Copyright, Fair Use and Electronic Reserve

Reserve systems in academic and research libraries have long been a staple among services offered to student patrons, especially in the post-World War II higher education boom that went into higher (sputnik-level) orbit with the GI’s baby-boom progeny in the 1960’s. As a staple, reserve has never received much attention in the research literature of librarianship – it’s one of those things academic libraries have done along with providing copy machines, Masterplots, and scissors at the reference desk. From an information retrieval viewpoint, reserve is mundane – known-item retrievals from a comparatively small document store, hardly more interesting than assigning Cutter numbers. Though still undervalued by administrators despite the large numbers of students served by them, reserve systems are, in my view, very interesting in one respect: they offer a model for class-associated document delivery which conforms to conditions of Fair Use as spelled out by Section 107 of the 1976 Copyright Act. This conformance between a library system and a clear provision of the Copyright Act gets especially interesting when we contemplate what the rapid growth and acceptance of the Internet offers us in terms of electronic delivery of documents. I hope in my talk this morning to convince my audience that interest in reserve systems, especially what has come to be known as electronic reserve, is indeed warranted in our field, and that the electronic reserve models now being developed in many library and computing organizations offer guidance for the future of electronic libraries generally.

 I argue ultimately that electronic reserve systems are not only of interest, but important, important not only to the droves of undergraduate students who may dutifully read what they are asked to read by their teachers, but to the progress of digital libraries. As a part of that argument that electronic reserve systems are both interesting and important, I will refer in my conclusion to another, older, debate that occurred at the intersection of commerce and library system design nearly twenty years ago, because I think the resolution of that older debate is instructive for our present situation. I need to start, however, by providing this audience with an overview of electronic reserve systems as they exist today, and an overview of the intellectual property issues, at least as I see them, based on the text of the 1976 Copyright Act, the conflicts that have occurred since then, and the more recent hoop-la surrounding the Conference on Fair Use (or CONFU). The review of copyright issues surrounding electronic reserve will outline what has come to be referred to, not initially by us but by the American Association of University Presses which endorsed it, as the "Northwestern model" for e-reserve copyright policy. I hope to persuade you that this model is both workable for other academic libraries and of benefit to publishers who are concerned about their intellectual property rights in a networked environment.

 What is the older controversy that the new clashes surrounding digital libraries and copyright causes me to recall at the end of this paper? It the fee or free debate – an issue which consumed much of the library press in the late ‘70’s. For benefit of those not in the field at that time, "fee or free" referred to the outcry over pass-through charges that libraries began levying in the early years of online literature searching. The dominant administrative view favored fee-charging for pragmatic reasons, while many front-line service providers decried the differential service standards such fees highlighted. The relevance of this now mostly-forgotten controversy for a current perspective on electronic reserve and copyright will be explained later, but in this preface let me just observe a common underlying text – the conflict over values brought on by technological change.

 Let me start now with Northwestern’s story about electronic reserve. Ours was not the first automated system to provide course readings online, but it has received fairly wide recognition for its ease of implementation and for the technologies it has integrated on a campuswide basis. The work done elsewhere at Rice University, Duke University, and San Jose State, most notably, is documented in the literature, and I’ll be glad to share what I know about that work in our question-and-answer period. I also want to make sure we also give a special nod of appreciation to a home-town fellow on the e-reserve front for his great work in maintaining a website on the topic, with pointers to many existing systems – Jeff Rosedale of Columbia University. Though our purposes for today’s conference are not to review technologies but to consider policy, I think it still useful to share some of the nuts and bolts, as the accessability of this technology now has definite implications for policy development among both libraries and publishers. (Fuller technical details on Northwestern's electronic reserve system were offered at a 1995 CAUSE presentation.)

 After I made a career move from the library world to the academic computing world at Northwestern, I was put in charge of developing electronic conferencing systems for Northwestern courses, and it was out of that area of responsibility that the work on designing an electronic reserve system for NU began. I had established both listserv and more advanced groupware applications for faculty teaching courses at Northwestern in 1993. In conversations with some of those faculty on how they were using those tools, I found that at least some teachers were interested in strictly one-way communication with a group of students, that is, interested in distributing their materials out to students electronically but not interested in the students having electronic conversations as a part of class participation. This is clearly a document distribution problem, not a conferencing application per se, and with tools easily at hand I developed a prototype electronic reserve system in the Summer of 1994 which, with terrific support, contributions, and critical input from library folks at Northwestern became operational in October of that year.

