WARNING: White space and line breaks may not be correct, but page numbering should be accurate. The index has been slightly changed so that it reflects the contents appropriately changes are marked with square brackets.
PAGE 1 LIBRARY OF CONGRESS + + + + + UNITED STATES COPYRIGHT OFFICE + + + + + HEARING ON EXEMPTION TO PROHIBITION ON CIRCUMVENTION OF COPYRIGHT PROTECTION SYSTEMS FOR ACCESS CONTROL TECHNOLOGIES + + + + + DOCKET NO. RM 9907 + + + + + Thursday, MAY 18, 2000 + + + + + The hearing in the above-entitled matter was held in Room 290, Stanford Law School, Crown Quadrangle, Stanford, California, at 2:00 p.m. BEFORE: MARYBETH PETERS, Register of Copyrights DAVID CARSON, ESQ., General Counsel RACHEL GOSLINS, ESQ., Attorney Advisor CHARLOTTE DOUGLASS, ESQ., Principal Legal Advisor ROBERT KASUNIC, ESQ., Senior Attorney Advisor PAGE 2 I-N-D-E-X Panel I: Siva Vaidhyanathan ...............................[.9 ] New York University Karen Coyle .....................................[.21 ] California Digital Library University of California Linda Crowe .....................................[.30 ] American Library Association Laura Gassaway ..................................[.34 ] American Association of Universities, American Council on Education, and the National Association of State Universities and Land-Grant Colleges [ Questions and Answers ...........................47 ] PAGE 3 P-R-O-C-E-E-D-I-N-G-S (2:15 p.m.) MS. PETERS: Good afternoon, and welcome to Stanford University Law School. We'll be hearing from one panel this afternoon, and we'll begin again tomorrow at 9:30 in the morning to hear two panels throughout the course of the day. The schedule for the Stanford hearings is available today outside, and it's also posted on our website. First of all, I'd like to thank Stanford University Law School for agreeing to host these hearings, and in particular we thank Professors Hank Greely and Paul Goldstein and Julie Viner, the Law School's Director for Special Events for all their assistance. We're very pleased to be here and we're grateful to the university and law school for making these facilities available to us. As you probably know, these hearings are part of the ongoing rulemaking process mandated by the Congress under Section 1201(a)(1) of Title 17 of the United States Code. Section 1201 was enacted in 1998 as part of the Digital Millennium Copyright PAGE 4 Act. And I look out and see there are some people who are not too thrilled about this Act. Nevertheless, it is in force, and Section 1201(a) provides that no person shall circumvent a technological measure that effectively controls access to a copyrighted work. However, this prohibition does not go into effect until October 28th of this year, which is two years after the DMCA went into effect. Section 1201(a) provides for this rulemaking in which it's the Librarian of Congress who may exempt certain classes of works from the prohibition against circumvention of technological measures that control access to copyrighted works. The purpose of the rulemaking proceeding is to determine whether there are particular classes of works as to which users are, or are likely to be, adversely affected in their ability to make non- infringing uses if they are prohibited from circumventing technological access control measures. Pursuant to the Copyright Office's Notice of Inquiry, which was published in the PAGE 5 Federal Register on November 24, 1999, we received 235 Initial Comments and 129 Reply Comments, all of which are available for viewing and downloading on our website. Two weeks ago, we conducted a first round of hearings at the Library of Congress in Washington. After the hearings here at Stanford, we will accept a final round of post-hearing comments. These post-hearing comments are due on Friday, June 23rd. In order to allow interested parties adequate time to respond to the hearing testimony, we intend to post the transcript of all hearings on our website as soon as the transcripts are available. We are also recording the testimony for streaming and possible downloading from the Office's website. The audio files from the hearings at the Library of Congress are currently available on our website. The transcripts will also be posted on the website as originally transcribed, but obviously everybody who testifies will have an opportunity to correct any errors in these transcripts. When those corrections are received, we will put the corrected PAGE 6 transcripts on the website. Those of you who are here to testify have already been advised that we intend to put the recording and transcripts on the website, and by your appearance here we understand that you have consented for us to do this. We are also putting written statements of testimony submitted on the Office's website until the transcripts are posted. The Comments, Reply Comments, Hearing Testimony and Post-Hearing Comments will form the basis of evidence for my recommendation to the Librarian of Congress. Before making that recommendation, I am to consult with the Assistant Secretary of Communications and Information of the Department of Commerce. We have already begun these consultations and expect to have more discussions with the agency that the Assistant Secretary heads - - which is NTIA, the National Telecommunications and Information Administration. After receiving my recommendation the Librarian will determine by October 28, which is the deadline, whether or not there are any classes of PAGE 7 works that shall be exempted from the prohibition against circumvention of access control measures during the three years that will begin on October 28th in the year 2000 forward. It is clear from the legislative history that this rulemaking proceeding is to focus on "distinct, verifiable and measurable impacts." Isolated or de minimis effects, speculation or conjecture, and mere inconvenience do not rise to the requisite level of proof. Any recommendations for exemptions must be based on specific impacts on particular classes of works. The panel will be asking some tough questions of the participants in an effort to define the issues. We stress that both sides will receive difficult questions, and none of the questions should be seen as expressing a particular view by the panel. This is an ongoing proceeding, and no decisions have been made yet. The purpose of these hearings is to further refine the issues and get the evidence that we need from both sides. In an effort to obtain all PAGE 8 relevant evidence, the Office reserves the right to ask questions in writing of any participant in these proceedings after the close of the hearings. Any such written questions asked and answers received will be posted on our website. What I'd like to do now is introduce our panel. To my immediate left is David Carson, who's the General Counsel of the Copyright Office. To my immediate right is Charlotte Douglass, who is Principal Legal Advisor to the General Counsel. To David's left is Rachel Goslins, who's an Attorney Advisor in our Office of Policy and International Affairs. And to Charlotte's is Rob Kasunic, who is a Senior Attorney in the Office of the General Counsel. We're about to begin. And we have been asked by our Reporter if any of the witnesses have written statements that they will be reading from, it would help them tremendously if you could give them a copy of your written statement. I see the panel is actually already in place, and I have received your order of preference. PAGE 9 So we will start with you, Dr. Siva Vaidhyanathan - - I can't say it. DR. VAIDHYANATHAN: Vaidhyanathan. MS. PETERS: Vaidhyanathan. DR. VAIDHYANATHAN: You were getting there. You would have been fine. MS. PETERS: And then we'll go to Karen Coyle, who will represent the California Digital Library. And then we'll go to the American Library Association, with Linda Crowe. And finally, we'll have Laura Gassaway, who will be representing the American Association of Universities, and the American Council on Education, and the National Association of State Universities and Land Grant Colleges. Okay, it's yours. DR. VAIDHYANATHAN: Good afternoon. My name is Siva Vaidhyanathan. I'm a media studies scholar and cultural historian at New York University. Thank you for allowing me to testify today. I am not a lawyer or a law professor. I am not a librarian. I am a user, a reader, a teacher, PAGE 10 a researcher and a citizen. Worse than that, I'm an unauthorized user. I am a fair user. I'm deeply concerned about the potential harm the anticircumvention power of the Digital Millennium Copyright Act will have on media studies and scholarship in general. I am just as concerned about the effects that this emerging leak-proof, highly regulated electronic regime could have on American culture and deliberative democracy. Today, most of the subjects of media studies research are widely accessible. A handful of works of film and early radio are even in the public domain. So scholars and teachers benefit from ample and easy sources. But that might change over the next few decades as more works -- even those already in the public domain -- become enclosed behind electronic locks and gates, and delivered in streams of digital signals. The potential for abuse of this technology and the legal power behind it is immense. You will notice that most of the tenses I am employing in this testimony are subjunctive and PAGE 11 conditional. As you may have gathered from all the previous testimony on this issue, this law has caused little harm yet, save the immeasurable and undocumentable chilling effect it might have had on those frightened by the combined cultural power of media companies and the state. Yes, my fears are speculative and alarmist. But they are not outlandish nor inconceivable. Not every media company is as harmless as a mouse. Not every government is invested in the free flow of ideas and information. Call me Cassandra if you must, but please imagine my classroom 35 years from now. As I do every semester, I plan to show my class a film that explores conflicting values and loyalties during wartime: Casablanca. But sometime during the 2020s, all the VEHICLES players at New York University fell into disrepair. The library has the tape, but nothing to play it on. Kim's Video Store on Bleecker Street is now just a Starbucks. Blockbuster is now a hand- held device instead of a large store. The only PAGE 12 means for showing this film to my class is to have it streamlined in via satellite feed into a video projector. Casablanca would have entered the public domain the previous year, assuming Congress does not extend the term once again. But it remains well protected, "double-wrapped" by both "click-wrap" contract and technological access controls. So my class settles down. On my palm computer I call up the interface page for either via-Disney-AOL-Warner-Mount or it's competitor MicroFox. I enter my "educator's code." I hit "play." Nothing happens. Once again, I must do my poor Bogart impression for the class in lieu of this film. So what happened? Well, perhaps this was my second class of the day and the service blocks fair users from watching a film twice. Perhaps the NYU Library could not negotiate a contract renewal with the company and stay within its tight budget. Perhaps my "educator's code" revealed me to be the one who wrote that scathing PAGE 13 review of the major summer blockbuster of 2034, Battlefield Earth IX: The Psychlo's Revenge. Perhaps the company identified me as someone who testified against the industry at a Copyright Office hearing way back in May of 2000. The Digital Millennium Copyright Act grants complete power to allow or deny access to a work with the producer or publisher of that work. The producer may prohibit access for those users who might have hostile intentions toward the work. This power could exclude critics and scholars. Most likely it would exclude parodists and satirists as well. The anticircumvention provision shifts the burden of negotiating fair use from the user, and the courts in the case of likely infringement, to the producer. The producer has no incentive to grant access to any user who might exploit the work for fair use -- including scholarship, teaching, commentary or parody. Under this regime, a user must agree to terms of contract with a monopolistic provider before gaining access. One must apply to PAGE 14 read, listen or watch. But why would a company restrict access to its product? In his testimony at these hearings in Washington, D.C., Bernard Sorkin, senior counsel for AOL-Time-Warner asserted that the content industries "cannot exist and prosper by barring their works from public availability," and any such fear "flies in the face of economic logic." Sorkin would be correct if his industry were perfectly competitive. But the very economic basis of copyright is that we need a state-granted limited monopoly to create artificial scarcity where natural scarcity could not exist. Once the content industry has a perfect, technological monopoly on high-demand back-catalog films such as "Casablanca," the industry has an incentive to limit the number of times it could be shown for free. Restricting free and "fair" use bolsters monopolistic pricing power. And companies have great incentive to restrict harsh critics and parodists from viewing their films. I am very concerned that the Librarian PAGE 15 of Congress is entrusted with composing a list of "classes of works" that might be exempted from the anticircumvention provision. As someone whose work spans from Twain to 2 Live Crew, and includes such sources as legal documents, private letters, diaries, movie soundtracks, and television and film, I have serious