Rights and Responsibilities of Participants in Networked Communities (1994)

Protection of intellectual property interests, whether by copyright, trade secret, patent, trademark, export law, or even encryption, is no simple matter even in traditional communications media. Restrictions on photocopying are violated routinely, though certain types of photocopying are protected by law under fair-use provisions. Pop music is created through "sampling" of other compositions, a practice of debatable legality that has parallels in electronic communications. National approaches to property protection differ, while the status of international copyright law is somewhat murky¹; and some U.S. laws related to intellectual property are so arcane that users may not even be aware of their existence or applicability.

Within this sphere of uncertainty, electronic networks generate special concerns, and indeed the Internet itself is playing a significant role in changing the nature of publication.² For example, in the research community, professional societies are investigating ways to offer their journals through the Internet, and some have already begun to do so. Many researchers would like to effectively circumvent the traditional publication process, which can lead to years of delay, and make their work available on the network for free. In other instances, many individuals hesitate to use the Internet at all, fearing—with some arguable justification—that material they created may be appropriated by others without permission.

The research community is not the only one affected. The on-line availability of various publications raises questions regarding the limits of electronic redistribution of print articles in general. In June 1994, the ClariNet Communications Corporation, publisher of an Internet newspaper, was asked by Knight-Ridder Tribune and its Tribune Media Services Division to cease its electronic publication of syndicated columns written by Dave Barry and Mike Royko, due to concerns about too much information piracy occurring on "the net." It seems that a subscriber to ClariNet, the electronic newspaper, sent a copy of a Dave Barry column from the ClariNet, presumably by e-mail, to a nonsubscriber mailing list, where it then reached a Knight-Ridder employee who reported it to executives at Knight-Ridder. In July 1994, the Working Group on Intellectual Property Rights of the administration's Information Infrastructure Task Force released a draft report on how copyright law and practices may need to be updated in an age of highly interconnected electronic networks. The report proposes, for example, that existing law be clarified to ensure that copyright law protects the creator of works that are disseminated through electronic networks.³

Although intellectual property is traditionally the domain of copyright, patents, and trade secrets, most of the discussion of intellectual property

matters in a networked environment is related to copyright, and discussions at the February 1993 forum reflected this weighting. Both scenarios in this chapter are university-based because the university is a domain in which the clash among competing values and rights (e.g., the value of the free and easy availability of knowledge and information versus the right to receive compensation for one's intellectual work) is often played out most clearly.

The Copyright Act of 1976 (Public Law 94-553) protects "original works of authorship" that are "fixed in a tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated either directly or with the aid of a machine or device."⁴ The owner of a copyright has exclusive rights to (1) reproduce the copyrighted work; (2) prepare derivative works; (3) distribute copies to the public by rental, lease, or lending; (4) perform the work; and (5) display the work publicly.⁵ These and other provisions are delineated in the United States Code Annotated, Title 17, Sections 101-120.

Copyright protection clearly applies to certain types of electronic materials. The 1980 amendments to the Copyright Act of 1976 expressly protect computer programs as literary works. The original expression of information in databases can be copyrighted, though the facts themselves cannot, a position reinforced by the Supreme Court decision in Feist v. Rural Telephone (see Chapter 3). This case eliminated copyright protection for white pages telephone directories, and some believe that it has had a significant impact on the business and practices of on-line database providers.

Original and creative text can be copyrighted as well, although enforcing this protection can be onerous. An example raised by attorney Lance Rose in the November 1992 workshop is that long e-mail messages could probably be protected under copyright law, but for practical reasons (i.e., the legal costs) there may be no legal recourse for violations. Other original and creative intellectual creations whose status is ambiguous under current law include the "look and feel" of an on-line service and a short e-mail message created by one person but posted publicly by another.