 Two very inexpensive technologies provided the building blocks for a system that has since been adopted and improved upon elsewhere: gopher and Adobe Acrobat. Gopher technology has of course now been superceded by the Web, but at the time of our’94 work still dominated as the technology of choice for campus wide information systems. Adobe Acrobat, which at the time we started was in version one and also not first in its product category, has in hindsight proved to be a good choice, as the PDF format for document exchange over the Internet is now widely in use, and the software continues to be developed very nicely. We knew through contact with Adobe that the Acrobat Reader software would become available at no charge by the end of 1994, and we felt confident that the integration of gopher and Acrobat would not only serve us well at Northwestern but also be easily picked up by other universities interested in implementing an electronic reserve service. The transfer from gopher to a WorldWide Web interface, accomplished by Claire Dougherty of our University Library, was very easy, and we expect the web interface to serve us for some time to come. No grant was required to do this, no big infusion of funding from an outside technology company, just an interest in doing it and a bit of time over the Summer to build it. The result has been a store of electronic documents, arranged by school, department, instructor, and course, which is available 24 hours a day to every NU student on and off campus and which can accommodate not only text but diagrams, color illustrations, hypertext documents, spreadsheets, and any other filetype which an instructor might demand.

 So the technology part is easy. If that’s all there was to it, end of story, and we wouldn’t be having this conference here today. But of course that’s not the way it is, and it’s the copyright issue that makes the story both longer and considerably more interesting. My initial perspective when we came to consider the system a production operation for e-reserve in the Library was that we should use it strictly for faculty-prepared documents, thus dodging the sticky issue of copyright in order to just have something that worked more effectively and economically than the listserv-based course material distributions that it was to replace. Thankfully, library staff at NU disagreed, taking the position that if a system couldn’t deliver the same sorts of published paper documents that faculty were used to providing the reserve department, it wasn’t good enough and didn’t deserve to be called electronic reserve. This strongly-held position led us to examine more carefully our assumptions about copyright and fair use, and in the end to put forward a policy formulation that we feel serves both higher education and the publishing community better than what we’ve had in the past.

 Any serious examination of the copyright implications of Internet-delivered documents must begin with a careful reading of section 107 of the 1976 Copyright Act. The rationale of the "four factors" which must be taken into account when claiming a fair-use exemption – the purpose and character of the use, the nature of the copyrighted work, the amount of the work, and the potential market impact – is well-known, and I think for the most part understood by libraries. But what a re-reading of the section brought to me was that the clause in chapter one of the section, directing that fair use leaves non-operative the protections provided copyright owners in Section 106, was so clear and unequivocal. I quote from the chapter: "...the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by ... section [106] , for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright." The four factors may be subject to interpretation as to their application, but, once it is determined that a use is fair, that these factors have been appropriately weighed in the context of a specific set of circumstances, then the application of the law is clear: the copyright holder’s rights in that context do not exist. This perspective, it seems to me, demands that those of us whose jobs entail service to the public be especially careful that the public interest be served through appropriate exercise of this section of the law. Judgment, care, attention to context and circumstance must be applied to the determination of whether the four factors are met, but, once met, there need be no further hesitation: The copyright law, from its initial embodiment in the United States Constitution, serves ultimately the public interest, not the interests of intellectual property holders. A retreat from this position would be to betray the public trust that has been given educators, librarians, and others whose responsibilities include the handling of information in all forms for any public purposes.

Seen from this perspective, a critical objective any electronic reserve system must meet becomes the provisioning of documents otherwise under copyright in such a way that the four factors are clearly addressed, so that a public right to utilize such documents is not eroded through atrophy in the electronic domain. If the promise of digital libraries, if the hopes within the Clinton-Gore administration, the educational policy development community, the Internet community are to be realized, intellectual property must have a public element as well as a private one. Society does not work well if everything is for sale, if new information which has value is only to be somehow bartered for as if it were all secret. The promotion of science and the useful arts is a public good, and the monopolies on intellectual property granted by the law must be understood to be limited by the terms of that law, not expanded because the vagaries of technological change may change the conditions on the playing field a bit. So let’s respect the rules, but keep the ball in play. Such became the consensus at Northwestern after a number of task force meetings, reviews with the University librarian, and conversations with the University’s General Counsel.

