From Free Public Libraries to Pay Per View:
by Robert L. Oakley Prepared for presentation as the Keynote Address at the
Copyright and Access to Information in the Digital Age
First Library Affairs Symposium on Information Issues at Southern Illinois University
April 24, 1998
Good morning. It is a pleasure for me to be here in Carbondale. I have been good friends with any number of the law librarians at Southern Illinois, dating back to Roger Jacobs and Liz Kelly, and now Frank Houdek. Because of the wonderful way they have spoken about the community, I have always felt that this was a place I should visit. So I really am pleased to be here today. I am also honored to have been asked by Jeanne Simon to present the keynote address for this important conference. Her leadership of the National Commission has been vital and important, and this conference is significant and comes at a very important time, because the copyright issues we are discussing are being debated in Congress right now, today.
I was asked to speak specifically about the current issues from the library perspective and to address what they mean for libraries and educators. I will do that, and I am sure that other speakers at this conference will go into even more detail than I can. But to help establish a frame for the rest of this conference, I wanted to do two additional things: provide an overview of copyright fundamentals and also consider the larger impact of these issues.
So, that's where we're going -- or at least that's where I'm going -- in the remaining part of the hour. First a copyright overview, then a review of the current debates, and finally, a discussion of the implications.
I. Why do we have copyright? What are the basics?
Conceptually, the principles on which copyright is based have ancient origins. Under very old principles of natural law, one is entitled to keep and benefit from the product of one's own work. A farmer who clears the land and plants crops can put food on the table. A hunter who killed an animal for dinner could bring it back for his family. Similarly, an artist could sell his paintings, and a writer could sell her poetry. From a very early time, then, there has developed the idea that one can keep and benefit from the fruits of one's own work and creativity.
But interestingly, even in early times, this principle wasn't absolute. There were limits. Ancient Biblical law -- Jewish law -- teaches that when harvesting your crops you should leave a little behind. You don't harvest all the way into the corners; you leave a little at the edges. Instead of taking everything you can, this ancient principle says you leave some behind for others who might need it -- today, we would say that you leave some for the public good. I don't think it's stretching things too far to suggest that this ancient idea of leaving something for the public good is an early forerunner of the copyright doctrine of fair use and some of the other limits built into the Copyright Act that we'll talk about later.
Well then, what is Copyright, anyway? One way to think about it is that it is a bundle of rights that Congress has granted to authors and other creators to provide an incentive for them to create and to allow them to make money and benefit from their work. In the creative environment, it provides a modern legal framework for the older natural law tradition.
But actually, the purpose of copyright is broader than just providing economic benefits to authors. The founders of our country saw a larger social purpose for copyright in the framework of a developing nation, because they understood that for this country to develop and thrive, it was important to stimulate creativity and scientific invention. So underlying the specifics of the law is a system of economic incentives intended to stimulate writing and scientific discovery and the development of America. With that understanding, the framers of the Constitution gave Congress the power to create a system of copyrights and patents. The Constitution says:
Congress shall have Power ... To promote the Progress of Science and useful Arts by securing for limited Times to Authors and Inventors the exclusive right to their respective Writings and Discoveries. (United States Constitution, Article I, Section 8.)
To emphasize: our Constitution says that there is a public purpose to copyright: "to promote the progress of science and useful arts." And in promoting the public interest, it should be expected that creators will be asked to leave a little behind at the corners of the field -- to balance the their own rights with the needs of others who want to learn from or build upon the work of those that have gone before.
With all that as background, what does the Copyright Act actually do. I said before that the Copyright Act gives copyright owners a certain bundle of rights. But what kinds of works are eligible to benefit those rights? Not everything is eligible. The Copyright Act protects original works of authorship from the moment of creation. Protection does not extend to ideas, scientific principles, or works completely lacking in originality, such as the white pages of the phone book. The requirement for there to be some modicum of originality is fundamental. Other than that, almost every original work qualifies.
Now I know you're puzzled. You thought that a work had to be published or something before it could be protected. Well, that used to be the case under the old copyright law, and lots of people think that it's still true, but that was changed with the new Act that was passed in 1976. Now it is said that a work is protected from the moment of creation. OK, but you're still puzzled. Doesn't it have to have a copyright notice on it? Again, that used to be the case, and it's still a good idea, but since we joined the Berne Convention in 1989, it's no longer required.
The end result of this is that so long as there's even a little bit of originality, everything you write is protected, whether or not it's been published and whether or not you have affixed a notice of copyright. That's weird, you think. Even my e-mails? For sure. And my grocery lists? Yup. If they exhibit any originality, say in the selection, they are protected. Do you care? Probably not.
Now we know that lots of stuff is protected. What does that mean? What is in that bundle of rights that I mentioned? This question brings us right into the nitty-gritty of the Act. The Copyright Act gives copyright owners five specific rights. They are:
- the right to reproduce the work in copies,
- the right to prepare derivative works,
- the right to distribute copies to the public by sale, or other transfer of ownership, or by rental lease or lending, [Please note that this distribution right is broad enough to encompanss lending. That would appear to create a problem for libraries if there wasn't something else in the Act for us. I'll come back to that in a minute.]
- the right to perform a work publicly,
- the right to display a work publicly,
Those are the five basic rights given to Copyright owners under the Act.
