New Legal Approaches in the Private Sector
In this presentation I hope to shed light on the current state of the public domain, especially with respect to technical and scientific data, and then describe several approaches that tend to eschew the public domain in its legal sense in favor of a rights regime that more subtly allocates power between author and downstream users.
When we talk about access and the public domain generally, what do we mean? It might be useful to discuss them in light of scientific and technical data that have occupied a large part of this symposium.
THE PROPRIETARY, THE PUBLIC DOMAIN, AND THE SPACE BETWEEN
For much scientific and technical data, typically one cannot assert a copyright—at least within the United States. These are the type of data for which there is not enough creative work in the expression to merit copyright protection. As a result, those who want to protect (or, depending on one’s point of view, hoard) these data are left to other devices, such as secrecy—sharing the data with some, but not the rest of us—and contract. In this case, when the proprietor chooses to let others access the data, they impose extra “private” law, created in the transaction between publisher and consumer.
Contract, however, is always limited by “privity.” I might have an agreement with you that you promise not to further share my data, but then once you do and the person with whom you have shared the data further shares them, it is no longer easy for me to limit consumption. That is, I might have an actionable disagreement with the person who violated the contract with me, but not usually with consumers downstream.
Furthermore, there is digital rights management, which Julie Cohen has addressed.1 Even if a work is not protected legally, one may simply “privicate” it—that is, publish it far and wide, but publish it in a manner, thanks to technology, that makes it hard for people to do with it what they will, simply because their respective computers will not let them.
Finally, tied to technologies of digital rights management are laws concerning circumvention of those technologies. If a person attempts to figure out how to do something that their computer will not permit, and if they then seek to apply or share that knowledge, in many instances they could go to jail.
See Chapter 15 of these Proceedings, “The Challenge of Digital Rights Management Technologies,” by Julie Cohen.
In the case of creative work, one can have all the protections previously mentioned for the scientific and technical data, with the addition of copyright protection. Because copyright is the default rule for creative works, copyright holders avoid the privity problem that occurs with contract.
How does this work? If I create a work, I can assert copyright in it, provided that it is creative enough that I can. I “give it to you”—I license a copy to you or you pay me to own a copy of the work—but if you further make copies and the people downstream from you make unauthorized, unprivileged copies, I can go after all of you as a matter of law. In fact, since 1976, I do not even have to put the copyright mark in a symbol on my creative work to have the copyright attach.
So if you are surfing the Internet and encounter a wonderful haiku and there is no copyright symbol on it, it does not mean that the work is not copyrighted. There are plenty of people—lawyers usually—who get up at lecterns like this one and sow fear, uncertainty, and doubt, warning that works published online are in all probability copyrighted, so if in doubt, do not do anything with the work.
If this is the regime we have for these two types of data, what kinds of material do we have in the public domain today? We have work created by nonhoarders, those “crazy” people who give their work away for reasons that in the last session were explained to be actually quite rational. That sort of work can become part of the public domain either by choice or by patron encouragement or even requirement. For example, if a researcher receives a U.S. government grant, the terms of the grant might require that the researcher share the data and let others make derivative works from them. Yet often the patron in these types of arrangements is a university—and as others have discussed during this symposium, universities today are torn between whether they want to be dot edu or dot com. While they are trying to figure that out, universities may not be the ones to rely upon to encourage or require materials to be shared freely; indeed, they may have the opposite agenda.
So that is how the system, generally, is working. For someone wishing to release into the public domain scientific and technical data for which one cannot assert copyright in the first instance, that person need simply fail to take the previously discussed steps to protect that type of data—that is, fail to keep the data secret, fail to write a contract, and fail to create and apply a digital rights management system.
But if the work at issue is a copyrightable work, under the current system one has to take certain steps to disclaim it. A person can choose to distribute this type of material to others, either formally or informally; however, they may not know the legal steps necessary to enter it into the public domain. Instead, it is more often the case that the person chooses not to enforce the rights that are legally retained and that other people come to know that.
A final way that creative material enters the public domain is through copyright expiration. Copyright is for a limited time, which at this moment for corporate-created works in the United States is 95 years. Of course, these copyrights have been extended retroactively repeatedly, rendering such entry mostly theoretical—unless the Supreme Court holds for the plaintiffs (for whom I am co-counsel) in Eldred v. Ashcroft, a case currently pending before the court.2
We also have the informal, de facto public domain, brought about by the existence of photocopiers, personal computers, and the Internet. Even with a set of “background” rights reserved to the author in any creative work, the fact is that most published works—whether or not they are published online—are largely available for use. We have had a culture that permits a certain amount of copying for personal use, and many activities that would count as legalistic violations of copyright are neither frowned upon nor fought against by copyright holders, much less the general public.
The situation as it stands, then, is in flux. The large-scale publishers who usually benefit from some level of copying and sharing are now well aware that photocopiers, personal computers, and the Internet exist—and that in their current incarnations, they represent a threat to prevailing business models. These publishers are unhappy, they are litigious, and they are hiring good coders to write digital rights management systems.3 They have a certain zealous righteousness to their position and freely use the language of theft to describe what is going on when, for
See Eldred v. Ashcroft 537 U.S. ___ (2003). The Supreme Court decided in favor of the defendant. For additional information, see http://www.supremecourtus.gov/opinions/02pdf/01-618.pdf.
example, consumers use peer-to-peer networks to transfer copyrighted music files among themselves. It seems that they hope that their indignation—arguably borne out of a newly expansionist cultural view of copyright—will be adopted by future generations.