misgivings about a government agency allowing greater access to some works over others. All elements of expressive culture are fair game for scholarship -- at least they are today and for a little while. If any categories of works should be exempted from the provision, then all of them should. The Librarian of Congress should not have the power to favor one type or subject of scholarship over another. But as Arnold Lutzker testified at your hearings in Washington, D.C., "classes of works" are not "categories of works." Privileging one "category of work" might let you exempt literary or scientific work but not music or film. And I assume that the Librarian of Congress recognizes this distinction and plans to execute his power based on PAGE 16 it. Any proposal that libraries and librarians enjoy some sort of special exemption from the legal threat inherent in the DMCA would not satisfy my concerns. First, libraries are not users per se, and much scholarship occurs outside of libraries. Second, such a move would turn librarians into "copyright cops," who would be entrusted to determine which uses would be fair and which would not. Fair use is something I as a user must be willing to employ without having to apply for it. All fair use is unauthorized. If a content company has a problem with my use, bring it on, let's go to court. But let's not involve a third party in the dispute, either by requiring her to preempt my use or by threatening her with liability for any infringing use I might make. Copyright was invented in the British Isles as an instrument of censorship, a way of regulating the traffic of ideas through the selective granting of licenses. Fortunately, PAGE 17 copyright has grown in the American context as something very different. Up until a few years ago, when it still embodied a balance among creators, publishers and users, copyright served as an essential foundation of democratic culture. Its very imperfections helped American culture and commerce thrive in the past 200 years. American users have benefitted from the proliferation of American cultural products, but they have also enjoyed four important safety-valves against the censorious power of copyright: the first sale doctrine; fair use; allowances for private non-commercial copying; and the idea/expression dichotomy which allows facts and ideas to flow freely while protecting specific displays of those ideas. Now, all four of these notions are under attack by the content industries through the World Intellectual Property Organization treaties. The DMCA is only the first step of this process. If the film and music industries continue to tighten their reins on use and access, PAGE 18 they will strangle the public domain and the information commons. This trend presents a much greater threat to American culture than just a chilling effect on scholarship. Shrinking the information and cultural commons starves the public sphere of elements of discourse, the raw material for decision making, creativity and humor. So what should we do about this pernicious trend? How can we revive the beauty and genius of the American copyright system and maintain its positive externalities on our culture and democracy? Well, for a start: the Librarian of Congress should exercise his power to exempt from the anticircumvention prohibition any works that are not easily and widely available for teaching, research and unauthorized reading in an unsecured format. Unsecured formats might include VHS videotapes, printed paper volumes or standard compact discs. That means these products must be archived in a public or university library somewhere. PAGE 19 Second, the Librarian of Congress should ensure that the anticircumvention prohibition does not apply in any case to material not covered by Title 107, the Copyright Act. Therefore, a publisher could not stifle access to works in the public domain, to government documents, or facts, or ideas or data. Third, the Librarian should exempt any works that enjoy technological controls that deny access based on editorial concerns. There are no bad readers, authorized or not. But ultimately, the Librarian's actions -- even if he provides as broad an exemption as possible -- will do little or nothing to restore the sense of public interest to copyright law. It would only be an endorsement of that value. Congress has granted the Librarian the power to exempt the use of certain classes of works from prosecution, but not to exempt the sale and distribution of the very anticircumvention technologies and devices that we users would require to exercise our rights in such an environment. PAGE 20 That's like granting us the right to record television shows for later viewing, but prohibiting the sale of video recorders. It's like having freedom of the press, but not the freedom to own a press. Congress should revisit this issue. I trust Congress would recognize the value of an imperfectly regulated yet balanced copyright system. The Digital Millennium Copyright Act is an absurd, Orwellian law, and it should be abandoned. If Congress does not fix it, I hope the U.S. Supreme Court -- which several times in the 1990s stood up for users' rights -- would once again rescue our copyright system from those who would corrupt it. On one final note, I offer an anecdote that should illustrate the value of unauthorized use. In December of 1906, Mark Twain donned his white suit to testify before a congressional committee on the new copyright bill. Twain expressed his desire for copyright to be expanded from mere expressions to ideas as well, and to be extended in perpetuity. PAGE 21 While Twain described the very copyright regime we seem to have built in his absence, a young actor in New York was busy reading a short story by Twain called, "The Death Disk," a fable set in the time of Cromwell's rule of England. The young actor made unauthorized use of Twain's story -- which Twain himself had lifted from Thomas Carlyle -- to make a short silent film in 1909 for the American Mutoscope and Biograph Company. In his short films, this enterprising young man worked out the technical challenges of narrative filmmaking. That man's name was David Wark Griffin, the father of American film. Thank you. MS. PETERS: Thank you. Next Karen. MS. COYLE: Good afternoon. My name is Karen Coyle, with the California Digital Library at the University of California. And I'm here to speak to you as a practitioner of library technology, not in any way as an expert in law. And I must say that what I say here are my own words. This is not policy of the University of California. I should give you a little bit of an PAGE 22 idea of my expertise and what I have been working on. The California Digital Library serves all nine University of California campuses. We have an online union catalog of about 10 million titles, and about 18 million holdings. We have available online 66 abstracting and indexing databases for our users, and we have eight of those we've actually mounted on our own computers. We provide our users with access to over 4,500 electronic journals and other digital works. My own expertise is primarily in the development of databases, and I estimated the other day that I have probably overseen the development of databases and the loading of about 50 million bibliographed records. Because there isn't a great deal of time, I chose three of the questions that were in your original call for comments here. And I will just answer those. All three of them have to do with technology. The first one is No. 2, "Do Different Technological Measures Have Different Effects on the PAGE 23 Ability of Users to Make Non-Infringing Uses?" And I had a very interesting experience just this week. You may know that Xerox and Microsoft recently announced that they were going to become content providers. And along with this, they have their own access control standard called XrML. And since this is part of my job, I tend to follow these standards, so I went out to the site to download it. In order to read it, I had to also download a special version of Adobe Acrobat, and I had to give my e-mail address so that I could be sent the key so that I could open up the document. I did this. Opened the document, it was 117 pages. I closed it and decided I'd look at it another day. It so happens that in my office I have two computers, and they're connected together and I store everything on basically a shared volume. And as far as I'm concerned, they're just two windows into my work space. So, earlier this week I got a chance to open up that document again, and I went to one of my computers and went to open it up. And I got an PAGE 24 error message. I realized it was the other computer and I hadn't downloaded the right version of Adobe Acrobat. I downloaded the right version of Adobe Acrobat, I'm still getting a rather cryptic error message. And it took me a few tries, but after a while I basically deduced that this document can only be opened on the computer where it was downloaded. Well, I decided to go back and read the legal "I agree" agreement, which of course I hadn't read the first time. None of us ever do. And there was no mention in there of access controls at all. So I went back to the web page where I downloaded it, and there was no mention of access controls. On Tuesday I went to the XrML site and said, "It looks to me like this is limited to just one CPU. Is this the case?" And as of this morning I still haven't gotten an answer from the developers. There are two sort of interesting technological aspects to this. One is that access controls may be invisible to the user. And these PAGE 25 were definitely invisible to me. I knew that the access control had to do with Adobe Acrobat, that I had to have a key. But there was nothing telling me that this was only readable on a single computer. The other interesting aspect is that the access control and the license may not be the same. Now, I can't find, really anywhere, a license that says what my license is in relation to this document. The license really has to do with relation to what I would develop using the XrML standard. But interestingly enough, the document seems to be licensed to me. They asked for my name, my address -- that had to be filled in -- my e-mail address. And yet the access is limited to an inanimate object on my desk, which is of a very temporary nature. Because as we know, computers get upgraded every three to five years. This one's four years old, it won't be around very long. I assume that when I upgrade my computer equipment, I'm not going to be able to read this file. Now, when you make this question about PAGE 26 "Do Different Technological Measures Have Different Effects?" the answer is yes. But when I sat down to try to think of all the different technological measures and all the possible effects, I realized that this is going to take a really serious study. I don't think we really know what all the effects are, and some of them are hidden, some of them aren't obvious. I really think that what we need for this technology is something that I like to call a social impact study. That when new technology comes up, that someone needs to look at it in terms of not just what does the technology do, but what's the impact it's going to have on society? And I turn to you, because at the moment I don't know of any other agency that might be in a position to bring together a group of technologists, or somehow charter an investigation of this nature. Of really learning what the controls are and what the impact they have on access. Because I don't think that we have an answer for that today. The next question that I wanted to look at was the one that says if there are works that are PAGE 27 available, both in basically digital copy and hard copy, is the availability of the hard copy essentially make it so that access to the digital copy isn't as important. And here I speak from experience that we've had in developing computer systems over about 20 years. Because I started in 1980 with the University of California on these systems. I mentioned that we have sort of eight core databases that are on our system. One of them is National Library of Medicine's Medline, which we made available 12 or more years ago, obviously to serve our medical and biology research staff and students. There is a paper equivalent Index Medicus. When we made it available we were very surprised by the amount of use, and we continue to be surprised by the amount of use. This database accounts for about 30 percent of the use on our system amongst these core databases. This is quite a surprise. There have been times when it actually rivaled the use of the online catalog. I can't explain it, no one else I know PAGE 28 can explain it. But what we do know is that by making information available digitally, we aren't giving the same access that we gave in the paper copy. And that our users are finding new ways to make use of the information, and are discovering new information. And I think we all know that when you search in a database you have the ability to discover information that you might not have discovered in the hard copy work. Because the ability to search is so much better. The exciting thing about working in digital libraries is that we're really developing a new kind of scholarship, and it is different to the scholarship that took place in the paper world. And I don't think we'd want to go backwards to that paper scholarship, and pretty soon we really won't be able to. And then the last question that I wanted to address--down here, it's No. 18: "In What Ways Can Technological Measures That Effectively Control Access To Copyright Works Be Circumvented, and How