Of these various rights, the reproduction right and the distribution right are key to the kinds of materials with which librarians and educators frequently deal. Taken together, they allow the Copyright owner to control what happens to their work, because both the making of copies and the distribution of those copies -- at least at the outset -- is something that is done only by authority of the copyright owner. With that in mind, you can understand why copyright owners have seen the photocopy machine as a major threat: it permits the making of unauthorized copies. Even more threatening than the photocopy machine, though, is the global network, because computers on the network can make it possible for someone without authorization to both copy and distribute perfect copies of a protected work to thousands, if not millions of people, almost instantaneously. Such copies, it is important to understand, are copies for purposes of this section, and, in such a quantity they are almost certain to be infringing copies.
In addition to reproduction and distribution, in the public display and performance rights, the word "public" would include a library, and the word "display" would encompass a work displayed on a computer terminal.
Finally, the right to prepare derivative works was originally intended to encompass movies and translations. Now it might include other kinds changes of form or format as well, such as the migration of a work onto a multi-media CD-ROM.
Those, then, are the five exclusive rights of the copyright owner: reproduction and distribution, public performance and display, and the right to make derivative works. But in creating these rights, Congress understood that to promote creativity, it was important to balance the rights of owners with the needs of users -- to leave something behind, if you will, for the public good. As a result, immediately following the enumeration of the rights, the statute sets out a series of limitations on those rights. There are several of these sections, and they are much longer and more complex than the section enumerating the rights of an owner. For now, I just want to focus on three of these sections that are most important to libraries and educators: fair use, first sale, and the library exemptions.
First, fair use. The first and most general statement of a limitation on the rights of copyright owners is fair use. Fair use is a rule of reason originally created by the courts to permit certain activities that seemed reasonable and in the public interest because they advanced the progress of science and art. As now incorporated in the Act, this section says that:
... the fair use of a copyrighted work, including such use by reproduction in copies ... for purposes such as criticism, comment, news reporting, teaching, scholarship, or research, is not an infringement of copyright.
Simply stated, this says that if it's one of the enumerated purposes, or similar to one of them and it's fair, then it's ok. The question, of course, is what is fair. One commentator said that it should be defined as whether a reasonable author would expect remuneration for the use in question. I often say to people to go with your gut. If, in your heart of hearts you think copying that software program to give to a friend is an infringement, then it probably is. If, on the other hand, when you copy an article from a magazine to read on the plane, you think you haven't deprived the author of any income, then probably it's fair. But, of course, the statute doesn't say to the court "go with your gut." Instead, it lists four factors the court is to consider. Those factors are:
- the purpose and character of the use -- is it like one of the purposes listed or not? Generally, the courts like education, but that's not always the end of the story.
- the nature and character of the work -- in other words making multiple copies of workbooks so you only have to buy one is not a good idea.
- the amount and substantiality of the amount used in comparison with the whole.
- and the effect on the marketplace. Although all four criteria are important, this last one -- the effect on the market -- often gets at what the real underlying concern is.
The second limitation on the rights of creators that I want to talk about is the so-called first-sale doctrine. This doctrine provides the legal foundation for lending libraries in America. Remember, I told you earlier that copyright owners had an exclusive right to control the distribution of their work. The first sale doctrine says that despite that right, the owner of a particular copy of a work -- like a library -- may sell or otherwise dispose of possession of that copy without the authority of the copyright owner. So, if you have a copy of you may lend it, sell it, or do whatever else you might want to with that copy.
The last limitation on the rights of copyright owners that I want to highlight is really a series of exemptions that were created expressly for libraries in section 108 of the Act. Especially with the advent of photocopying technology, when the Act was passed in 1976, libraries felt the need for some certainty about what they could and couldn't do. Fair use was important, but because fair use is determined on a case-by-case basis, libraries wanted more assurance about some of their activities. Could they photocopy an occasional article for a faculty member? Could they obtain a photocopy of an article from another library for the use of a faculty member? Could they put vending photocopy machines in their libraries? Could they use photocopy technology to preserve deteriorating works in their collection? These were the kinds of questions that were asked, and subject to some conditions, they were answered in the affirmative.
In general, then, section 108 does provide libraries with a safe haven for certain activities. But it also sets out some conditions. First and foremost, all copying done by the library under section 108 is limited to single copies. 108 does not authorize the making of multiple copies for any purpose, not even for preservation. In addition, the library must not be making the copies for a "commercial purpose." Thus if a library wants to make some money by running a document delivery service, they cannot take refuge in section 108. Finally, all the copies must carry a notice of copyright.
Subject to those constraints, under section 108, libraries are allowed to do a number of things. First, under certain conditions, they are allowed to make single copies in facsimile form -- that is not digital -- of both published and unpublished works for purposes of preservation. This provision was completely non-controversial when it was passed. Unfortunately, as library preservation programs have developed, it is no longer adequate. National preservation standards require the making of three copies, although only one of them is for actual use. Moreover, reproduction technology has developed, and the limitation to facsimile form is now a serious constraint.
Second, under certain circumstances, libraries may make copies for a user for purposes of research or scholarship. They may even obtain those copies from another library. The key here is that these things should not be done in a way that substitutes for a subscription or a purchase.
Third, libraries may provide self-service copy machines, provided they post apropriate warnings about copyright.
All right, that's our thumbnail review of copyright in 15 minutes or less. Before we go on, let's step back again and review the big picture. It should be apparent, by now, even if it wasn't when I started, that the Copyright Act really is a carefully constructed balance. First, the Act sets out the rights of the copyright owner -- you remember, there were five of them:
- reproduction and
- public performance and
- display, and
- the right to make derivative works.
