget Footnotes: Hyperlink." New York Times, July 16, 2000.

      3. Judy Redfearn, "OECD to Set Up Global Facility on Biodiversity," Science 285: 5424 (July 2, 1999): 22-2

      4. David E. Rosenbaum, "Database Legislation Spurs Fierce Lobbying," New York Times, June 5, 2000, A14.

      5. 111 S.Ct. 1282 (1991).

      6. J. H. Reichman and Paul F. Uhlir, "Database Protection at the Crossroads: Recent Developments and Their Impact on Science and Technology," Berkeley Technology Law Journal 14 (1999): 837.

      7. Peter Jaszi, "Toward a Theory of Copyright: The Metamorphoses of 'Authorship,'" Duke Law Journal 2 (1991): 455-502; Peter Jaszi, "On the Author Effect: Contemporary Copyright and Collective Creativity," in The Construction of Authorship, ed. Martha Woodmansee and Peter Jaszi (Durham, N.C.: Duke University Press, 1994), 29-56.

      8. Keith Aoki, "Neocolonialism, Anticommons Property, and Biopiracy in the (Not-So-Brave) New World Order of International Intellectual Property Protection," Indiana Journal of Global Legal Studies 6 (1998): nn. 84-86, <http://www.law.indiana.edu/glsj/vol6/no1/aoki.htm>.

      9. Nicholas Wade, "Background Paper," in The Science Business: Twentieth Century Fund Report on the Commercialization of Scientific Research (New York: Priority Press, 1984), 30-31.

      10. Carole Ganz-Brown, "Patent Policies to Fine Tune Commercialization of Government-Sponsored University Research," Science and Public Policy 26: 6 (December 1999): 403-14.

      11. John Barton, "Reforming the Patent System," Science 287: 5460 (March 17, 2000): 1933-34.

      12. 1980 Computer Service Copyright Act, enacted December 12, 1980, PL 96-517. See also Ann Branscombe, Who Owns Information (New York: Basic Books, 1994), 142.

      13. 797 F.2d 1222 (3d Cir. 1986).

      14. Seth Shulman, "Software Patents Tangle the Web," Technology Review 103: 2 (March/April 2000): 68-79, <http://www.technologyreview.com/magazine/mar00/shulman.asp>.

      15. Julie E. Cohen and Mark A. Lemley, "Patent Scope and Innovation in the Software Industry," California Law Review 89 (January 2001): 1-58, <http://papers.ssrn.com/paper.taf?ABSTRACT ID=209668>; 149 F.3d 1368 (Fed. Cir. 1998).

      16. Final Act Embodying the Results of the Uruguay Round of the Multilateral Negotiations, April 15, 1994, Annex 1C, Legal Instruments—Results of the Uruguay Round, vol. 31; 33 I.L.M. 81 (1994), <http://www.wto.org/wto/intellec/1-ipcon.htm>.

      17. Directive 96/9/EC of the European Parliament and of the Council, March 11, 1996, <http://www2.echo.lu/legal/en/ipr/database/database.html>.

      18. WIPO Diplomatic Conference on Certain Copyright and Neighboring Rights Questions, Geneva, December 2-20, 1996, <http://www.uspto.gov/web/offices/dcom/olia/diplconf/>.

      19. James Boyle, Shamans, Software, and Spleens: Law and the Construction of the Information Society (Cambridge, Mass.: Harvard University Press, 1996), x-xi.

      20. Robert Mazzoleni and Richard R. Nelson, "The Benefits and Costs of Strong Patent Protection: A Contribution to the Current Debate," Research Policy 27 (1998): 273-84.

      21. Gary Taubes, "Computational Molecular Biology: Software Matchmakers Help Make Sense of Sequences," Science 273: 5375 (August 2, 1996): 588-90.

      22. Aris Persidis, "Bioinformatics," Nature Biotechnology 17: 8 (August 1999): 828-30.

      23. Sylvia J. Spengler, "Bioinformatics in the Information Age," Science 287: 5456 (February 18, 2000): 1221-23.

      24. John Barton, "Reforming the Patent System," Science 287: 5460 (March 17, 2000): 1933-34.

      25. U.S. Patent 5,933819, August 3, 1999.

      26. 383 U.S. 519 (1966).

      27. Fed. Cases 342, No. 4901 (C.C.D. Mass. 1841).

      28. 62 U.S.L.W. 4169.

      29. Carole Ganz-Brown, "Electronic Information Markets: An Idea Whose Time Has Come," Journal of World Intellectual Property 1: 3 (May 1998): 465-93.

      30. Westinghouse v. Boyden Power-Brade Co., 170 U.S. 537, 562 (1898).

      31. Pamela Samuelson, "Encoding the Law into Digital Libraries," Communications of the ACM 41: 4 (1998): 13-18.

      32. The case A&M Records Inc. v. Napster Inc., 99-5183, is being considered in tandem with a similar suit filed by a group of music composers, Leiber v. Napster, 00-0074; <http://thomas.loc.gov/cgi-bin/query/z?c105:H.R.2281.ENR> has the full text of the Digital Millennium Copyright Act (DMCA) H.R. 2281; Scott Carlson, "Metallica Sues Universities and Napster, Charging That Students Engage in Music Piracy" Chronicle of Higher Education, April 28, 2000, A50.