PAGE 29 Widespread Is Such Circumvention?" Well, I do not know of any library that has a job title, Librarian/Cryptographer. We really -- we don't have cryptographers on our staffs. I don't expect us to have them in the near future. I was thinking about the other day that -- I believe it was last year or the year before, the Electronic Frontier Foundation did a crypto- experiment in which they spent about a year -- actually, a little over a year -- building a special computer to the tune of about $250,000 so that they could experiment with breaking through 56-bit DES encryption. And apparently -- John Gilmore just told me that it actually took 56 hours. And then once they had that computer built to do it, and it's just coincidence that it's the same number as the number of bits. Clearly, it is not really economical in most cases for a library to use this type of technology in order to gain access to works. Encryption really is a question of economics, and PAGE 30 the cryptographers will always tell you there is no unbreakable encryption. There's just encryption that it's too expensive to break for what you're going to get out of it. I can't imagine libraries having the ability to break through strong copy controls. And I think that this is, in a way, unfortunate because I am quite convinced that we will lose some works. If a library does find that it needs to invest its time and resources in trying to free a work in order to make it available to the public, I feel they are doing a great public service and we should support them in that. It is not something that I can see that any library is going to undertake idly. I don't have a recommendation for you as to what the wording should be, in terms of what exemptions there should be for libraries, because I couldn't begin to speak that language. And I will let the lawyers do that for me. But thank you very much. MS. CROWE: Good afternoon. I guess I'm going to start with the disclaimer, since I'm far from a copyright expert. This testimony is PAGE 31 short and, I hope, to the point. My name is Linda Crowe, and I'm the Director of the Bay Area Library and Information System, the Peninsula Library System and the Silicon Valley Library System. Each is a consortium of public libraries covering the core of the Bay Area, including all the public libraries in the counties in alphabetical order: Alameda, Contra Costa, San Francisco, San Mateo and Santa Clara. The service area consists of 25 individual jurisdictions, and over 175 outlets or main libraries and branches. All of these outlets have public access to electronic resources. Some with only a single terminal and some with more than 100 terminals open for public use. And they are in use from the time the library opens until it closes. Public libraries see themselves playing a critical role in providing accurate access to information. And it's particularly important that new technologies support and enhance, not impede, the ability of public libraries to provide these services. Many consider the public PAGE 32 library as the public transportation to the Internet or the superhighway. And I suppose that is somewhat descriptive of their role. Much has been made of the digital divide in this state and throughout the country. The area that the libraries we serve represent, one, if not the most wired area in the country. Yet, there are information haves and have- nots, and the digital divide is as real here as anywhere else. Where else can many teenagers who live in East Palo Alto, just down the road, parts of West Oakland or Bayview/Hunter's Point go to access the resources he or she needs to complete a homework assignment, or do research on a subject of personal interest? As three consortias, we are spending more and more scarce dollars on resources in electronic formats. For example, this fiscal year we will spend close to $1.5 million on electronic databases. Next year we will probably spend more, and we are constantly trying to meet requests for people who want more and need more. PAGE 33 We need to be able to assure our users that within the limits of fair use, that the people who need them will have them available. I mentioned the digital divide, and that public libraries may be the only place some people may be able to use these resources. We also find that more and more people who have access to the Internet elsewhere, come to the public libraries because librarians have organized the information and can help access what the user really needs more quickly and more effectively. These users need research done in whatever format is available. And public libraries need to be able to supply these formats without undue technological constraints, costs or charges. At this point, most public libraries are not talking at meetings much about the DMCA, copyright and fair use because they have lived and accepted the principles that they've had for years. Now, we have this broad new law that confuses and concerns us, because of the ambiguity PAGE 34 and apparent contradiction. On one hand we have the anticircumvention section 1201, and on the other hand -- as I understand it -- we have the provision to 21201 that says, "nothing in this section shall affect rights, remedies, limitations or defenses to copyright infringement, including fair use under this title." We need a precise, a clear precise sense of what is and is not proper, so we can exercise those rights. Without this preciseness we are likely to err on the side of caution, possibly restricting access to information to those who need it, and denying them the rights to use it in ways that are legal under current copyright law. I would urge the Librarian to issue exemptions that protect the rights of content owners, but allow us to serve our public. That is, the people who use and depend on public libraries. Thank you. MS. GASSAWAY: Good afternoon. My name is Laura Gassaway. I'm here today on behalf of the Association of American Universities, the National PAGE 35 Association of State Universities and Land-Grant Colleges, and the American Council on Education. Thank you so much for the opportunity to appear before you. I will make four points today. First, an examination of the purpose of today's hearings. Second, our experience to date with access controls and their first cousins, license restrictions. Third, how the proposed business models presented by copyright holders will interfere with the use of copyrighted works in teaching, learning, scholarship and research. And lastly, what this means for your task in the rulemaking proceeding. The bottom line for us is exempting from the realm of prohibition on circumventing conduct any uses for which the user had lawful initial access. Further, we believe two types of works that were identified in our opening statement -- fair use works and thin copyrights -- are particularly vulnerable. And equities are stronger toward exempting them. Congress was concerned that the PAGE 36 possibility of technological protection measure, TPMs, would be applied by copyright owners in ways that interfered with lawful uses of copyrighted works. Ways that upset the copyright balance that has long served owners and users of protected works. You've heard copyright holders in these hearings state that their future economic health will be completely compromised if there is an exemption for their works. Nothing could be further from the truth. The risk to copyright holders is negligible. Nothing in this rulemaking affects the availability of the prohibitions contained in Sections 1201(a)(2) and 1201(b). For example, the manufacture and distribution of circumventing devices and the performance of circumventing services. Nor will any copyright remedies for infringement be exempted. In any case of infringement Section 1201(a)(1) is redundant. Moreover, nothing in this rulemaking will stop copyright holders from applying TPMs. The question is whether we invoke federal courts to PAGE 37 reinforce anticircumvention. Additionally, contractual rights will continue to apply. Our experience with access controls to date, whether technological protection measures or licensing, have been varied. We, of course, have had much more experience with licensing than TPMs to date. Most of our experience has been with passwords, which are the kind of basic or primary access control technology that does not cause us concern. In fact, our own Institutions are using passwords on web pages, course materials, and the like that we develop. We have also dealt with location restrictions, especially in license agreements where the university pays a license fee but access is restricted to on-campus use. This has been a problem for us when we're dealing with distance learning students, medical interns and the other students who are enrolled but who are physically located elsewhere than the campus for that particular semester. And we've had to deal with authentication of those users to create access for PAGE 38 them. Sometimes publishers have not allowed us to do that. They have restricted it to the domain name, regardless of the fact that these were enrolled students. So they have restricted it to the place, regardless of the fact that these were enrolled students. We've also seen restrictions on who may access the material. Usually it is students, faculty and staff -- pointed out to you by earlier witnesses in these hearings. The problems for state-supported Institutions who also have responsibility to serve citizens of their area and their state. Increasingly, this is causing difficulty when we have sole source government information that's embodied in these electronic databases, et cetera. Another problem that we've had with licensing has been the removal from databases of materials during the license period, with no advance warning. And I'm specifically referring to the removal of the French legal materials from the Lexis PAGE 39 database very recently. Now, one could certainly argue that Lexis has violated the license agreement with law schools in doing that, but that's really not our purpose here today. Just to point out that these are the difficulties we've had. There have also been licensing restrictions on use, where a particular product could be used for teaching and demonstration but not for research. Sometimes licenses to use have been denied because the copyright holder sees their market as a non-educational model. It just doesn't fit the use that we asking, so they have refused to license entirely, or even to respond to our request for a license. Today's contractual restrictions are tomorrow's technological protection measures. With a license agreement, however, institutions have had some ability to negotiate the terms. With TPMs, the ability to negotiate is lost. We have had one relevant experience in my own library with TPMs and the problems with them. PAGE 40 I call these "disappearing CD-Roms." Actually, we still have the CD-ROM, it's the content that's disappeared. Apparently they were date- sensitive, although this was not included in the license agreement, and there was no advance warning. The library was left with nothing. This happened to us with Westlaw CD-ROMs. The publisher admitted that it was a mistake and agreed to replace them. But we were several weeks without the material. So far, publishers have not implemented many such controls. But according to their testimony during these hearings, this is about to change. The institutions represented by our organizations are seriously concerned about what we've heard from copyright holders at these hearings. Clearly, they intend to merge access and use controls. The business models they discuss make it clear. At what point do access controls and use controls merge? One could argue that when a university acquires access to materials through a license agreement for its enrolled students, and yet PAGE 41 off-campus but enrolled students are not permitted access, this is a type of use control. We paid for access for these students, but they still cannot use the material. The copyright holder is differentiating between users, all of whom are enrolled students. Is this access or is this use? We don't know, and we can't tell when we are liable for such conduct should we circumvent. Copyright holders clearly want to merge access and use, as their testimony indicates. They say they want to keep anyone from breaking into the bookstore and stealing a book. What it seems to me, and to us that they are saying is that they want to stop anyone from breaking into the book, even after they have lawfully acquired access to the book. The pay-for-use world that publishers and producers have discussed at these hearings are use controls for higher education. Such merger is completely inconsistent with the congressional scheme. Congress treated access and use controls very differently in the statute, and PAGE 42 the Copyright Office should take into account what copyright holders have said during these hearings because the risk to users of copyrighted works is considerable. The proposed business models presented, we believe, will interfere with the use of copyrighted works for teaching, learning, scholarship and research. Fair use is fundamental to educating America's students, producing scholarship, research and the like. College and university libraries acquire copyrighted works by purchase, gift or license, faculty, students and staff then have the right to use these works for education. How will an educational institution be able to function with pay-for-use? For example, the single listen. A faculty member plays the song in class once. The students then ask to have it repeated because they didn't quite understand or get enough