For users and system operators, it can be difficult to identify

copyrighted materials, because current legal practice does not require posting of a copyright notice (though notice is still required to qualify for recovery of certain damages). Thus, the burden is on the user to know the law. Nevertheless, at least some network operators, notably commercial information services, strive to keep copyrighted materials off their networks. Commercial software is fairly easy to identify and remove, but graphics are more difficult, according to Stephen M. Case, president of America OnLine. (Even if images are copyrighted, it is relatively simple to strip copyright notices from them in such a way that there is no indication that they are protected by copyright.) Case takes this responsibility to heart, even though he employs contractors to review system files.⁶

A 1993 federal court case, Playboy Enterprises Inc. v. Frena, described in Chapter 3, underscores the need for caution, as it held that a finding of copyright infringement did not require an intent to infringe. In another case involving the posting of copyrighted video games on a public system with the knowledge and encouragement of the system operator, a federal court granted a temporary injunction in favor of the copyright holder on the grounds that it was unlikely that a fair-use defense would be successful.⁷

The library of a large university connected to the Internet subscribes to an electronic database (accessible via a gateway on the library system) and an electronic journal (to be distributed to subscribers by electronic mail). The database owner provides access for up to six users at a time, for a flat fee to be paid by the library. A student exceeds the six-user limit on the database.

Forum participants generally agreed that the database subscription agreement, or license, should be enforced by the university, not the provider, in Scenario 1. Karen Hunter, vice president and assistant to the chair of Elsevier Science Inc., said her company asks its licensee universities to "reinforce the notion of copyright" and to stress the ethics of database use with faculty and students. Hunter's chief concern is not how but whether the university tries to enforce its agreement. The six-user limit in the scenario presumably applies to a price, she noted; if the university cannot keep the system secure, then the price needs to be renegotiated.

Robert A. Simons, general counsel and secretary for DIALOG Information Services, said universities have a duty to comply with their subscription agreements; the question is whether their controls are reasonable and, if a student somehow bypasses reasonable controls, whether the university has violated the contract. Such an incident probably would not concern the provider, he said, unless it occurred frequently and the university obviously was not exercising any control.

If a faculty member violated the user limit, then the infraction would become more serious, because a faculty member may be viewed as an agent of the university, Simons said. "[For] faculty members, I think there is an even higher duty, not only on their part but particularly on the university's part to educate them as to what is right and what is wrong," he said. "I think the university has an affirmative [duty] to publish 'do's' and 'don'ts' with respect to its network system, much as it has the duty to post a sign on its photocopy machine in order that it not be deemed a vicarious or contributory infringer."

Peter W. Martin, a Cornell Law School professor and former dean, said the university should have a rule and a disciplinary mechanism for dealing with the student in Scenario 1. University rules and regulations traditionally address issues that "get at the core of the academic enterprise" and also affect the university community, and this situation meets those criteria, he said. "[T]he traditional mind-set would be to say to any third party …, 'That's an internal matter for us; you can't force us to discipline.' And unless a faculty member or library administrator or some other student cares and sort of launches a disciplinary proceeding, it won't happen." At Cornell, Martin said, administrators would care enough about the contractual agreement with the database provider that they would initiate a disciplinary proceeding.

Nancy M. Cline, dean of university libraries at Pennsylvania State

University, was concerned about the need to be responsive to copyright issues and said that expectations of behavior can be conveyed through policy guidelines, faculty contracts, and student codes of conduct; once such expectations are conveyed, the personal responsibility for appropriate use rests with individuals. Beyond that, she was concerned about the prospect of being asked to enforce certain types of contractual provisions having to do with access, pointing out the difficulty of doing so as well as the concern for privacy rights. Her purchasing decisions are based in part on the expectations and obligations tied to particular products. For example, Cline can enforce concurrent user limitations but has difficulty predicting the number of users (a total some providers request to assure profits). It is difficult to "predict safely some of those boundaries without jeopardizing the nature of scholarship," she said.