With that resolve, Northwestern undertook a careful review of reserve policies, in both the electronic and paper realms, to ensure that application of the four factors in Section 107 of the law would be consistent, fair, and protective of the legitimate interests of copyright owners while at the same time not constraining rights under the law which served educational purposes. We looked at existing guidelines, noting the inconsistencies and anomalies so well teased out by Kenny Crews’ 1993 book, Copyright, Fair Use, and the Challenge for Universities: Promoting the Progress of Higher Education (University of Chicago Press), and tried most of all to be conservative in our rule making. What we came up with can be found on our electronic reserve web site. Let me briefly outline our practices here.

 First of all, we designed the gopher site, then the web site, so that it would serve the needs of course-based instruction rather than any broader purposes of document providing. A user even at Northwestern, in other words, cannot easily retrieve some popular or well-known document simply because it has a high likelihood of being made assigned reading in some course, and therefore likely to be in the electronic reserve document pool. Intellectual access to specific documents is enabled only by a browse function in which the document seeker must specify the particular class within which the document is to be used. Jughead in the gopher environment, and now SWISH in the web environment are offered as search engines (free, incidentally) to enable users to quickly traverse the hierarchical tree of schools, departments, and courses in which specific course readings are arranged, but the search indexes are explicitly built to stop at the course level. I can search for any courses taught by Professor Smith, for instance, or for my course on early childhood psychology, but not for Donald Campbell’s well-known book chapter on quasi-experimental design. This strategy protects the copyright owner’s right to capture potential market value for the intellectual property in question in contexts other than the specific teaching purpose of the course.

 Several other protections of the interests of copyright owners are built into the handling procedures the library uses in processing e-reserve materials. Among these are the following: A copyright warning statement appears as the initial page of every Acrobat document in the system, informing users that further dissemination of the electronic document beyond the class is prohibited. All non-Northwestern accesses to the gopher/web site are blocked by IP domain below the search and browse levels, so that all outsiders can see are our introductory e-reserve page and the "About" pages which detail our policies and offer free download of the Acrobat Reader software for those who may still need it. Copyright use permissions are sought for any materials which are used by an instructor in a course if the course is offered more than once, and royalties paid as appropriate. Materials in the reserve document pool are made available only during the period in which the course using it is taught. No more than one chapter of a book nor more than one article in a journal issue will be placed on reserve for a course. In all these ways we work to protect the interests of copyright owners, and we continue to evaluate our procedural and technological options to ensure owners’ rights are respected.

 What do we say to publishers who argue that electronic reserve systems such as ours violate their copyrights? First, we say no, they do not; Section 107 of the law exempts those uses we make of their intellectual property from copyright protection. Second, and I think far more usefully, we ask publishers to provide us material already in electronic form at reasonable prices so that we may continue to improve a service that is enormously appreciated by students and faculty alike and that has become the single most widely used application on our network, and in our labs after word processing. We envision licensing arrangements for e-reserve items as a new and important revenue stream for publishers committed to quality intellectual production and the goals of humanistic education. As institutions, we’re certainly willing to pay for product that serves our goals and benefits our students. As we are with major database purchases and major software site license acquisitions, we’re looking for publishers who "get it." We’re not interested in publishers who whine to get changes in the rules because the wind changed direction.

 Before I go on to consider elements of the aborted CONFU process for electronic reserve, I’d like to offer more reasons to encourage publishers to support the development of e-reserve systems rather than fight them. If there are publishers in this audience, and I hope there are, please reflect for a couple of minutes on the emerging technological environment within our institutions of higher education, our still practically prenatal public sector information infrastructure being nurtured now mostly by public libraries and grassroots community organizations, and the consumer services starting to gain their feet, such as America Online, WebTV, and the any number of imaginary baseball league web servers out there on the Internet. Is it chaos? Of course. Is it driving students back to their textbooks, citizens into the municipal archives to review the minutes of last month’s city council meeting, or 50-year-old men to the microfilm room to calculate how many pop fouls Yogi Berra caught in the first half of the 1957 season? I haven’t been counting, but I don’t think so. The electronic document era is approaching, and the ways in which documents can move around electronically are astoundingly diverse. Yes, electronic watermarking technologies are well along in development. Yes, micropayment systems for electronic publishing ventures are coming. But the concept of a document has been transformed from something precious and rare into something as ubiquitous as a grocery store receipt, and on the ‘Net there is functionally hardly a distinction worth making between the two.