of the material for the educational purposes. The individual students then need to listen additional PAGE 43 times for reinforcement, or to study for exams. How is this going to work for higher education? We also strongly support the statements of library associations concerning the preservation of digital works. The cultural and scholarly record is critical for researchers and teachers, and indeed, we believe for society. The makers of silent films saw them only as works for entertainment. Fortunately, there were libraries that preserved these works as important cultural records. How much poorer would be our understanding of society, and of early movie-making, had these works been lost to the world. The same is true for things like greeting cards, postcards, old photographs, advertising posters, things that clearly were not originally intended for education. Fifty years from now scholars will want to look at the early digital materials to determine how the industry developed, and what it said about people's taste and interest. What scientific information was deemed worthy of distribution in PAGE 44 digital format. One witness dismissed archiving of digital works as irrelevant to society. The Copyright Office and the Library of Congress needs to think hard about turning these decisions on preservation over to owners with this attitude. With this information, how should the Copyright Office exercise this rulemaking power? Congress intended primary controls on access when it established the distinction between access and use controls. For that reason, we believe the classes exempted from the act of anticircumvention should be those for which the user had lawful initial access. In the Academy, we do not differentiate between works for entertainment and works for teaching, learning and scholarship. The discipline in which they are used makes a great deal of difference. My colleague from NYU pointed out a good bit of that. We have faculty who study the history of rock and roll, so how can sound recordings -- works originally intended for entertainment -- not be PAGE 45 subject to defeating TPMs in cultural studies, history, et cetera? The works were intended for entertainment originally, the use that it's made for them in education is quite different. It is instruction and research. Use of these works in the Academy has been fair use for 200 years. There are two classes of works that probably have greater universal use in higher education than others. In other words, all disciplines make use of these works. Therefore, these works are those for which the balance leans most heavily for a broad exemption. And the failure to do so will significantly hurt teaching, learning, scholarship, et cetera. First of all, fair use works. Works that due to their nature are likely to be lawfully used under the fair use doctrine. This would include, at a minimum, scientific and social databases, textbooks, scholarly journals, academic monographs and treatises, law reports and educational audio/visual works. PAGE 46 You should probably also think about writing a regulation that would exempt a work not on this list, but because of its use in a particular context, where it was highly likely the use might be a fair use. And an example might be motion pictures in a film school. The second type of work that we believe is especially -- needs a broad exemption are factual works, those with thin copyrights. Those that contain limited copyrightable subject matter, and are fact-intensive, or that contain significant public domain materials. Examples of these works would be maps, some databases, histories, statistical reports and abstracts, encyclopedias, dictionaries, newspapers and the like. We believe that the exemption you are considering should be broad for scholarship, education and libraries. The United States has the finest academic system in the world. Likewise, we have the strongest copyright industries. Both have thrived under copyright and with fair use. Responsibility PAGE 47 falls on your office and this rulemaking to see that the balance is preserved. Thank you. MS. PETERS: Okay, thank you. We want to go to the panel questioning. And the questioner can ask a person specifically, or can throw it out to the panel. If they throw it to the panel, anyone who feels like jumping in and contributing, please do. And even if a question is directed to an individual, if another person feels that they have something to add, please feel free to add and say "I want to add to that." We're going to start the questioning with Rachel Goslins of the Policy and International Staff. MS. GOSLINS: Good afternoon. This panel is especially valuable and helpful to us because of all the types of people that we have technologies before us. Librarians, academicians and users have the most kind of hands-on experience of both the works and the technologies that we're talking about. PAGE 48 So my questions are going to be largely focused on actual problems or conditions that you experience in the course of administering your libraries or teaching your classes. First, in a way we're relatively lucky in this study, because we're asked to look at access control technologies, which are probably the oldest technologies we have around in the relatively young world of the digital environment. Password protection and encryption, IP domain names, protections have been around for a while, and so there should be some backlog of experience with them. And my first question is directed to the whole panel. Considering that these testifying have been around for a while, and considering that up until -- and at the moment the act of circumventing them is not illegal, so I'm not asking anybody to confess to anything. Are there times or situations in your day-to-day businesses where you have to circumvent these kind of protections or forego the use of work? PAGE 49 DR. VAIDHYANATHAN: There are examples where I would like to circumvent, but it doesn't mean I either have or have been able to. For instance, I use Lexis/Nexis rampantly in my research. And I'm licensed through New York University Library to read Lexis/Nexis database information from my IP address on my university- issued computer. But when I travel, I can't. I wish I could. I wish I could have access to that, but once again the license is sort of computer-specific, or machine-specific as opposed to licensing the access to me as a scholar. MS. GOSLINS: And what is it, exactly, that prohibits you from circumventing these controls? DR. VAIDHYANATHAN: Well, because the IP address allows the server to let me into that particular page. It's what checks whether I'm okay. MS. GOSLINS: So in this case, it's the technology? It's not the lack or existence of a legal prohibition in doing so, it's just that the technology is effective? PAGE 50 DR. VAIDHYANATHAN: Right. MS. GOSLINS: Are there other experiences that people have had? MS. COYLE: I think that describes most of our experience, which is that in things like IP address checking we have vendors who limit -- most our vendors, actually, limit to certain IP addresses. And it's not that our users or our librarians wouldn't sometimes like to get around that. But most people don't know how. So the technology is actually effective. And I think that's why we don't have a lot of experience in trying to circumvent controls. Some of them virtually cannot be circumvented. I mean, it can be very difficult. The economy of circumventing these controls is really prohibitive. MS. GOSLINS: That's actually a point I wanted to pick up on. MS. PETERS: May I comment on this? Which is, you said that the control came about with regard to your contract, your license. So, actually, I guess your university signed a license PAGE 51 that binds its employees. So is your complaint against the license, or is it the license plus the technology? DR. VAIDHYANATHAN: It's the license plus the technology. I mean, one requires the other. MS. GOSLINS: I'd like to pick up on Ms. Coyle's point. But before doing that, I'd just like to make sure nobody else on the panel has examples of times that they have to circumvent. MS. GASSAWAY: There have been times that we've had to circumvent, specifically with license to -- MS. GOSLINS: Arrest her. (Laughter.) MS. GASSAWAY: Specifically with license to Westlaw, because we're under the law school contract for each individual's personal password. You know, if someone comes in off the street with a reference question, but it is -- we've already looked at it, we know whether or not -- it's not for law practice. It's for a general question. PAGE 52 If we then use our own passwords for that, in a way we are circumventing. Not letting them do it, but to answer a reference question for them. And we have done that on occasion, simply because that's the only access that we have to some of the material. Now we can't answer that French legal question that they were going to ask, period. MS. GOSLINS: If I could just follow-up on that for a second. Is it really, then, that you're -- when, in effect, you're doing there would be circumventing the license terms, right? Not necessarily the -- I mean, the access -- you're not actually breaking the access control protections. You're just circumventing the terms of the license? MS. GASSAWAY: Yes. I mean, I guess it depends on which way you look at it. Because of the personal password situation, it's a little bit different. Because ours apply to any machine, no IP address. It's personal. MS. PETERS: For certain use, right? MS. GASSAWAY: Right. For educational use. PAGE 53 MS. PETERS: Right, yes. MS. GASSAWAY: Not only educational use, law school use. MS. GOSLINS: And are there other -- okay, picking up on Ms. Coyle's point, which is something we've heard in several of these hearings, that circumventing access controls is expensive and time-consuming and difficult. And generally you need some design or some product or service that is illegal to manufacture, design or produce them. At the risk of being argumentative, why is this exemption important, then? If these technological protections are effective, if realistically libraries wouldn't be able to circumvent even if we were to exempt all classes of works, what will a possible exemption give you? MS. COYLE: Right. And the reason why I see that as being an argument for the exemption is the fact that should a library get in the position where it does have to dedicate the $250,000 and a year's worth of development in order to circumvent, it's because it was an extremely important piece of PAGE 54 knowledge that that library felt it was worth investing the money to preserve. It's not that you can't do it, it's that it has to be -- you have to match the value of what you are preserving with what it's going to cost you to get to it to preserve it. DR. VAIDHYANATHAN: And I would like the Library of Congress to stand up for the principle. MS. PETERS: For the -- DR. VAIDHYANATHAN: The principle of preservation, the principle of first sale, the principle of fair use. MS. GOSLINS: Okay. This is another sort of broad question. Obviously, one of the things we're looking at is to what extent materials are available in alternative formats, and to what extent they're available only in digital formats. So I'm just curious. I don't mean specific numbers, but off the cuff, how much of the material that you deal with in the operations of your libraries is available only in digital form, and how much of it is available elsewhere? PAGE 55 MS. CROWE: In the public libraries, often the databases that we buy are databases of periodicals and that sort of thing that are -- they're also available in print. But we have to make choices about which format we're going to buy, because we can't afford to buy both formats. So, it has become more and more popular to buy them in electronic format because people can access them from home, or from the office, or from wherever is convenient. Therefore, although they may be available in two formats, we have to choose. And we're choosing the one that we think is more convenient for people. MS. GOSLINS: But, in general, the formats -- you're choosing between two formats, and in the day-to-day world it's also available in an analog format? MS. CROWE: Yes. Often. Not all the time, but often. MS. COYLE: Now, many of the e-journals that we carry never were published in print, so we do have many thousands of electronic journals that PAGE 56 are available only in this format. And, as you know, there's things like Highwire Press that in the academic world, in academic production of knowledge, that there is a consciousness of the need to both provide these things electronically, and to preserve them. And what we're finding is that consortia and library agreements are being developed so that these things are being maintained for perpetual access by some institution. Research Libraries Group has taken on that for some materials. This is definitely true for the archival materials, because as people digitize archival materials, even though the archival material is still there, it has a very low possibility of use. So when you have an archive of very rare photographs, you cannot let people have access to them. Therefore, the digital product really becomes a functioning surrogate. And there is a great consciousness of providing perpetual use to those materials. However, oftentimes those materials are materials that are owned by the library, and PAGE 57 therefore there isn't a copyright issue. MS. GASSAWAY: And we certainly have some things in the big legal databases that are available in print. But we're also getting many more databases that do not exist in print, like EDGAR, with all the SEC material. Some