Then, for the sake of the public good, the Act limits those rights. We talked about a few of the limits very quickly including:
- fair use
- first sale
- and the library exemptions, which, in turn, include
- reproduction to support private scholarship or research under certain circumstances
- interlibrary lending
- self-service copy machines
For purposes of giving you a shorthand which may be useful for the rest of this conference, let me digress for just a few seconds to tell you where, in the Act, these things appear. The rights of the Copyright owner are set out in Section 106. That is followed by the fair use limitation in section 107. The library exemptions are in Section 108, and first sale is in Section 109. So, if someone says 107, they're talking about fair use; if they say 108, they're talking about library issues.
- In many ways, this historic balance in copyright seems threatened today. Copyright owners, themselves, feel threatened by the development of digital technology and networks. As a result, they are seeking increased protections. Unfortunately, many of us feel that these increased protections will come at the expense of the public interest.
Without a doubt, the development of digital networks has increased the potential for piracy and changed the way in which information providers make their money. That is rightfully a cause for concern, and we librarians and educators should be interested in maintaining a healthy information industry, because they are the source of the information we use. Unfortunately, however, to many of us, it appears that the changes now being sought by the industry will eliminate the balance and leave little or nothing behind for the public interest. In the next 20 minutes or so, I want to review with you a series of issues now under discussion. They will include a proposal for the extension of the term of copyright, encryption and the proposed new copyright treaty, copyright liability for online service providers, which might include libraries and educational institutions, a proposal to protect databases that wouldn't otherwise qualify for protection under copyright, and a proposal to make the Copyright Act largely irrelevant through the mechanism of licensing. As you might guess, that's a lot to cover in a short time, but if you work to stay with me, I think we can do it.
First, term extension.
One of the most basic limits on the rights of creators that I did not discuss earlier is the length of the copyright term. It started out in 1789 as a fourteen year, renewable term. Over the years, however, it has gotten longer and longer -- first to a 28 year renewable term, and then up to 75 years, total. The 75 year term still applies to most older works, but for new works, the term is now the life of the author plus 50 years. To me, life plus 50 already goes well beyond what is needed for an incentive to create -- well beyond being a reasonable period to exploit a work -- and well beyond being a limit on the rights of creators. Now it is really just a limit on the ability of the heirs to benefit from the work of their ancestor.
Despite the existence of Copyright terms that extend well beyond the life of the creator, a bill was introduced in the last Congress to extend the term of copyright by an additional 20 years. For new works, such a change would result in a term of the life of the author plus 70 years. For many older works not yet in the public domain, the new maximum would be 95 years, and works published in the 1920's and 30's would not come into the public domain for an additional 20 years. In effect, this proposal would create a 20 year moratorium on any new works entering the public domain.
The stated reason behind term extension is "international harmonization." A few European Countries have begun to migrate to the longer term, and it is argued that without a similarly longer term in the United States, our authors will be disadvantaged. Perhaps more important for U.S. political purposes is the fact that the film industry and the music industry are both strong supporters of the extension because without it some valuable properties from the 1920's will soon enter the public domain. Specifically, the Disney Company and the Gershwin estate are both working hard to protect their interests. Many wonderful Gershwin tunes were written in the 20's, and even the venerable Mickey Mouse will lose his copyright protection in a few years if nothing is done. With such powerful industries behind the proposal, it seemed clear at the outset that the bill would ultimately succeed.
From the library perspective, however, the longer term presents a number of problems. In general, the extension could interfere with a library's efforts to serve its clients, even when a work was old and out of print, and when the copying did not interfere with any identifiable market. Moreover, the issue is closely linked to the preservation problem. Because of the uncertainties of copyright, many librarians working to preserve their collections choose to follow the secure path of preserving only those materials that are already in the public domain, i.e., published before 1923. A twenty year moratorium on new works entering the public domain will represent a twenty year setback to the library preservation effort.
With these issues in mind the library community sought an exemption from the extension so long as the library was not doing the copying for commercial purposes and so long as the copying did not interfere with the market for the copyrighted work." 
Extensive discussions were held throughout 1996 on term extension generally, and on the library proposal in particular. Copyright owners were concerned about the open-ended nature of the potential uses in the library, and they also wanted stronger language to protect their markets. By the end of the Congress, compromise language was developed by the Register of Copyrights that would allow libraries to make copies of protected works during that last 20 year period so long as the work was not currently being commerically exploited by the owner and it could not be obtained at a reasonable price.
This approach seemed to be a workable compromise. At the very end of the legislative session, however, the publishers' representative insisted that "normal commercial exploitation" would include works made available through a licensing. Such a provision seemed to the library community to make the whole proposal a nullity because anything could then be included on a list of works available by license and negate the whole purpose of the proposal. The discussions foundered at the end of the Congress on this issue. The bill was re-introduced in the 105th Congress, and the Library Community attempted to re-engage the parties on the issue. It proved impossible to re-ignite those discussions, however, so the library community felt reluctantly that it now had to oppose term extension altogether.
Encryption is not really a legal issue -- or at least no one has made it one so far -- so in raising the issue here, it is in large measure as a prelude to the discussion about copyright, the internet, and the treaty proposal now before Congress.
I have long considered encryption to be the copypright owner's trump card because it gives them total control. Forget the corners of the garden. With encryption, they'll leave the corners if they want, but no law -- other than the Biblical admonition -- will direct them to.