      33. Shane Ham and Robert D. Atkinson, Napster and Online Piracy: The Need to Revisit the Digital Millennium Copyright Act (Washington, D.C.: Progressive Policy Institute, 2000), 1-2, <http://www.dlcppi.org/texts/tech/napster1.htm>.

      34. P.L. 105-304; Jeri Clausing, "Report Proposes Update of Copyright Act," New York Times, May 22, 2000, C-6.

      35. Roger Parloff, "Newbies vs. Netwits," Law.com, September 12, 2000, <http://www.law.com/>.

      36. MPAA v. 2600 (S.D.N.Y., August 8, 2000).

      37. Opinion, <http://eon.law.harvard.edu/openlaw/DVD/NY/opinion.pdf>; final judgment, <http://eon.law.harvard.edu/openlaw/DVD/NY/finaljudgment.pdf>.

      38. 38 Jane C. Ginsburg, "International Copyright: From a 'Bundle' of National Copyright Laws to a Supranational Code?," Journal of the Copyright Society of the USA 47 (2000): 265-91.

      39. Jenny Lyn Bader, "Forget Footnotes, Hyperlink," New York Times, July 16, 2000, Week in Review Desk.

      40. Martin H. Sampson, "Hyperlink at Your Own Risk," Law Journal Extra, June 24, 1997, Outside Counsel, <http://www.phillipsnizer.com/artnew18.htm>.

      41. eBay Inc. v. Bidder's Edge Inc. (N.D. Calif., No. 0-99 21200, filed 12/10/99).

      42. Brenda Sandburg, "Judge Halts E-Bay's Unwanted Hits," law.com California, May 26, 2000, <http://www.callaw.com/opinions/stories/edt0526b.html>.

      43. UMF Recordings Inc. v. MP3.com Inc. (S.D.N.Y., No. 00-0472).

      44. Lessley Anderson, "Judge Explains His Ruling against MP3," Industry Standard, May 4, 2000, 07: 45 PM PST, <http://www.thestandard.com/article/display/0,1151,14807,00.html>.

      45. Michael Learmonth, "The Price of Digital Music," Industry Standard, June 12, 2000, <http://www.thestandard.com/article/display/0,1151,15920,00.html>.

      46. Alec Foege, "Record Labels Are Hearing an Angry Song," New York Times, June 11, 2000.

      47. Steven Bachrach, R. Stephen Berry, Martin Blume, Thomas von Foerster, Alexander Fowler, Paul Ginsparg, Stephen Heller, Neil Kestner, Andrew Odlyzko, Ann Okerson, Ron Wigington, and Anne Moffat, "Who Should Own Scientific Papers?" Science 281: 5383 (September 4, 1998): 1459-60.

      48. The patent system has no work-for-hire doctrine. However, in the United States by the 1930s, the patent system initiated in 1790 was under attack on the grounds that its original purpose had been rendered obsolete: Corporate enterprise had undoubtedly displaced the solo inventor as the primary performer of inventive activity. Industrial organizations obviously had not eliminated originality, however, but simply transferred it from the individual scientist, now largely an employable factor in research and development to the corporation.

      49. 490 U.S. 730 (1989).

      50. 751 F. Supp. 474 (S.D.N.Y. 1990).

      51. Dan Carnevale and Jeffrey R. Young, "Who Owns On-Line Courses?: Colleges and Professors Start to Sort It Out," Chronicle of Higher Education, December 17, 1999, A45, Information Technology Section.

      52. Scott Carlson, "When Professors Create Software, Do They Own It, or Do Their Colleges," Chronicle of Higher Education, July 21, 2000, A29, Information Technology Section.

      53. 945 F.2d 500 (1991).

      54. 147 F.3d 195 (2d Cir. 1998).

      55. Faye Buckalew, "Joint Authorship in the Second Circuit," Brooklyn Law Review 64 (Summer 1998): n. 217.

      56. Nature Opinion, "Debates over Credit for the Annotation of Genomes," Nature 405: 6788 (June 15, 2000): 719.

      57. Michael A. Heller and Rebecca S. Eisenberg, "Can Patents Deter Innovation?: The Anticommons in Biomedical Research," Science 280: 5364 (May 1, 1998): 698-701.

      58. William Gardner and Joseph Rosenbaum, "Database Protection and Access to Information," Science 286: 5445 (November 26, 1999): 1658.

      59. Reichman and Uhlir, "Database Protection," 808.

      60. Gary Taubes, "Science Journals Go Wired," Science 271: 5250 (February 9, 1996): 764.

      61. Eliot Marshall, "Journals Launch Private Reference Network," Science 286: 5444 (November 19, 1999): 1459.

      62. "The Writing Is on the Web for Science Journals in Print," Nature 397: 6716 (January 21, 1999): 195-200.

      63. Reichman and Uhlir, "Database Protection," 808.