In that vein, she warned against focusing strictly on tight access controls—whether through strict limitations on the number of users or through the imposition of "per-access" fees that would effectively exclude the majority of students—without considering the purposes of education and research. Universities should not be "blocking off access [to knowledge] and controlling it so rigorously through contract or other types of containment processes … that it would change the very nature of scholarship and research. … There's a vitality to a lot of research that is predicated on people being able to discover and find and use things that may not necessarily come to them through the usual channels." She continued, "There is a great deal of discovery that goes on in a university context that is serendipitous. People wander into things and thereby branch out and discover a new technique, a new application, whatever. … I would hate to see a graduate from microbiology go out into the workplace never having been able to use MEDLINE⁸ or some of these other [network] resources just because he or she was the twenty-first student in a 20-workstation contract." She further argued that limitations on access to certain materials reduce the likelihood that those materials will be used and that if people cannot readily use these materials, there may be no need for these materials to be generated in the first place.

Panel members could not offer an ideal mechanism for restricting use. Simons said counting individual users is almost impossible and makes little sense: "Publishers were almost unanimous in agreeing that counting noses or heads is not the same as counting workstations. …

It's almost impossible, and it's an impractical approach to try to count noses. It's as if one is saying that without a copyright issue, I'm going to post something with a thumbtack on a bulletin board, but only if you allow 20 people to look at it, and not 21. … That doesn't make much sense in either a copyright environment or a commercial environment." On the other hand, the number of workstations with simultaneous access to the relevant files could be restricted. But Ronald L. Plesser, an attorney specializing in new information technologies, questioned whether physical access should be limited. He noted that many universities issue personal identification numbers that can be used at various workstations. He said that pricing strategies for handling such situations will evolve in the marketplace.

Pamela Samuelson, a professor of intellectual property law at the University of Pittsburgh, expressed concern about requiring users to know and abide by all the details of many different licensing agreements. She agreed that universities should foster discussions and establish guidelines regarding appropriate behavior on electronic networks, suggesting that universities help foster a consensus about what is "reasonable within that environment."

Plesser and Simons said current copyright law works fairly well in the electronic environment, though improvements are certainly possible. Plesser said the breakdown comes in enforcement. There are two problems, he said: how to know when someone takes proprietary electronic material and plagiarizes it or alters it, and how to identify the perpetrator so that he or she can be held liable for damages or payment. "We've got to create maybe some new mechanisms of enforcement, collection, [and] payment," he stated.

Emphasizing the distinction between licensing and copyright protections, Plesser said licenses may be issued both for copyrighted works and for databases that may not be copyrighted, and so copyright law is not always relevant to property protection. "The fact that you can't copyright [material] or may not be able to enforce the copyright does not necessarily mean you can't license it, control the use, and control the issue," he said. "So I think the law comes from two places." He noted that the Feist case (described in Chapter 3) raised some of these issues.

Simons predicted that a mechanism will be developed whereby users may obtain duplicates of copyrighted electronic materials for a fee; a notice specifying acceptable and unacceptable uses of the materials could be placed on copyrighted material so that users would be sure of the restrictions. A system of this nature would transform electronic networks used for information access into an electronic

bookstore, he suggested. ''Once we do that, those publishers and authors who are currently worried about placing their valuable data into this morass … will say, 'Aha, it's now a marketplace where I can go and buy and sell my wares. …'"

Computer and network technology increases the ease with which electronic text can be copied, distributed, and altered. Thus, David Johnson wondered "if the copyright statutes that arose out of the print mechanisms are themselves becoming hopelessly outdated." He went on to note that we may need to "give up on an effort to protect particular artifacts of electronic text as if they were property and instead focus on how to control access to information through particular channels"; the proper route, he said, was to address "the repetitive distribution of information through channels like libraries and universities and institutions" and what that type of distribution would do to the nature of life in the affected communities.

Licensing agreements for the use of electronic documents alter a university's approach to discipline significantly, by effectively imposing a new definition of fair use,⁹ according to Martin. As a condition of providing service, the provider has nearly unlimited power to place any restraints whatever on usage as long as the university agrees to them in advance, he said. Martin described the agreement as overriding fair-use provisions, in that "the leverage is there for the provider to assert through license his own view of what is appropriate distribution of that information, without regard to what fair use under the copyright laws would allow." Martin further argued that "what's new about [the networking environment] and its licensing framework via the universities is that universities cannot, as they

could with print materials, put up signs next to the copy machines [describing appropriate use] and then leave faculty and students to their own judgments, protected by fair-use provisions."