of that does not exist in print. If you're in Washington and can go by the Securities and Exchange Commission, you can get hold of some of it. But some of it is just simply not available to us. The ERIC database is another one in which there's a good bit of material that's available only electronically. MS. GOSLINS: What is the name? ERIC? MS. GASSAWAY: The education. And until Marybeth's colleague at the Patent and Trademark Office got right about publishing patents, the only access to them was through an electronic database. But now they are available on the web. The Bureau of National Affairs is creating all kinds of libraries, as they're calling them, of their materials. Like in Healthlaw, which PAGE 58 will have a combination of materials that are available in print, but a big chunk of their stuff is not. And it is the bringing it together that is making it valuable. So I think we are on the upswing of seeing things that are available only electronically. MS. GOSLINS: Okay. Can I ask you a very specific question? Do you remember what witness it was that said that archiving is irrelevant to society? MS. GASSAWAY: Yes. Richard Weisgrau of ASMP. MS. GOSLINS: All right. And I have another specific question for you, Lolly. One of your suggestions is that what you call fair use works should be exempted from the anticircumvention prohibition for, I believe, libraries and universities, right? And in that group you include textbooks and audio/visual and other educational materials. Does this create a similar problem that PAGE 59 we saw in distance education, which is a proposal to exempt works precisely in the markets for which they're designed. MS. GASSAWAY: Well, remember that that applies only to works for which we already lawful access. So if these are licensed work, we paid the license fee. It's just they exempt them from the anticircumvention provision. MS. GOSLINS: And the initial access point would be -- for instance, if you -- as an example of that, if you purchased access to an online database for six months, and then that six months ended. You would then, from that point on, be exempted from the prohibition of anticircumvention because you had initial use? MS. GASSAWAY: No. I think when you're talking about a termed license period, it is only for that period. I think it's only for that period. I do not personally see how we could ever advocate that -- you know, with a journal subscription now, an analog subscription, when you stop the subscription you don't continue getting the volumes. PAGE 60 So, I mean, I think that the move to the digital world is the same thing with those. If you paid for 12 months of access, that's what you get. MS. GOSLINS: So how would it work, that if you had initial lawful access and it's an audio/visual work produced for education purposes. You would then need to circumvent the access control protections? MS. GASSAWAY: It's not that whether we would need to, it's whether we would be liable if we did do it. That's really the issue. We don't know whether we would need to. Suppose that it's one of these timed ones that disappears? You know, something like that. And yet it's still within our contract period of having paid for. Why should we be liable for circumventing if we've already had lawful initial access and the contract period has not expired? MS. GOSLINS: I'm sorry that I'm following up on this. I think I'm just still slightly confused. I understand that example. That seems like that example would, however, apply only PAGE 61 in situations where somebody made a mistake and you weren't having access to something that you were entitled to. That seems like a pretty narrow exemption. MS. GASSAWAY: Well, all I can tell you is from my own experience as a law librarian, and that's the only one I personally dealt with so far. So I don't know how rare it is, because there were a bunch of us who had to deal with that. MS. GOSLINS: Okay. And then I just have one more question for Mr. Vaidhyanathan, something like that. One of the things you mentioned, or you suggested as a criterion for exemption is technologies that didn't deny access based on editorial concerns. I'm just curious. Are there technologies now that do that, and are capable of doing that? DR. VAIDHYANATHAN: So far -- I'm sure they're all capable of doing that. When I get into NYTimes.com, I have to enter my e-mail address and a password. So far I have done nothing to justify NYTimes.com from keeping me out, but there's nothing PAGE 62 to prevent them from keeping me out. Now, it's a -- MS. GOSLINS: Okay, it's sort of an FBI Top Ten Wanted list? DR. VAIDHYANATHAN: Or, yes. Whatever. It's not inconceivable to think that certain sites or databases would be willing and able to exercise editorial control over access. You know, it would be a very simple way of regulating readership. MS. GOSLINS: But you're not aware of anybody that does that now? DR. VAIDHYANATHAN: No, I'm not. MS. GOSLINS: Okay. All right. I think I'm done. MS. PETERS: Charlotte? MS. DOUGLASS: You said in pursuing a little bit of that adverse effects, and the difference between an adverse effect and a mere inconvenience -- well, if you can't get material at all but for circumvention, that's one thing. But suppose you have the ability instead of circumventing of digitally-encrypted work, to go to PAGE 63 12 other sources and get access. Would you think then that the availability of the work in 12 other areas would mean that that was an adverse effect, when you could not get the digitally-encrypted version? Or would the availability that took you 12 times as long mean that that was a mere inconvenience? I'm just trying to hone in on what's an adverse effect and what's a mere inconvenience. MS. COYLE: Yes, and I'm trying to understand it. I'm not quite sure what it is. But it sounds to me like what you're talking about is something that comes up in the area of preservation of hard copy works, which is that if your work is deteriorating you don't immediately copy it. The idea is that you're supposed to go out and try to find out if you can reasonably get another copy. And you seem to be saying that if for some reason your digital access is broken, should you be required to go out and try to find other digital access before circumventing. Is that -- MS. DOUGLASS: No, I'm sorry I wasn't PAGE 64 clear. I'm trying to distinguish between, on the one hand, what is a real adverse effect and what is a mere inconvenience. For the reason that Congress tells us that we are to consider things that are adverse effects in trying to establish these particular classes of works that are to be exempted for non-infringing use. But Congress also tells us that we should not pay as much attention to mere inconveniences. And so I'm trying to decide whether -- what are they talking about? MS. COYLE: You're still trying to define inconvenience, yes. MS. DOUGLASS: What is an example of that? And so I came up with an example that, suppose you had a digitally-encrypted work but you can actually get the same material by going to 12 other places, taking 12 times a long. Does that mean that it's -- that the fact that you can go and take 12 times as long, that's a mere inconvenience or is it an adverse effect? MS. COYLE: I think it's very hard to PAGE 65 sort of answer that question in the abstract, because it depends on whether or not that user is actually going to continue to follow-up 12 times as long. Or if that 12 times as long is 12 days rather than 12 seconds, which there's a big difference in time. And so there has to be some kind of concept of what it's reasonable to expect users to go through, or for libraries to go through. DR. VAIDHYANATHAN: And Ms. Douglass, I have actually three observations about that subject. First of all, if there were a condition, for instance, that the researcher were on a heavy deadline, then -- and an article is not going to get published if he or she can't get that information within 28 or 48 hours, that's a real effect. Secondly, another real effect might be if the user is for instance, visually disabled and has software available to create an audio presentation of a digital work and is basically unable to read or understand printed text, then that's a case of real harm. PAGE 66 Third, I think it's very important to recognize that in the realm of First Amendment law a chilling effect is a real effect. And it wouldn't be very hard to be able to come up with examples of a chilling effect. And then, for instance, just off the top of my head before I read this, when the anti-DSS cases hit the press, sites that had DCSS software on them shut themselves down before it was ever litigated. In other words, they were protecting themselves. That is a real chilling effect. MS. DOUGLASS: Thank you. MS. GASSAWAY: I think also that there are works that are available only electronically. So, clearly the adverse impact or adverse effect is there for those works. Your example of there are 12 other sources available says that going through the decryption or whatever, would not be very time or cost-effective, except in the instances that Siva just mentioned to you, or something like that. But we do have to focus on the fact that increasingly there are works that are going to be PAGE 67 available only digitally. And on those the adverse effect is clear. MS. DOUGLASS: One of the questions that we asked in our Initial Notice was, speaking still of adverse effect, with respect to any adverse effect is there an explanation for the adverse effect other than the presence of technological measures that effectively control access to works. That means that but for the presence of circumvention, is there -- was that adverse effect caused, or could it be something else? Could it be because of the licensing restrictions, could it be for -- I know this is highly abstract and I'm sorry I can't give you a hard example of it. But we have to somehow try to link cause and effect here because that's what Congress said to do. So I want to know if there is, or if there could be other reasons for the adverse effect except for the prohibition on circumvention? Does that make sense? MS. GASSAWAY: I think it does make sense. I think, at least from my standpoint, the PAGE 68 concern that we're going to see some of the license restrictions on use converted to TPMs is really where we focus. And where we really have to think about, sure, some of the concerns today are just pure licensing concerns. But if you can turn that contract into a technological protection that also protects not only access, but use, you know, then it's too late. We've already lost the access to the work unless that we have the broad exemption from circumvention. MS. DOUGLASS: And your broad exemption goes to fair use works and thin copyright works? MS. GASSAWAY: Well, first we would prefer that it be any work for which we have initial -- MS. DOUGLASS: Lawful. MS. GASSAWAY: Lawful use, right. And then in the alternative, if that is broader than rulemaking can encompass, we'd say that there are these classes of works that are specifically unique to higher ed. But we would also hope that there was a rule that said for particular circumstances there PAGE 69 would be the ability to bring other categories into that. And I use as the example films in a filmmaking school, or something like that, which are not in those two categories necessarily. MS. DOUGLASS: Now, I'm trying to recollect. You said that classes of works is not necessarily categories of works. MS. GASSAWAY: I didn't say that. MS. DOUGLASS: Oh, you said that. Okay, I'm sorry. So this may be an unfair question, but can you figure out how to get to classes of works that are not categories of works according to the legislative history? DR. VAIDHYANATHAN: Well, let's see. I think you're empowered to read the word "categories" as distinct from "classes." And I think that you're empowered to do that because they are two distinct words and two distinct areas of the code. If they had meant -- if the Commerce Committee had meant categories, it could and probably would have said categories. What is a class of work? Well, you can PAGE 70 define a class of work functionally, and I think that may be the only reasonable way to do it. A class of work is a work that is available in certain ways and used in certain ways. And then it's up to you to fill in the blanks what those ways are. MS. DOUGLASS: Do you have any comment on this, Dr. Gassaway? MS. GASSAWAY: I'm pretty sure it's not categories of works, because that's how I first read it when the legislation was being drafted and trying to figure it out. I think we've done the best job we can with looking at those fair use works and factual works. Suppose it could also be defined by length of term? You know, we could say that after the first so many years the work is no longer something that we need to worry about that for. I don't think that will be very popular with copyright holders, but you know, you're looking at different ways we could cut across what's a class of work. Old stuff. That's a class. Bad stuff. MS. DOUGLASS: Old stuff is a class PAGE 71 particularly if the copyright term has already expired. MS. GASSAWAY: Oh, then it's public domain. I'm talking about old stuff that's still under copyright. DR. VAIDHYANATHAN: But to articulate clearly that items in the public domain should not be covered by the anticircumvention prohibition because they're not covered under Section 107. MS. GOSLINS: I think that's pretty clear from the legislation. DR. VAIDHYANATHAN: Well, okay. We'll see if it's clear in practice. MS. DOUGLASS: What's not so clear, maybe, is public domain material that is covered by what some consider to be a thin