 Let me get more to the point. At our university, which was last month judged the second most wired college campus in America by Yahoo Magazine, faculty and grad student teaching assistants are building web sites with the ease of walking down the hall to the departmental copier, and more cheaply. Our popular campus conferencing system, which offers complete privacy to any course using it and enables document distribution with such ease that a small freshman writing class can share literally dozens of document drafts in a week’s time, enables faculty to set up their own, non-library-mediated, electronic reserve systems with a mouse-click, and thus to distribute any sort of electronic document they choose to effortlessly. No records are kept. No rules, however we might make them, are enforceable. Northwestern’s environment today is Slippery Rock State’s environment tomorrow. It is my view that centrally managed and coordinated document delivery systems such as the electronic reserve systems being put in place in many colleges and universities, hold more promise for financial viability among established publishers than ad hoc, chaotic ones. The longstanding friendly and mutually-supportive relationships between the publishing community and the library community deserve to continue, and I hope that publishers can see advantages in partnering with libraries as electronic document delivery systems evolve. In fact, at Northwestern, it is becoming increasingly recognized that the hierarchical arrangement of e-reserve course materials by school, department, and course offers a well-understood and convenient retrieval model for any number of information objects, and that other academic information services can piggy-back on the electronic reserve web. The department and course hierarchy can be extended to offer access to class software, to electronic class discussions, to electronically-maintained course evaluation surveys, to web-based textbook ordering systems. Such clarification of the geography of information space will be critical to the success of electronic markets in the years ahead.

 While I’m still on this soap-box, I’d like to say a bit about the Conference on Fair Use, which will be covered more thoroughly this afternoon by another speaker. Established as an outgrowth of the Lehman NII White Paper, the Conference on Fair Use, or CONFU, did convene a working group to develop guidelines for electronic reserve, guidelines which ended up NOT being a part of the final CONFU report. We at Northwestern did communicate with this group as outside interested parties, as they solicited input through contact with the e-reserve ACRL discussion group that Jeff Rosedale convened. One change we had a hand in effecting in those guidelines was the dropping of a strict requirement that a password be required to authenticate every e-reserve user, a requirement which we felt entailed considerably more technological effort than was warranted, and would make a very-easily implemented system such as we designed much more difficult. Does anybody in this audience know of a reserve desk which not only requires a student ID for every transaction, but also requires a check that every student requesting an item is enrolled in the class for which that item is assigned? …. I didn’t think so. I was glad that the CONFU group moved to a less restrictive position on that matter, but was very disappointed in the end when it was revealed that the e-reserve group simply could not reach consensus on the fundamentals. Those who stuck with the process to the end – Kenny Crews, Lolly Gasaway, and others who tried to seek balance despite the position taken by some of the publisher representatives – deserve our thanks, and I hope that the e-reserve guidelines in the end are taken up by libraries despite their non-inclusion in the CONFU report. The low acceptance by relevant communities of the CONFU guidelines that WERE published tells me that there were flaws in the whole process, beginning with the appointment of a patent expert to get the process started rather than someone more sensitive to the very different intricacies of copyright. I think that those who stuck it out to complete the e-reserve guidelines should in fact hold the non-inclusion of their work in the final report as a matter of pride.

 I promised at the start of this talk that I would relate matters of electronic reserve and copyright policy to the old library "fee or free" debate of the late 1970’s. How do I see them related? First recall how that debate was left in the literature by the time it exhausted itself around 1980 or so. The clear consensus was that the pragmatists won and the starry-eyed idealists lost. Of the ARL research libraries which offered online literature searching in 1981, only two out of a hundred or so provided searching without charges levied to patrons by my survey at that time. The business orientation dominated, and library services were compromised with the sense that economic interests were all-important.

 Yet where are we now with online searching? Lo and behold, the online services supported by pass-through charges are but a tiny speck compared with the amount of online searching that occurs in all our libraries without charge. First online catalogs, then CD-ROMs, and now the Internet offer far more satisfying results to far more users than Dialog, BRS, and SDC were ever able to in their heyday. Over time, and with the changing economies of computer hardware and cycle costs, database publishers have come to understand that less restrictive, licensing-oriented distribution of their wares is far more appealing to large audiences than a nickel-and-dime approach in which every bibliographic record was charged for.

 It is my hope that the appeal and success of electronic reserve systems may soon reinforce the same message to publishers as we work to bring the digital library closer to reality. An older bean-counting approach to copyright cannot be sustained in the new environment in which copies multiply with the speed of the sorcerer’s apprentice broom and water pail. Fair use is not something that publishers should view with only rank dread; it is a principle that sustains the ecology of our information environment, and should be nurtured. I hope that electronic reserve systems can help sustain the livelihood of publishers who deserve it, and show the way for more publishers to work with educational institutions to design information services which are both economical and valued.