What is encryption? Essentially it is a technological means of giving the owner control over his or her work even if it is no longer in her possession. One of the easiest ways to think about it is to think of it as an electronic envelope. The copyright owner puts the work into the envelope, seals it, and then writes on the outside the conditions for opening and reading what's inside. It might say, for example, if you want to read this article, it will cost you $2.95. No forwarding, printing or downloading is permitted. If you wish to proceed, please enter your VISA or Mastercard number now. If you authorize payment, you will be permitted to read the article; if you don't, the envelope will remain sealed. This kind of system is called "pay-per-view," and it has been the goal of the publishing industry for a long time. The use of encryption technology makes such a system possible.
Third, the proposed copyright treaty and its implementing legislation.
Currently in Congress, already passed by the House, and perhaps about to be acted upon by the Senate, is legislation to approve and implement a draft treaty. In the United States, this legislation started out in 1995 with a report from the Working Group on Intellectual Property, which had been convened to make recommendations about the intellectual property issues involved in developing the Internet. The proposals of the Woking Group clarified the rights of copyright owners by adding the word "transmission" to the list of exclusive rights of copyright owners. They also endorsed encryption and licensing as a solution to the piracy problem, and, to permit the encryption system to work, the bill proposed to make devices that could circumvent encryption systems illegal.
The library community and many in industry had some major concerns with the proposal, and made those views known on Capitol Hill and elsewhere. The seeming expansion of the rights of copyright owners to include "transmissions" raised many questions as did the reliance on technological solutions. Among other things, the library groups asked that if Copyright owners were getting the extra protection they felt they needed, then at a minimum, there should also be an addition to the fair use section of the statute to ensure that it would apply in the electronic environment and not be limited to the analog technology of the time when it created. Although many people assured us that the vitality of fair use would continue, they were not willing to put that simple proposition in writing, in the statute. And, by the way, they still aren't.
Ultimately, the Working Group proposal did not pass because so many concerns were raised about it. However, the proponents of the legislation took many of the same issues to Geneva where they were able to secure a draft treaty. In something of a victory for the public interest, the Library groups who were in Geneva won a concession that nations could provide for limitations on the exclusive rights otherwise granted to copyright owners. That was a victory, because even though such limitations have been a part of United States law for a long time, the concept had not been recognized before in international treaty.
The treaty does a number of things, but two of them seem important here. Although it did not establish a transmission right, it did establish a "right of communication" which grants:
authors ... the exclusive right of authorizing any communication to the public of their works, by wire or wireless means, including the making available to the public of their works in such a way that members of the public may access these works from a place and at a time individually chosen by them.
Plainly, that is intended to encompass the placement of protected materials on the net. Although it sounds like a new right, experts think it can be inferred from our existing rights of reproduction and distribution, so that language is not included in the implementing legislation. Make no mistake, however, since a treaty is the supreme law of the land, when the treaty is adopted, the communication right will very much be a part of U.S. law.
Unfortunately, the Library community and many of its coalition partners have some of the same concerns about this legislation that they had about the earlier one, so I thought I should go through those.
The first concern is a heavy reliance on encryption and the barring of anti-circumvention technology. The treaty supports the use of encryption systems and requires countries to provide legal protection against its circumvention. The WIPO implementation bill proposed by the administration, and currently moving forward in Congress, deals with this issue by outlawing the development of technology that could be used to circumvent the system of encryption, even if the circumvention would be used for a lawful purpose, such as fair use. For the reasons I've indicated, many of us have doubts about encryption and believe that the unfettered use of encryption will lead to an imbalance in the system as we have known it to be. But even if you do accept the necessity for encryption, it seems wrong to protect it by outlawing the development of new technologies.
Consider this: when the photocopier was invented, many people thought that it's primary purpose was to defeat copyright. When the VCR was invented, so many people thought that infringing copyright was its primary purpose that a lawsuit was brought against SONY, the company that made it. We cannot know what new reproductive technologies will emerge in the future. But if this bill is passed, the development of such new technologies will come to a halt, and we will never see the next generation of reproductive technology.
The library community and its coalition partners in industry believe that we can comply with the treaty without interfering with the development of new technology by focusing -- as has traditionally been done in copyright -- on the intent of the person using the technology. The treaty does not require a broad prohibition of software and hardware. The same goal can be achieved by holding the person liable who, for purposes of infringement, knowingly circumvents the operation of an encryption system. Such an approach would allow technology to continue to develop and place the blame where it belongs: on the person committing the infringing act.
In addition to this concern about anti-circumvention, the library community has been concerned about a number of things that are not in the proposed legislation, but should be if the Copyright Act is being updated for the digital age.
Fair use. We would like to see the fair use section of the Act amended slightly to ensure that a finding of "fair use" may be made without regard to the means by which a work has been performed, displayed, or distributed. In that way, just as teachers, librarians, researchers, and others may make fair use of copyrighted works in the analog world, they would be assured that they could do the same thing in the digital world.
Library Preservation. The preservation problem has reached critical proportions as brittle books continue to deteriorate on our shelves. This problem will get even worse if the term extension bill is passed. In the late 20th century, librarians should be permitted to use digital techniques, not merely analog techniques, to be able to preserve the materials in their collections.
Distance learning. The current Copyright Act supports distance learning by permitting the distribution of works for such purposes using television. The Copyright Act should be updated to permit the same thing to be done digitally. Failure to make this change could hamper the development of distance learning which has become increasingly important everywhere, but especially in rural areas.