veneer of copyrightable works. For example, an introduction, an index, a table, with all of these public domain facts from the SEC. And you happen to be in Denver, Colorado. So, are you advocating any particular exemption with respect, possibly, to that kind of PAGE 72 material? DR. VAIDHYANATHAN: The only way I could envision that working -- I would love to see an exemption for the public domain material. In other words, no one should be able to prosecute me for circumventing access to the complete works of Mark Twain on a protected CD-ROM, for instance. I would like that. However, I recognize that the complete works of Mark Twain are available in several other forms, not enclosed, not protected. So you may find too broad a definition on work along -- or work against the principle of the legislation in front of you. MS. GASSAWAY: Also talking about works abandoned in the commercial market, which are fixed and obsolete technology. It's another class of works that we could look at. MS. DOUGLASS: Thank you. Ms. Coyle, the summary of your statement refers to reformatting material, and the need for circumvention in connection with preserving material for archival purposes. Have you ever needed to PAGE 73 reformat audio/visual works or the like for storage reasons? For example, DAT or DVD format, to avoid maintaining items on more space-consuming media? Am I making myself clear? MS. COYLE: Yes. We do make copies of everything that we receive, and this is part of our licensing. And we do copy -- I mean, we have data going back to 1978, so I've had the privilege of going through system upgrades and having to recopy hundreds of thousands of files. So, yes, this is something that occurs actually quite regularly. MS. DOUGLASS: And you do it now as a matter of a license? MS. COYLE: Right, right. MS. PETERS: Is that something that you require in all of your licenses, or manage to get in all of your licenses? MS. COYLE: You don't manage to get it from all of them, no. MS. PETERS: Okay. What do you do when you don't get it? PAGE 74 MS. COYLE: I talk to the people who do licensing, because I don't do that directly. You know, I do the bits and bytes, other people do licensing. But I figured this question would be asked. What happens at that point is it seems that we go into prolonged negotiations with the vendor. And we have had contract negotiations last 18 to 24 months until we reach an agreement. MS. PETERS: But you keep going for an agreement. Maybe I should ask, what do you do with your archived material? Does that become the base from which you serve, or is it more like a doomsday kind of -- MS. COYLE: Yes. Actually, most of the material that we've archived is bibliographic records. And those at least were -- now they're stored in Oakland and San Francisco. There was a time when some of them were stored in Nevada as well, so that when California slid into the ocean our data would still be there. Fortunately, that hasn't happened. But yes, things are stored with PAGE 75 the idea that we think we have to keep it forever. Now, I should mention about digital preservation, because I think there's some misunderstanding about that. Which is that digital preservation -- with book preservation you wait 50 to 200 years, and as the book starts to deteriorate then you preserve it. With digital preservation you have -- preservation really begins on Day One. That preservation is really a kind of preventive kind of thing. MS. DOUGLASS: So you would not necessarily advocate having an anticircumvention exemption, because you would take care of it up front? MS. COYLE: Well, no. You don't take care of it up front. The problem is that -- I mean, if I get a CD-ROM and data is on CD-ROM, that data is protected. I can make a copy of that CD-ROM, the data is just as protected as it ever was. And, you know, I have no more access to that new CD- ROM than I did into the old one. PAGE 76 If I feel that that is data that I have to -- and that I have a right to keep in perpetuity, I am going to want it kept in a format that I know I can get to in 10 years. And a copy of that CD-ROM is not going to do that for me. MS. PETERS: So, you put it what? What kind of format would you put it on? MS. COYLE: It isn't just -- I mean, you're talking about the physical format? MS. PETERS: I mean, obviously -- MS. COYLE: The physical format isn't the question. The question is the data format. MS. PETERS: Right. But what does that become? MS. COYLE: Depends on what kind of data you have. MS. PETERS: So what do you do for a CD- ROM to preserve it? MS. COYLE: Well, it's a -- see, it isn't a question whether it's CD-ROM, it's what's on it. So, for example, our standard for preserving images is a certain level of TIF format. And if we PAGE 77 have images that we've received -- and again, most of the images that we receive are from our own archives. So it isn't a question of having to circumvent anything. But we have images that we receive. We put them into that format, because that's the format we expect to be able to read in 20 to 30 years. MS. DOUGLASS: I'm going out on a limb here. Dr. Vaidhyanathan? DR. VAIDHYANATHAN: That's good. MS. DOUGLASS: All right. Your summary, the first to a broader picture of information comment, and a shrinking of the public domain. And you say that it's going to affect decision-making and creativity and humor. And from the lack of humor here, that's a serious charge. So I'd like to know if you can connect that really broad charge to anything regarding circumvention. DR. VAIDHYANATHAN: The connection would be really clear under cases where copyright holders exercised editorial control over access. The minute that starts happening, then certain classes of PAGE 78 people get access to certain works or information, even data, if we're not careful how we write these rules. In which case, certain classes of people would have much higher ability to manipulate public discussion and debate. And perhaps people, economically marginalized or socially marginalized, would not have access to central texts, ideas and tenets of our society that might be worthy of satire. And as a result -- I mean, in connection to that, and I'm going to add this -- might as well add it now, as long as there are no follow-up questions about it, because I just learned about it. Cyber Patrol, the filtering software, the filtering service, apparently had been suppressing speech. It prevents you from viewing certain places on the Web, for instance. Apparently the encryption of the block list was broken. And as soon as activists discovered that Cyber Patrol was blocking sites not particularly defined by its policy, Cyber Patrol PAGE 79 blocked those websites that carried the criticism of the policy. Does that make sense? MS. DOUGLASS: Yes, it does. I've heard it before. DR. VAIDHYANATHAN: Glad you did. MS. DOUGLASS: And I suppose there's not been any similar -- I guess we could call it adverse effect. Or has there been any action taken? Have you heard, for example, that there's some sort of a parody of The Matrix on the Web? DR. VAIDHYANATHAN: Of The Matrix I haven't heard. But I have heard about a parody of an Elian Gonzalez photo, for instance, that also simultaneously parodied a major beer ad. And both - - I guess people received cease and desist letters as a result of this parodic manipulation. That's not a control over -- it's not a technological control over access to this stuff. Nobody really has a problem with access to Budweiser ads. If only we did. But it was a case where the cultural power of the copyright system is used to try to stifle parody and free expression. PAGE 80 MS. DOUGLASS: Thank you for adding humor to this testimony. MS. PETERS: Thank you very much. Rob? MR. KASUNIC: Good afternoon. Going back, there was a lot of broad concerns with the things that are probably outside the scope of the technological controls. And I think in some of the comments, we've seen that there is a -- Ms. Coyle's testimony that it's too early for any of us to make any definite statements about some of this. And also Dr. Vaidhyanathan, that the potential for abuse is there. And also admitting that the fears are speculative and alarmist. That these are maybe significant concerns, but it's not clear that we've reached a certain point yet in the number of works. In addition to -- Congress set up, in addition to this triennial review that the Copyright Office is empowered to do every three years, that Congress also in the legislative history anticipated that the market would be a factor in controlling PAGE 81 this. That if controls got too tight, then the market would compensate, and that there would be effects on that side. And in addition, the Copyright Office is there to review this at another point in time, if some of these situations do get worse. Is there any evidence, in any of your views, that this is not likely to be the case? That Congress was wrong, that the market or that the pressure of knowing that the Copyright Office would be reviewing this again would not be enough to alleviate some of these potential fears? DR. VAIDHYANATHAN: The second concern first. Yes, your triennial review is not likely to have a direct effect on mitigating any of these harms. For the simple fact that technologies and devices will still be illegal. So once again, it's the right without the ability. Addressing your first concern, once again a chilling effect is a real and tangible factor in the way that the public and creators interact with media companies and the copyright law PAGE 82 system. And any gap in understanding of the nuances of the Digital Millennium Copyright Act in general, or the anticircumvention prohibitions specifically, are not only likely to have a chilling effect, I'm sure that's already happened. If you don't realize what your specific rights are, chances are you're not going to exercise any of your given rights. MR. KASUNIC: Did anybody else have -- yes? MS. GASSAWAY: I should mention that we have a little bit of concern about how the market for education is being viewed, generally. And I'm not talking about just for materials that are designed specifically for that market. But as copyright holders talk to you all, everything they talked about market seemed to be aimed at an individual. You know, how are we going to deal with getting access to these works in the educational context, if everything is set up so it's an individual who gets access? As opposed to a license that we're dealing with now. PAGE 83 And the market simply has not worked that well, when it comes even to licensing. Especially some of these works for entertainment. I mean, look at the evidence we saw when we were talking about distance learning, about just being denied even after the school had purchased the work, denied the right to use it for distance learning. I know we shouldn't talk about that it's too expensive, but I'm not talking about that. But quoted fees that mean you really don't use it. And so I'm talking about exorbitant, not just a little on the expensive side. But which is clearly a way of controlling what's going on in education. And I think that's one of the concerns I have. It's really this whole First Amendment. I mean, what are we going to be able to use for teaching? Especially with things like cultural studies. It's really a control on what is going to be taught to your kids. MR. KASUNIC: Well, then, I guess that gets back to a comment that was raised before about what is the purpose of the exemption in this, and PAGE 84 how will this really help. Because if the market has not been doing its job in maybe restrictive licensing, that's -- this exemption isn't going to change that. And it's also not going to have an effect on the technology itself. The technology can be as restrictive as anyone wants to make it. All we can deal with in any potential exemption is the prohibition on circumventing that technology. So what will the effect of an exemption be here? MS. COYLE: I actually think that the exemption will have an effect on the licensing and contracting. Because I think that it gives a message, and it gives a message that we expect libraries to be providing information to the public, and to be archiving the information. And I think that it helps support what libraries are trying to get into their licenses, which is the ability to do just that. MR. KASUNIC: Well, then, in terms of the message that is to be sent -- and there does seem to be an interest in the Copyright Office in PAGE 85 sending some kind of message here. But under certain restrictions on what kind of message we can send. And we do have to identify classes of work. In terms of how -- it was raised, going back to the classes of works and the determination of categories. Were given in the legislative history something that did, to a certain extent, tie this to the categories, where it was talked about that, given examples, that this could something narrower or should be something narrower than an overall category of works, and not something like audio/visual in general. But more narrow as in motion pictures, but maybe not so narrow as in some particular genre within there. So how do we take this out of that area of limiting it to one particular category, to a broad -- to having a class of works which spans a number of different categories? One where we do have this legislative history that does seem to narrow the scope a bit. MS. GASSAWAY: We gave you a bunch of examples in the testimony that I delivered, talking PAGE 86 about factual works and fair use works. And named within those specific types that appear in different categories in Section 102(a). MR. KASUNIC: Okay. With the thin copyrights, and that was one of the