Samuelson, however, said that fair use is determined not by publishers but rather by law. "And I think it's really important to note, especially when you are dealing with university contexts, that there has to be some room for fair use that the community can participate in and not really rely on publishers to define what the scope of use is." Copyrighted electronic materials are covered by the fair-use provisions in copyright law, according to Samuelson; notions of fair use have been extended to the electronic environment in at least two precedent-setting legal cases.¹⁰

In general, Samuelson said, users are becoming increasingly sensitive to the boundaries of fair use, and a consensus is evolving that caution should be exercised in reproducing and widely distributing the material of other authors without explicit permission. She suggested that copyright law would remain as "a kind of net" but that new mechanisms for handling these issues will evolve as the Internet is used-increasingly for commercial services. She went on to predict "a kind of transition from today's environment, in which people basically exchange information without the expectation of the commercialization of that environment, to an environment where commercialization is a more routine [mechanism for distribution]."

Providers indicated that the question of who should determine fair use is difficult to answer. Hunter said publishers are not so much "dictating" fair use as trying to fill a legal void. Part of the problem is that, although the copyright guidelines of the National Commission on New Technological Uses of Copyrighted Works (an advisory commission established in the late 1970s to address copyright issues as they related to new technologies) address fair use in library photocopying, there appears to be no legal precedent for electronic networks. "I think [fair use] will be defined by … negotiation and use and a lot of other things, but I don't think it's fair to say that the law provides for fair use in the electronic environment," Hunter

said; "I don't think we have that yet." The confusion also was noted by Plesser, who represents trade associations and private companies.

Some forum participants suggested that concerns about property protection keep potential contributors out of the electronic environment. In running a legal information institute at Cornell that distributes materials on the Internet, Martin finds that many individuals are afraid to contribute because any contribution they make can be appropriated by others: "They consider it like dumping some valuable watches into Central Park—[they] will be gone." He also discerned considerable confusion over the application of copyright law on electronic networks, particularly with regard to how copyrighted information may be used without the author's permission. Plesser agreed that many academics refuse to put their software, texts, or other materials on the Internet, for fear of losing "the value of their creation."

Hunter said the provider community is divided on this issue. "There are a lot of publishers who are very reticent about the whole network environment," she said. "There are others who say it's a reality, it's the future, and we have to get in and test it." (In the period since Hunter made these remarks, such tests have increased in number.¹¹) She likened the situation to a story about the designer of a new campus who constructed the building but postponed laying down sidewalks for about six months, to "see what paths people generate. That's the kind of world we're in. I think we're trying to see how … [people] want to use the information."

Cline, on the other hand, said some groups of faculty are collaborating on the Internet with great intensity, having established community understandings about network use. Acceptance of the technology has gone so far that some institutions accept electronic publications in tenure evaluations, she said. She acknowledged the legitimacy of concerns about the quality of material "published" electronically, but she did not believe that such publications, suitably juried or reviewed, would be handicapped in comparison to print publications.¹²

A university is connected to the Internet. Under a joint effort of its alumni relations and industrial liaison programs, the university also provides library and Internet access for Company X, a small start-up business founded by university alumni, in return for stock options in Company X. To facilitate private communications, the university provides RSA-based public-key encryption software¹³ on its host computers, encourages the software's use, and maintains databases that facilitate the lookup of the public-keys of all users using the university as a node. Someone from Iraq accesses the university computers and downloads the software [which is not approved for export to Iraq].

Martin, as the owner of an Internet mailbox that may be accessed from foreign locations, said he feels no duty to monitor users who browse through his system. He is not concerned about someone adding or manipulating files because he controls those types of activities.