First Sale. Like distance learning, the first sale doctrine needs to be updated for the digital age. We believe that it is not unreasonable for a person who has a legally obtained digital copy of a work to transmit it to someone else, so long as the original is deleted from the sender's system at substantially the same time.
Ephemeral copies. When a work is transmitted over a network, many temporary or ephemeral copies are made along the way. We believe that for the web to thrive, it should be made clear that such ephemeral copies, in and of themselves, are not infringing copies.
Fair use, library preservation, distance learning, ephemeral copies, first sale. None of these are radical; certainly they do not "gut" the bill, as one advocate has written; all they do is attempt to carry forward the existing balance of rights into the digital age. These ideas have been incorporated into a counter proposal to the Adminstration bill that has been put forward by Senator Ashcroft and by Representatives Boucher and Campbell. Unfortunately, as reasonable as they seem to us, and although the House bills have attracted a number of co-sponsors, these issues have not yet been dealt with.
The fourth issue I wanted to address is the question of liability for online service providers.
Let's say you are running a library, and you have a server connected to the network to provide access to resouces on the net. Or perhaps you are a system administrator for your campus. Now, let's suppose a student uses a computer connected to your network to access a web-site that contains infringing material. Should you be liable for that? What if the library provided a pointer to that site? But what if when the pointer was put up six weeks ago, the site contained no infringing material and it's changed in the meantime? What if a faculty member decides to post reading materials for his course on the network instead of on reserve in the library? What if it turns out that some of those readings are infringing? What if the student puts infringing material on the University's computer and sends out an e-mail to all his friends to "come and get it?"
At what point in this rather large range of transactions, should the University or the network service provider be liable? That has been the subject of discussions for a couple of years now in Washington. It is an issue on which the industry is very concerned. Some companies, like America Online, find themselves very threatened if they might be liable for the transgressions of their millions of members, over which they have no control. On the other hand, the content community -- especially the music industry is very concerned that illicit activities may be taking place on AOL or other ISP computers and wants them to take whatever steps they can to stop it.
At one end of this range of activities is something that has come to be called the "mere conduit." That is the situation when an internet service provider acts like the phone company, simply providing wires and switches for the transmission of information, but having nothing to do with the content. At the other end of the spectrum is when the provider creates a site themselves and makes it available to their customers. At each end of the range, there seems to be a large measure of agreement. If someone creates a site with infringing material on it, they should not be able to avoid liability. On the other hand, if they are a mere conduit, with no involvement in the content, then most people agree it is equally clear there should be no liability, especially if when notified that there is an infringing site on their system, they take prompt action to remove it.
But in between those extremes, there seems to be little consensus. What if aol hires someone to create a site for them? Probably liable. What if they didn't create the site, but merely highlight it on their opening screen for other people to know about? What if a library provides a hyperlink to a site? Is that different from highlighting? What if that site was non-infringing one day and infringing the next? If a copyright owner notifies you that you have infringing material on your site, what actions do you need to take to avoid liability? Somewhere in between the extremes is the line of liability, and for the sake of certainty and security, the industry would very much like to see that line defined.
Early in this Congress a bill was introduced to move these issues along. Unfortunately, all it did was to grant an exemption for the mere conduit, leaving for another day -- and probably a court -- the resolution of more of the intermediate issues. More recently, there has been discussion of adding to that an exception for simply providing links, as many libraries do, and also for providing space for an individual to create their own site over which the provider has no control. In both cases, the exception would not apply if the provider knew about infringing activity, and if notified of the infringement, they must take appropriate steps to remove the material. These changes could go a long way to giving internet service providers the security they need. They don't, however, deal with all the issues for the academic community which must consider academic freedom and fair use in the educational environment. The discussions on those issues are still underway, but the online service provider bill is likely to get folded into the treaty implementation bill and be passed together, sometime this spring.
Now I need to turn to something a little different -- database protection, which is the fifth out of six issues on my list.
The database issue is different because, strictly speaking, it is not about copyright. On the other hand, it arises as an issue because some proprietors believe that the Copyright Act does not protect them in the way that they would like to be protected.
The story begins in 1991 when the Supreme Court decided the Feist case. That case held that the White pages of the phone book were not copyrightable because they lacked sufficient originality. Noting that originality is the bedrock principle of copyright, the Court in Feist explicitly overruled a line of cases that had allowed copyright protection based on the so-called "sweat of the brow" rationale, where the basis for protection had been hard work and investment, not originality. The Court said that to obtain copyright protection some modicum of originality must be found, if only in the selection, coordination, or arrangement of the material. Mere investment or hard work was not enough.
The case dealt only with the white pages, but the holding was broad enough to call into question the copyrightability of many databases, especially those created automatically or from materials in the public domain, unless there is some demonstrative selection, coordination, or arrangement of the material in question.
About a year after Feist , the Commission of the European Union presented a proposal to the Council for a Directive on the protection of databases outside the traditional systems of copyright. Over the next few years, that proposal progressed through two readings, and eventually it was finalized as a Directive and transmitted to the member states for implementation.