areas you talked about in the types of work, with factual works. And this is something that was raised with a number of the database owners and interests. That there is a claim that this is something that's covered under Title 17. That while there is a scope of protection is arguable, that it may not be as completely as broad, that this is under Title 17. And that the technology is controlling a work that's protected by Title 17. How can we work with that restriction, that it is something that's covered. If the technology is covering both copyrightable elements and factual material, how -- and is not differentiating between the two, is that something that should be able to be protected under Section 1201? MS. GASSAWAY: I think it's relatively easy to do it. You simply would say that for PAGE 87 educational, scholarly research purposes, even though those works are protected under the Copyright Act, and because we've had lawful initial use, then they are exempted from anticircumvention for these library, education, scholarly and research uses. DR. VAIDHYANATHAN: And are you actually talking about, for instance, databases with some original arrangement and -- is that what your question is about? So there's partial copyrighted material on a particular database, but the data itself -- which is not covered under the Copyright Act -- but you're saying how can you help draw that distinction, or what should you privilege? MR. KASUNIC: Right. Well, if we have a technological control measure -- that as long as there is some element that would be copyrightable, that that can be applied to the overall work. While that may contain public domain elements, do we open up the -- certain copyrightable elements? How does an exemption differentiate between the two? DR. VAIDHYANATHAN: You should err on the side of public interest, you should err on the PAGE 88 side of factual availability and the free flow of information. You should, for instance, say that not just teachers and scholars should have access to non-copyrightable elements of a particular work, but all potential users should have access to that information. Especially if it's the only place one can get it. MR. KASUNIC: So that would be a restriction, then, on it. That if you have a database that's a sole source of that, and if this was something that was available in some other form, then that would not be -- that would not fall under that exemption? DR. VAIDHYANATHAN: Well, once again, you have to take into account accessibility for all users. Users in Alaska, users who are visually impaired, all of whom should have an equal ability to manipulate factual information. MR. KASUNIC: Well, I guess, Ms. Coyle, that that was something that -- raising about with the sole source. And it's unclear that this has increased the benefits to society, having some of PAGE 89 these in a digital format as opposed to if it's maybe available only in hard form. That having some of these work available digitally has increased the number of users and types of uses that are -- MS. COYLE: That simply changed the kind of use, although this has just begun and we don't really know exactly where that's going. But I think we're seeing a change, actually, in the type of scholarship that takes place because of a new kind of availability of information, which was previously available in a different form. MR. KASUNIC: And are you aware of anybody who has looked into doing some of -- you've mentioned social impact studies. And I know there was some interest in the Copyright Office doing that. But is that something that is going to be looked into by libraries and other areas, to determine what some of these adverse effects are? MS. COYLE: I don't know of anyone who's really planning to do something that I would consider to be a study of that type, no. It's going to take effort, it's going to take people's time, PAGE 90 it's going to take gathering together a group of experts. MS. GASSAWAY: I think I would worry a little bit about -- talking about these databases and things that are copyrighted. We're not arguing about the copyright status of those works, but that the veneer of copyright should not be used to bootstrap circumvention prohibition for all non- protected material. I mean, I think it's turned the other way. We're simply saying the whole point is we've had initial lawful access, so perhaps we've had a license, whatever. These are thin copyrighted works because of the use that we are making of them in education, research. Talking about students, faculty, all the way up, libraries. That because of all of that, there should be an exemption in this anticircumvention for that class of works, those that have thin copyrights to begin with. DR. VAIDHYANATHAN: Access to a copyrightable veneer of a database is not infringement of a copyrightable veneer. PAGE 91 MS. GASSAWAY: That's right. DR. VAIDHYANATHAN: And it's really important that we not conflate use and infringement. Or somebody has to because Congress didn't. MR. KASUNIC: Well, just to clarify one point, in terms of -- when you said that the initial lawful use, or initial lawful access to a work -- and that's something that had been raised in some of the other testimony by Peter Jaszi and Arnie Lutzker as well, about initial lawful access being a criteria. That access would then, I think in Professor Jaszi's statement that that was lawfully acquired work. So you're talking about this being expanded to lawful access in terms of the licensing, but being restricted to the terms of that license so that not -- okay. Just wanted to clarify that. Thank you. MS. PETERS: Can I just follow on what was Rob's question? Is your focus on initial access against a pay-per-use model? So that you have to somehow trigger a payment, or another dime, whatever PAGE 92 it is, to get use -- to be able to look at it again? MS. GASSAWAY: Not necessarily. It could be that you took a blanket license of some kind. It could be that you acquired it by gift. You know, I mean, when we look at works, you might have purchased it. If it's an outright purchase. So it's sort of all of those ways one lawfully acquires a work, whether possession or access to it. MS. PETERS: Okay. Let's say that you got it by gift. And it has, what, an access control such as a password. So you don't have that password, but because it came to you lawfully as a gift, then you have the right to circumvent that access control? MS. GASSAWAY: Yes. Unless the license to the person who acquired it initially required that they not be able to give it away. MS. PETERS: Right. MS. GASSAWAY: If their license did not prevent that, I guess we could look at software under 109(b)(2)(A), whatever those long numbers are. You know, the library and the education exemption PAGE 93 to the computer software amendments. Because we are allowed to give it away to another educational institution. So we give it to them, they don't have the password. Yes, they should be able to circumvent that. Because under the statute we're allowed to give it away to them. MS. PETERS: Take software as an example. MS. GASSAWAY: Okay. MS. PETERS: Under software, you -- well, I won't say you. People, libraries have the right to lend that software. MS. GASSAWAY: Yes. MS. PETERS: Is the software ever, like, password-protected so that when the people get it home, they have a problem using it? MS. GASSAWAY: I'm probably not the best one to ask about that, Marybeth. I'll bet some of the people who are in other kinds of libraries -- in law libraries we don't do much of that, loaning software. PAGE 94 MS. COYLE: Yes. It's not very -- so, I don't know. I mean, I assume that if that's the case, then you would, along with lending the software, you would have to give them the access password. Because otherwise they couldn't use it, and why would you have lent it to them if they couldn't use it? MS. PETERS: Okay. MS. GASSAWAY: So they could look at the floppy. MS. PETERS: Okay. David? MR. CARSON: Well, following up on your second to last question. Professor Gassaway, and really everyone, are any of you aware of cases -- and I think I've heard one or two, but I just wanted to get sort of a checklist in my own mind of cases in which technological measures have restricted access to works, beyond existing contractual restrictions? MS. GASSAWAY: The only personal experience that I've had is the one of the disappearing CD-ROM content. That's the only one I PAGE 95 have personally seen to date. MR. CARSON: Okay. That was the Lexis French database? MS. GASSAWAY: No, that was the Westlaw CD-ROM. MR. CARSON: And that one, I think you said, was a mistake, right? MS. GASSAWAY: It was a mistake, but they said it was a mistake. But we don't know whether it was. MR. CARSON: Did they correct it? MS. GASSAWAY: Yes, but it took them seven weeks to correct it. So we were seven weeks - - MS. PETERS: You had no access for seven weeks? MS. GASSAWAY: We had no access for seven weeks. MR. CARSON: Okay. Anyone else aware of any cases in which something -- technological measures restricted access beyond terms that were in a license that you had? PAGE 96 DR. VAIDHYANATHAN: The Cyber Patrol case would fall under that. MR. CARSON: Okay. All right. MS. COYLE: Well, I still think it's interesting because you're assuming that the only time that people can't get in is when the technology deliberately is keeping them out. And I think we can't assume perfect technology. And I think that the example that Lolly gave is a very good one of that. Technology fails. It actually fails quite regularly, and so it fails even though you may still be within your contract. DR. VAIDHYANATHAN: A real fresh example of that that just happened a couple of days ago. There's a new subscription-only website for media critics and scholars called "Inside.com." It's planning to charge $20 per month for an access fee, and therefore it's going to be password-protected. For their start-up they sent out e-mails to specific people on a specific list, saying "We're going to give you a month of free access. Go to this page and register with us, and we'll let you PAGE 97 in." So I got the e-mail, I went to the page, I registered, gave them all the information they needed. And their link, their connect button was not hot. There was a glitch in the system, so I didn't have access to their information, even though I gave them everything they asked for in our contractual deal. MR. CARSON: Do you have any reason to think that was anything other than a mistake on their part? DR. VAIDHYANATHAN: No, no. Whether it was a mistake or not, their effect is the same. MR. CARSON: Okay. Let's take that further, though. Because what we're talking about here is whether there should be classes of works with respect to would you circumvent. And let's assume we were to make a class of works as being those works which, by mistake, access has been denied you, even though you have a contractual right to. DR. VAIDHYANATHAN: Yes. That would be PAGE 98 great. MR. CARSON: I'm not sure that's an appropriate class under the law, but let's just assume that for a minute. DR. VAIDHYANATHAN: Well, it might be an essential class under the law. When I'm teaching that class in 2035, and I plug in my access code and the film doesn't come, it may not have come because of some evil intent. MR. CARSON: Oh, I understand. DR. VAIDHYANATHAN: It may have not come because of a mistake. And if I have a really brilliant student who's willing to hack the system right then and there to get me in, I shouldn't be prosecuted for lawful access to that film. MR. CARSON: Okay, you've actually started to answer my question, anyway. Because the question is -- well, first of all, let's have another -- let's build another assumption in this. Let's assume, because no one thus far has had a contrary experience, that when that access has been denied by mistake, the content provider, once being PAGE 99 advised of that mistake, takes corrective action. It may take him a little while, but they do take corrective action. Let's assume good faith by the content provider. As a realistic matter, is the ability to circumvent something that you could take advantage of? I mean, is it something that, when you have encryption or whatever that is preventing you from getting in there, that you could virtually instantaneously circumvent anyway, quicker than it would take the content provider who's acting in good faith to correct the problem? DR. VAIDHYANATHAN: It's impossible to predict. Because it's impossible to predict the level of technological expertise among those who seek access. And it's impossible to predict the level of technological barrier set up by the content provider. It's also impossible to predict the chain of communication it would require through any complex system, to correct the situation. And my semester might be over before Casablanca plays. MR. CARSON: Let's take the Westlaw PAGE 100 example that we did have. Do you have any reason to believe that you would have had means to circumvent that in a timely way? MS. GASSAWAY: No. The only way we could have circumvented it is if another library had the same title, and we just simply copied it. Because there was nothing on the CD. We had just a blank CD. MR. CARSON: Okay. Let's take the Lexis French database. That's something that I assume you had online access to, and at some point it just disappeared? MS. GASSAWAY: That's right. MR. CARSON: I'm not sure how relevant it is, but do you know why it disappeared? Was there any explanation? MS. GASSAWAY: No, we don't. MR. CARSON: Okay. Can you tell me how a legal right to circumvent technological access control measures would have prevented the problem that you ultimately had? MS. GASSAWAY: It wouldn't have, David. PAGE 101 I was just, at that point, really talking about -- you asked for examples of what had happened so far. That one was very fresh because of having happened recently. Because there was no announcement, no explanation why, just one day it's gone. MR. CARSON: Okay. I'd like all of you now to assume for a moment what may or may not be the case. Which is that we decide that we are going to exempt only those classes of works with respect to which you can demonstrate that users have already suffered serious adverse impacts on their ability to engage in non-infringing uses. Are you with me so far? Okay. In that case, can you tell me what those classes are, and what are the impacts that have already occurred that you can identify, that would justify selecting those classes? I guess I'm asking you to tell me what classes there are with respect to which there have already been those serious adverse impacts. MS. COYLE: What classes? Okay, ask me that again. I'm lost. What classes? PAGE 102 MR. CARSON: First of all, the premise is that we will exempt only those classes of works with respect to which users have already had serious adverse impacts on their ability to make non- infringing uses. MS. COYLE: In other words, we've had an experience with this in the past? MR. CARSON: Absolutely. MS. COYLE: Oh, okay. I had understood your statement differently. MR. CARSON: All right. If that's the case, if the law says we can't exempt a class unless we've already made that finding. MS. COYLE: There's proof that something's already gone wrong somewhere, yes. MR. CARSON: Yes. Then can you tell me what classes there with respect to which that condition has been met? DR. VAIDHYANATHAN: Yes. There are people who can't play digital video disks because they didn't buy, perhaps even couldn't buy a particular brand of digital video disk player. PAGE 103 Let's say, for instance, they have a computer with Linux running on it, and they want to play their lawfully acquired digital video disk. Then, yes, that particular situation would come up. MR. CARSON: We expect to have some long conversations on that subject tomorrow. Anything besides that? Okay. Professor Gassaway, the first class of works that you asked us to examine -- and correct me if I've got this wrong -- is works with respect to the user has initially obtained lawful possession, is that correct? MS. GASSAWAY: It's lawful access. MR. CARSON: Lawful access, okay. How do you square that with the requirement that we identify a particular class of works? Is that a class of works within the statutory meaning? MS. GASSAWAY: I think you have broad discretionary power here to accomplish that however you want. And if you want to define the class as that to which the user had initial lawful access, I think you can do that. MR. CARSON: Well, you're a law PAGE 104 professor. What do you find in the statutory language or the legislative history that suggests we have that much discretion? MS. GASSAWAY: I don't find anything that says you don't. (Laughter.) MR. CARSON: Fine. Let's move on to your class of fair use works. I happen to be a strong believer in fair use. And I guess my question is, aren't all works fair use works? In fact, my experience is that some of the most interesting fair use cases, and the ones that I find myself believing most strongly about are the cases in which the work -- with respect to which fair use is being made, are highly creative works. MS. GASSAWAY: Right. MR. CARSON: And if that's the case, then are we exempting everything? MS. GASSAWAY: Well, I think we started out by saying, "Look, for all of higher education there are two groups of works that we think all disciplines use, and maybe have -- the equity's just PAGE 105 even a little stronger than anything else." And that was those fair use works and the thin copyright. But we did say, in addition, we thought that you might think about writing a regulation that would exempt a work not in those categories, these creative kinds of works. Because of the surrounding context, like use of motion pictures in a film school. Where we wouldn't say those are works that would automatically fall into that, because of the context they well might. And that's what I said, I think in higher ed we do not differentiate between the types of works. You know, we just don't. We consider an audio/visual work the same thing that we consider a literary work. The Copyright Act differentiates them, but teachers do not. DR. VAIDHYANATHAN: I also do think it's important that we not be in a position to, for instance, license teachers and professors to have greater access to works than, for instance, my mom. I mean, all users should have equal access to these PAGE 106 works. And therefore, actually coming up with a notion of an actual -- I'm afraid it's a category rather than a class -- fair use works might disrupt that. That's why I'm not really on board with a specific definition of a class of fair use works. All works are potential fair use works. MS. GASSAWAY: I don't agree with that. I think that then doesn't that gut the whole provision of the law? You're going to exempt everything. DR. VAIDHYANATHAN: Yes, that would be great. MS. GASSAWAY: And that's why I think we said, you know, given our druthers, we would start with this. But we also have to look at the fact that classes of works did not mean the 102(a) categories. What does it mean? And there are different ways to cut it. And we've mentioned date and some other things. But the ones that really made the most sense might be those fair use and thin copyright works if you PAGE 107 cannot go as broad as looking at that initial lawful use-- initial lawful access. I m sorry. MR. CARSON: Okay. I'd like some help with my legal analysis here, so I'm going to primarily look to you, Professor Gassaway, on this one. MS. GASSAWAY: Hot seat. MR. CARSON: This is a question of interpretation of Section 1201. MS. GASSAWAY: Oh, great. MR. CARSON: When we -- let's assume that we recommend that the Librarian exempt a particular class of works, and let's assume that he accepts that recommendation and exempts it. Is it your understanding that if a particular class of works is exempted, all users of that class of works are entitled to circumvent technological measures that control access? Or alternatively, only that users who are engaging ultimately in non-infringing uses are entitled to circumvent? MS. GASSAWAY: Now, I'm not able to -- like I can in 108 and 107 and 110 -- spit out the PAGE 108 sections without doing much looking. But I thought that it said any class of works that are subject by -- non-infringing uses by persons who are users of the work are who are likely to be affected. So I think you can cut it different ways. It could be everyone, it could simply be because of the public good of education and libraries. It could be that way. I think you have a lot of discretion there. Because I think it does talk about particular persons and users. MR. CARSON: Okay. Well, let's say we decide that the databases are going to be exempt, and that's all we do. Because let's assume for a moment -- because I think this is probably the plain reading of the statute -- that all the Librarian does is say, "The following classes are classes I designate as falling within those categories." So the ultimate regulation just says we find the following category, database. Would that mean, in your view, that anyone can circumvent a technological measure that controls access to that database? Or would it mean, on the other hand, that PAGE 109 only people who are engaging in non-infringing use of that database can circumvent? MS. GASSAWAY: I think the words of the statute say that non-infringing uses by persons who are users of a copyrighted work. So I'm reading from the bible, and it says non-infringing uses. So that's certainly my own interpretation, that it is - - MS. PETERS: They're in D. MS. GASSAWAY: Pardon? MS. PETERS: You're reading D, right? MS. GASSAWAY: I'm reading D. And I really -- now, this may just be my own foggy notion of it, but all along I thought that not only could you define classes, but classes for particular users. That it did not necessarily have to be as against the public, generally. That would be great. But it could also be against particular classes of users, from the way I've read this. But Ms. Peters and I were both at a conference where we heard a copyright law professor say not only would it be nice if the statute could PAGE 110 be read and understand by normal human beings, it would be nice if it could be read and understand by intellectual property professors, so-- DR. VAIDHYANATHAN: Getting back to that example, I'm not sure how one could infringe upon a database. I don't think we've come up with a set of situations, unless you're actually talking about infringing the copyrightable portion of that database. MR. CARSON: Well, sure. DR. VAIDHYANATHAN: So, yes. So exempting databases would be an irrelevant exercise. MR. CARSON: Well, I think the database owners might disagree with you on that. MS. GASSAWAY: Yes. I disagree with that. I think that there certainly are portions of databases that are copyrightable and therefore subject to infringement. So that certainly could be one. I would be surprised if it would be exempted as against all uses, because that would also include competitors for the database, rather than those users for what are traditionally fair use PAGE 111 purposes. MR. CARSON: Professor Gassaway, you also said that some copyright owners have previously testified in this proceeding that they intend to merge access control measures and use control measures. Did I get that right? MS. GASSAWAY: Yes. MR. CARSON: Okay. Just a suggestion. It would be very helpful for us if either when you correct your transcript or in post-hearing comments, if you could identify those particular people and where in their testimony we could find that, you'll save us a little bit of work. MS. GASSAWAY: We'll do that. MR. CARSON: Okay. MS. GASSAWAY: Naturally, they didn't use exactly those words. MR. CARSON: That's why it would help for you to identify exactly what it was they said, so we can come to our own judgment as well. Ms. Coyle, you said you're not a lawyer and therefore you can't -- PAGE 112 MS. COYLE: Definitely not. MR. CARSON: Well, congratulations. (Laughter.) MS. COYLE: Thank you. MR. CARSON: And that you can't describe in legal terms what exemptions we can recommend, and I can certainly understand that. But as someone who's out there in the field, struggling with these issues, can you tell us as a practical matter what kinds of things should be exempted from this anticircumvention provision? Leave it to us to come up with the legal language. You tell us the problem and what kinds of works really are at risk here. MS. COYLE: I think, as you've heard in the other testimony, I can't think of a type of work that isn't at risk. As long as it's digital and it's protected, I believe it's at risk. MR. CARSON: Okay. Professor Vaidhyanathan, you said that one of the types of works you'd like to see exempted would be works that are not easily and widely available in unsecured PAGE 113 formats. Can you give us concrete examples of what kind of work you're talking about? DR. VAIDHYANATHAN: Let's see. Well, this is a skimpy concrete example. There are certain articles that are available only on the New York Times website, not available in the paper product. If the New York Times website were protected completely, which it basically is password protection at this point. Then, yes, that material would have to be exempted under my model. Exempted from the anticircumvention provision. MR. CARSON: Okay. You're giving me what I think is really a hypothetical. Because you started saying "if." DR. VAIDHYANATHAN: Well, it is protected by a technological gate right now. You can't get into -- MR. CARSON: Which anyone can get into, having done it myself a number of times. DR. VAIDHYANATHAN: Well, you and I aren't everyone. We don't know if everyone can, and we don't know for how long, and we don't know under PAGE 114 what conditions they still say yes or no. They've only said yes, as far as your experience or my experience indicates. But that doesn't mean they can't or won't say no. And we haven't yet found the person to whom they've said no. However, that is a technological gate. And circumventing it in order to get access to a particular article that's not available in print form should be exempted. MS. PETERS: Okay. I did something brilliant. I went and let my very able staff go first, and I'm looking at all the questions that I have. And actually I think almost all the questions that I had, I've asked throughout or others have basically answered them. So, I think maybe for me I don't have anything at this point. Does anyone else on the panel have anything that they're dying to ask? No? Okay. Let me just make a note. In the proceeding I noticed at least one person raising their hand. And I didn't recognize that person PAGE 115 because this is a formal hearing in which people had to give notice, and they had adequate opportunity to testify. If there's anyone in the audience who feels very strongly that they want to say something, we do have another comment period. And you certainly can file comments by June 23rd. I certainly would like to thank the witnesses. You've been extremely helpful, and we've kept you quite a while. So thank you very much. And for those -- I see some people who will testify tomorrow. We hope to see you here. And anyone else who wants to come. Thank you. (Whereupon, at 4:18 p.m., the hearing was adjourned, to be reconvened Friday, May 19, 2000, at 9:30 a.m.)