Other participants felt obliged to monitor their systems. Hunter said providers have some obligation to inspect their materials for violations of law. In the case of journal publication, inspection may be a responsibility of the editor or reviewer of articles to be published. Simons views the export issue as so important that his licensing agreements often require publishers to refrain from using any material that would constitute "encryption data" under the International Traffic in Arms Regulations or U.S. Department of Commerce rules.¹⁴ Otherwise, it would be possible to inadvertently "export"

such data and thus break the law, he said. He suggested that according to certain views of export control, the mere presence of foreign nationals in attendance at an unclassified domestic conference could be considered an export of technical data.

Lance Hoffman wondered whether faculty and universities understood their obligations at all. He cited his experience in teaching a class on encryption that includes students who are foreign nationals; all assignments are done on-line. This combination of factors raises the possibility of inadvertent "export" of materials that could be considered encryption data, which are treated as munitions and barred from export by U.S. law (the International Traffic in Arms Regulations). Hoffman said he doubted that university counsels are familiar with this law.

Participants said the university would be obliged to turn over certain records to authorities as specified by law. Alan McDonald, a special assistant to the assistant director of the Technical Services Division of the Federal Bureau of Investigation (FBI), said an FBI request for transaction records would require cooperation from the university, like any other service provider. The situation probably is covered either by the Electronic Communications Privacy Act (ECPA) or the general process of federal criminal procedure; the FBI might obtain a subpoena under United States Code Annotated, Title 18, Section 2703, he said.

Plesser agreed that the university would have to comply with the ECPA, assuming that the university, for billing purposes, kept records of which materials were accessed and that law enforcement authorities requested that information. The action required would depend in part on whether the request was for transactional or content data, a distinction that is difficult to make in the scenario, he said. (This issue is discussed further in Chapter 7.)

"The university can't just hand [the data] over on request," Plesser said. "They have to make sure that there is a legal process and that

it's quite defined, and that they are not notifying the person to whom [the information] relates."

The protection of intellectual property proprietary interests is problematic even outside the networking context. Concerns for such protection have become more widespread with the advent of easily available copying technologies such as photocopiers, tape recorders, and video cassette recorders (VCRs). Seen in this light, computer networks are yet another technology to facilitate copying.

At the same time, computer networks have properties that set them apart from other copying technologies. For example, computer networks make easy the large-scale redistribution of copied information. With photocopiers and VCRs, a separate copy must be made for each new recipient. However, on computer networks it can be as easy to send a copy of an electronic document (a book, a musical recording, a photographic reproduction of a painting) to a thousand people as to one person, and sending it to one can be trivial indeed.¹⁵ Moreover, while the laws of physics impose definite limitations on the extent of analog copying, digital copies can be made with perfect fidelity; thus digital copies can be made ad infinitum, vastly increasing the potential scope of redistribution through the remote access capabilities enabled by networks. While a photocopy of a book or a pirated video cassette recording or even a floppy disk must be physically carried from place to place to transfer information, a computer network can transfer information without such restrictions. Thus, laws and regulations based on impeding the movement of physical objects across geographical boundaries are increasingly difficult to enforce in a networked environment.

Previous copying technologies such as tape recorders and photocopiers have resulted in a legal regime based on the concept of fair use; much of the debate today revolves around the extent to which now-traditional concepts of fair use can be sustained in a networked environment. In addition, electronic digital technology changes the economic calculus. Before digital media and the network, the limited number and quality of copies and the limited distribution of those copies made individual copying insufficiently significant economically

to merit enforcement. Digital reproduction and network distribution enable much broader circulation of higher-quality reproductions at the same expense of money and effort and thus have very different economic consequences and implications.

Finally, digitally stored information can be altered with relative ease. The same technology that enables perfect reproduction also allows changes to be introduced in reproductions with relative ease. A few keystrokes can modify an electronic document; absent the use of technology to guarantee that no changes have been made, such changes are undetectable.¹⁶ For example, digital photographs can be modified to show scenes that never existed, and the viewer will never know that they have been modified. Coupled with the ability to send anonymous or pseudonymous transmissions, the potential to alter information in largely undetectable ways raises new concerns and reinforces old ones about maintaining the integrity of information and protecting the rights of the creators of such information.