Two months later, in May of 1995, a substantially similar proposal was introduced into our Congress as H.R. 3531 by Representative Moorhead. It was supported by the Information Industries Association, but aroused concern among many other groups -- including many in the industry -- who were concerned about the breadth of the bill and its potential to significantly undo the copyright balance in the United States. Let me highlight four key provisions of this new bill for you. First, it would have overruled Feist, permitting protection based on investment rather than originality. Second, the definition of database was so broad that it could have included essentially anything, even literary works. Third, it would even have protected collections of public domain government information. Fourth, protection for the database would last indefinitely, as long as the database continued to be updated or otherwise modified. In addition, the proposal made no provision for fair use or other similar uses of the data. This bill split the information industry and caused alarm in the library community and among scientific and medical researchers who felt that their reliance on free public domain data was threatened, and it was never passed.
Recognizing the flaws in this approach, the proponents of database protection returned with a different approach in the current Congress. Instead of trying to protect data under copyright, the current bill would protect it against misappropriation, or taking the data in a way that interferes with the compiler's market. At first, this approach seemed more promising.
Nonetheless, there remain a number of issues. Most importantly, there is nothing in the record that suggests that additional protection is needed for the protection of databases. No studies have been done, and from all that anyone can see, the database industry is thriving, with companies like Thomson and Reed Elsevier purchasing database companies for record prices. As a result, there is no indication of a need to protect investment, rather than creativity. Moreover, the information industry itself remains split on the desirability of this proposal, with the legal publishers being the chief proponents and other major companies like Dun and Bradstreet opposing it.
In addition, some of the key definitions are troublesome. For example the definition of information is very broad, as it was in the earlier bill. The bill defines "information" as:
facts, data, works of authorship, or any other intangible material capable of being collected and organized in a systematic way.
That sounds to me as though it is broad enough to include pretty much anything that is copyrightable as well as facts and data that are not. As a result, this bill could supplant copyright in many cases. It prohibits someone from "extracting, or using in commerce, all or a substantial part of a collection of information gathered, organized, or maintained by another person through the investment of substantial monetary or other resources, so as to harm that other person's actual or potential market"
Again here, the criterion for protection is investment, not creativity. Moreover there is no definition of harm, nor is there any explanation of the meaning of the phrase "potential market." How is that to be determined? If a library downloads an item, couldn't it be argued that that interferes with the market?
No governmental entity at any level can protect its information under this bill, nor can a private contractor under an exclusive arrangement with the government, so one might think that government information cannot be protected. However, a smart company that simply goes out and collects the information before it disappears from the government source can, through this legislation, gain a monopoly over government information that would be in the public domain under copyright.
Responding to criticism of the original bill that there was no "fair use" kind of provision, the drafters did include a section that permits the use of an "insubstantial" part of the database, but without defining "insubstantial". They also included a provision for not for profit educational, scientific or research uses. But again, if the market includes pay-per-view transactions, then any use could be argued to interfere with the potential market, so these exceptions do not really help the library and education community.
The database bill remains troublesome. It sweeps much too broadly, threatens the principle of free public access to government information, and does not include meaningful exceptions for the library and education community.
Finally, the sixth issue I want to highlight for your consideration is the issue of licensing.
If encryption was one trump card for information providers, the issue of licensing is another complimentary one. Remember the description of encryption as an envelope with a lot of terms and conditions written on it. Well, maybe not all the terms are there on the outside of the envelope, but some of them are inside, after you open the envelope, or in another envelope clipped to the first. How many of you have installed software or downloaded something from the web where there was a license agreement to read and a button to click saying "I agree" or "I accept?" Leave your hands up, please. For those of you that have done that, leave your hand up if you always actually read what you're agreeing to. The licensing proposals moving forward now validates those agreements, even though they are rarely read and they've been very controversial.
These license agreements are commonly known as shrink wrap licenses, because they often accompany software, and frequently you can only read them after you open the package and find the shrink-wrapped envelope with the software inside. Or maybe you don't even see it until you're in the middle of the install program. In the new parlance, these licenses are now called "mass-market licenses," to include clickable licenses as part of the software or on the web.
When licenses such as these have been reviewed by courts, they have frequently been held to be invalid, because they were not negotiated between the parties, and they cannot reflect a meeting of the minds of the parties, because there was often no opportunity and even more often no inclination to read it until there's a problem.
One court -- the U.S. Court of Appeals for the Seventh Circuit -- did find such a license valid in a case called Pro-CD v. Zeidenberg. Now, following the lead of that court, a prominent legal group is drafting an amendment to the Uniform Commercial Code that would also validate such contracts.
The Uniform Commercial Code has been adopted in all 50 states and the District of Columbia as the basic law governing commercial transactions. Until now, it only covered contracts for the sales of goods -- like cars, or produce, or a load of steel. Several years ago, however, some people began to feel that since this was the information age, our basic commercial law should be updated to include licenses for information as well as contracts for the sales of goods. UCC Article 2 is the existing section, and the proposed new section would be Article 2B.
The drafting process has been underway for several years, with a panel of experts doing the drafting, subject to the review of two different national organizations. If those two groups approve the draft, then it will be sent to the 50 state legislatures for their enactment into state law.
The drafting meetings have been dominated by representatives of the information industry, representatives of the software industry, and so on. Few consumer groups have been represented there, and those that have tried to make an impact have run into a brick wall. (I speak from experience.) Because of the lack of consumer input, Consumers' Union has called for the project to be delayed until more of their issues can be dealt with, and Ralph Nader recently criticized the effort, saying that consumers are (excuse my French) "about to get screwed."
Central to the concern of libraries and other consumer groups is the validation of mass market licenses. The reason this is important from a copyright perspective is because by means of a mass market license, a licensor can get you to waive any or all of the rights you might have had under the Copyright Act. That includes fair use, first sale, or any other right you might wish to make use of. All that, without your even being aware of what you've done. You may also agree to other conditions or waivers of warranties that you would be unaware of. The library community agrees that for the internet to develop as a commercial medium, there needs to be some way -- such as through a click-agreement -- for individuals to manifest assent to a contract. But we do not agree that people should give up important rights without even knowing that they are doing so, let alone having had the opportunity to negotiate over it. At a minimum, the library community believes that when a consumer is asked to give up any of their rights under the Copyright Act, that fact should be brought to their attention and they should have to manifest assent to that specific provision.
Like the database bill, the definition of information in the draft U.C.C. is broad enough to include everything: facts, databases, literary works, everything. So now you can see why this bill is a second trump card against the consumer. First everything is locked up by means of an encryption system which it is a criminal offense to defeat even for a legitimate fair use purpose, and then to see what's inside the sealed envelope, you may be asked to waive your rights.
* * *
Well, now we have two very different pictures in front of us. First, I gave you an outline of some of the key provisions of the current Copyright Act. I hope from that that you got a good sense that, imperfect though it may be, the Copyright Act really does make a good effort to balance the different interests -- to set up a system by which a creator can obtain a reasonable reward from her work for a reasonable period of time and at the same time provide some assurance to researchers and other users of information that they can make reasonable uses of the information they find to further the progress of science and the useful arts.
Unfortunately, the picture of the developments now in progress looks very different. Copyright owners want a longer terms of copyright, to extend even longer beyond their own lifetime than the law already provides. They want to protect things that are not protectible under copyright, things that are completely lacking in originality. They want to use encryption technology to guarantee that they are in total control and do not even have to permit fair uses without payment, and they want to use non-negotiated licenses to ensure your agreement with whatever terms they choose to impose. Taken altogether, the picture for the public interest is not a pretty one.
- In what remains of my time this morning, I will speculate a little about the reasons why we seem to be headed in this direction and comment on the important societal values that I think will be lost if these initiatives succeed without some modification.
Despite the picture I have painted, I believe that owners of intellectual property do have a worry, and I believe that it is a legitimate worry. I understand the publishers' concern. They are fearful that as they begin to make their material available over the web, it will be possible for people to copy and redistribute their works at will and without remuneration. No doubt, there is good reason for this worry. Copyright piracy is a serious global problem. That is why copyright owners want to establish absolute control over their works. But American students and researchers are not all engaged in wholesale piracy, and I am afraid that if we provide the total protection now being sought by the industry, something important will be lost. I believe that in designing a solution to the problem, it is important to preserve what we have in the current system that is positive and beneficial to society.
You know, some people think that librarians are just out to get publishers. Not too long ago, I was attacked by a publishers' representative who said "Librarians just want to get everything for free." Nothing could be further from the truth. My library has a $7 million budget, and I spend nearly half of that buying books and acquiring other materials. In the aggregate, librarians spend a lot of money to support the publishing industry, and they understand the importance of maintaing a healthy industry.
So, contrary what some people apparently think about me, I believe that proposed solutions to the problem of copyright on the web should accommodate both groups -- meeting the needs of scholars and the needs of copyright owners as well. Publishers should have a right to appropriate remuneration for the use of their materials on the web. But librarians, researchers, and students in our schools should also have the ability to make certain limited uses of material on the web without each time having to ask for permission or pay for the privilege.
Let me give you an example. Not very long ago I was using the Washington Post on the Web. It is an excellent site, supported by advertising, and it provides a valuable public service by being available free for two weeks after initial publication. The Post has also made a backfile available. Searching on the backfile is free. But if you want to actually read any of the articles you find in your search, it will cost you $2.95 each. This is the system of pay-per-view and is the wave of the future if the current proposals in Congress succeed. Everytime you want to read something, no matter how short or how long, you will have to pay. And if you think about it, the costs will mount up very quickly. I have a twelve year old son. Let's suppose he is doing a school paper on, let's say, civil rights issues in the Washington, D.C. area. At $2.95 per hit, is there any reasonable way he could do that research? Not with my credit cards, he couldn't.
So what is wrong with this picture? How is it different from what we have today?
Libraries in our society embody important values, values that are very different from the pay-per-view model toward which we seem to be headed. First of all, libraries are places where a very wide range of material is available. That includes public domain information which is readily available on the web already and a broad range of commercially available material as well -- much broader than what the web currently provides. Second, libraries are generally open. A library user who is a member of the community that created the library does not have to pay to walk in the door. Moreover, a library user does not have to ask for permission or pay for each book they take off the shelves or for each article they might want to read or skim. It is not, of course, that libraries are free; they are not. Nor is it that libraries do not pay for the commercial materials they make available. Rather, it is that the library is a shared resource, and the end user does not have to pay on a transactional basis.
A few years ago, Professor Henry Perritt published an article on copyright management. In that article, he wrote:
One of the many profound observations by Ithiel de Sola Pool was that copyright always has depended upon technological bottlenecks for its enforceability. The printing press was the original enforcement bottleneck .... As technologies change, old bottlenecks disappear and enforceability requires a search for new bottlenecks. .... The problem with distributed publishing on an open architecture internet is that there is no bottleneck ... If new bottlenecks are to be found, they almost surely will be found at the origin and at the point of consumption. Encryption and decryption techniques ... concentrate on those bottlenecks as points of control. .... In such an architecture, the law either will be relatively unimportant because technology can be counted on to prevent free riding, or the law will need to focus not on prohibiting copying or use without permission, but on preventing circumvention of the technological protections.
I believe that much of the potential of the Net will be unrealized and unrealizable if we focus on the creation of a complex system of charges and bottlenecks. One does not need a Ph.D. in fluid dynamics to understand that a bottleneck slows down the flow. And it seems counter-intuitive to me that we should be building a high speed network only to layer on top of it a system of bottlenecks that will slow down the flow. I believe that we need to avoid thinking of information users as potential free- riders. Instead, I would prefer to design a system that provides appropriate compensation to copyright owners but then, to promote the progress of science and the arts, as Justice O'Connor said: "encourages other to build freely upon the ideas and information."
Some works can easily be distributed with no bottlenecks and at no cost. Among the materials that can be readily available without any copyright issue are non-copyrightable works, such as works from the U.S. government, works in the public domain because their term of copyright has expired, and works for which the author does not wish to exercise his or her copyright.
Just as certain works should be available without a fee, so too certain uses of existing works should be permitted. In particular, "fair use," however that is defined should be available regardless of the medium. It is true that we cannot know now how the contours of fair use might evolve in the electronic environment, but as an equitable rule of reason, it should continue to be available to researchers, educators, and others making use of the information. Fair use is the means by which researchers learn, and in turn write for the benefit of society. It is the fuel that drives the engine of progress envisioned by Justice O'Connor in the Feist case.
As I have thought about what was important about fair use and the library model we have known, I realized that the key element was the absence of transaction-based fees. As we all know, any user fee will act as a barrier to use. The greater the fee, the greater the barrier. So if our goal is to limit the use of information on the net, we will design a system that imposes fees on individual users. Those fees will be large enough to be noticed and will be immediately recoverable regardless of fair use.
Contrast this approach with libraries as we have known them. Libraries are well-used and socially important not because they are free, which they are not, but because the individual user does not have to pay for their usage on a transactional basis. As a shared resource, users can take advantage of as much or as little of the service as they desire and there is no difference in the cost to them.
So I wondered if there were other models that might point us in a direction that could provide a source of revenue for the benefit of copyright owners, while avoiding, to the maximum extent possible, transactions fees for individual users. Libraries, of course, came to mind, where the community or the university, or some other organization establishes a budget and then all members of that community are free to use the resources available in the library.
Interestingly, the Internet, itself, also came to mind. One reason that has often been cited for the explosion of use of the Internet on university campuses is the fact that the costs are not borne by individual users, nor even by individual departments in most cases. Rather the costs are typically borne centrally. Moreover, the fee is fixed and independent of individual transactions.
In my mind, these models have suggested a new direction. (I must note here that I am not speaking for A.A.L.L.; these ideas are purely my own.) These models suggest that everyone could benefit if we could find a way to cover the necessary royalty payments to copyright owners by creating a pool of revenue from a higher level than the individual end user or on some other basis independent of individual transactions. Such an approach would recognize the importance of individual use of networked information while avoiding putting up barriers to that use. It would also recognize the need for compenation to copyright owners. But by removing the barriers to individuals, it would stimulate a higher level of usage, while at the same time providing a source of revenue for copyright owners.
Under this kind of system, the various metering systems used by encryption technologies could be useful, not for charging users, but for allocating available revenues to individual owners. All copyright owners, including individual owners, as well as publishers, would be compensated with a portion of the central fund, based on usage of their work.
With this kind of system in place, I believe that we would preserve some of the traditional values of libraries in the new information environment. End users would benefit because they would not have to worry about individual transaction fees. Libraries would benefit because it would allow them to provide access to information in the new formats without having to worry about tranaction fees that would be an uncontrollable cost item in their budget. And copyright owners would benefit because the system would provide them with an assured source of revenue and maximum use of their product.
None of this is to say that this idea would be simple to carry out. There are lots of questions that would have to be answered. But it is a concept that, if it could be accepted by all the parties, could make "the copyright problem" go away. It would also preserve the library values about access to information that have served us so well for so long and which have so effectively promoted the "progress of science and the useful arts."
With all due respect, I believe that Congress is on the wrong track as it heads toward giving proprietors nearly total control over their works. I believe that creative solutions are possible that will provide appropriate rewards to creators while also preserving the kind of societal values for which libraries have always stood.
It may well be too late but I would like to call upon Congress to avoid such narrow thinking and to try to devise a system where the tillers of the soil will reap an appropriate reward, but where the edges and the corners will be left for the benefit of others.
Thank you for your patience. It was a pleasure to be here today.
- Copyright 1998 by Robert L. Oakley
- Robert Oakley is Director of the Law Library and Professor of Law at the Georgetown University Law Center. He also represents the American Association of Law Libraries on information policy matters in Washington, D.C.
- H.R. 989, 104th Congress.
- Proposed "Library Exemption" to H.R. 989, September 19, 1995.
- Intellectual Property and the National Information Infrastructure, the Report of the Working Group on Intellectual Property Rights, Washington, D.C., Information Infrastructure Task Force, September 1995. See preliminary analysis of the white paper by James M. Sabina in the 1995 Bowker Annual at 248.
- H.R. 2441 and S. 1284.
- WIPO Copyright Treaty, Adopted by the Diplomatic Conference on Certain Copyright and Neighboring Rights Questions, Geneva, December 20, 1996.
- Feist v. Rural Telephone Co., Inc.