An acknowledgment of the ever-widening gulf between the attitudes of intellectual property producers and consumers is evident in a recent report on the creative industries published by the British Patent Office.4 The report sought to consider ways of improving “the public’s perception of the need for intellectual property and its relevance to so much of what we do in our jobs, at home, at leisure and in education.”5 The report states: “There is general acceptance that terminology used by IP practitioners as a result of legislative authority is cumbersome and not user friendly. An important step in achieving greater understanding and acceptance of licensing and the value of intellectual property is therefore finding an alternative way of referring to intellectual property.” (The British Patent Office evidently has no favored replacement for the phrase, suggesting only that the matter be placed under further consideration.) Another recommendation in the report is that “school children should recognize their own creativity by including the copyright symbol on their course work.”6
This reveals the cultural battle between society looking at the intellectual property regime as something that is merely holding them back from what they reasonably want and by all rights could do (before the sleeping giants awoke) and society viewing the intellectual property regime as a useful instrument with which to protect their intellectual fruits from misappropriation.
Attempts to manipulate cultural views aside, the current situation remains that there are two separate baselines for making use of others’ work. For informal use—the kind of use that I have made by including others’ clip art in my accompanying Powerpoint presentation—our consciences are basically the limit. For formal use, however, we turn to lawyers. If I wanted to take this presentation and publish it as part of a book, I would have to obtain copyright permissions. I am working on an Internet law casebook right now with four coauthors. Putting this book together, I am obliged to send out clearance letters for every fragment of others’ work that I want to incorporate. It is a formal publishing enterprise; in a formal situation, with a company that represents a viable target for legal action, all of the defaults are reversed. Unless we literally get clearance, I cannot include the fragment in my book. Yet, again, if the book were simply a presentation for my students, all the defaults would flip back.
It is in light of these tensions between producers’ and consumers’ perceptions of IP and between the different standards for using other’s work that we turn to approaches to promoting the public domain and open access to public domain or near-public domain materials.
NEW APPROACHES FOR PROMOTING THE PUBLIC DOMAIN
You have heard in this symposium about approaches used in software licensing. As to that, I believe software approaches are blazing a trail, yet they are also legally untested. In other words, we have no idea whether some of the approaches I am about to describe actually work as a legal matter.
While most recent cases have involved music and movie publishers, other major publishers have been waging this battle for a long time, and other speakers at this symposium noted the proprietary mindset of many scientific publishers, especially the elite ones. Cases from the mid-1980s involving Mead (Reed Elsevier’s predecessor) and West arose in large part because of the new dimensions of infringement that were made possible by advances in technology. In addition, in Williams & Wilkins Company v. U.S. (487 F.2d 1345 (1973) (aff’d by equally divided U.S. Supreme Court at 420 U.S. 376 (1975)) claims by a publisher of scientific and medical journals against the National Institutes of Health and the National Library of Medicine for royalties on articles photocopied from copyrighted publications were denied. Williams & Wilkins was an early and influential case on the subject of the ramifications of the availability of photocopiers on infringement claims and fair use defenses. More recently, Reed Elsevier and Thomson have been among the most forceful lobbyists for strong database protection legislation in the United States and elsewhere.
Report from the Intellectual Property Group of the Government’s Creative Industries Task Force, available online at http://www.patent.gov.uk/copy/notices/pdf/ipgroup.pdf.
Id. at pg. 3.
Id. at Recommendation 7.11, pg. 23.
The first approach has been employed by the GNU organization to promote the free software movement. What is free software? There are four components to it: the freedom to run the program—even if you did not write it yourself—for any purpose; the freedom to study how the program works and adapt it to your own needs; the freedom to redistribute copies so you can help your neighbor; and the freedom to improve the program and release your improvements to the public so the community as a whole benefits.
Freedom here is of course described not in terms of author but rather consumer of the work, or someone who may want to do something with it. The ideology of GNU’s founder, Richard Stallman, is that it would be best if all software worked this way—and copyright is not something, if all its rights are asserted, that allows those freedoms. So what is Stallman’s approach, short of repealing copyright law for software (which, to be sure, he would like to do)? He does not urge software authors to release their code into the public domain simply by failing to copyright it. This approach is disfavored because in the absence of any information about copyright, the consumer may be confused, not knowing whether or not the program is truly free to be used as he or she wishes. Another concern that would arise if software were released into the public domain is the “proprietizing” of that software. If somebody takes a piece of software released into the public domain and makes something even better with it—one starts with Mosaic and ends up with Internet Explorer—she can copyright the result. It is entirely legal to take material from the public domain and use it as the basis for new, derivative, copyrighted work. So to prevent the privatization of what had originally been free, Stallman rejects the idea of releasing work into the public domain. Instead, he suggests asserting copyright with a carefully crafted “copyleft” license. One restriction of copyleft is that when software is based on an original work under copyleft, the new work must inherit the copyleft license. This makes it so that all derivative works are covered by a license whose substantive terms operate to keep the work free—free from downstream proprietization.
We have seen some other approaches of this sort in the private sector. Mozilla.org is a recent entry to the marketplace, providing the Netscape source code but putting certain restrictions in its corresponding public license so that downstream changes must themselves be free under the same kind of license. Sun Microsystems has taken a similar approach, using a self-described open-community process—and trademark as the instrument of control. With this approach, people can do what they want with certain implementations of Java, but if they stray too far in ways that might proprietize Java, Sun asserts trademark infringement. The derivative product can then no longer be called Java.
We also have a third example from the private, nongovernmental sector: the Internet Engineering Task Force, a group that has developed certain fundamental protocols, such as simple mail transfer protocol and transmission control protocol/Internet Protocol, that make the Internet work.7 This is an example of what James Gleick called “the patent that never was.”8 The Internet Engineering Task Force does not release these standards into the public domain. Instead they have an organization, namely, the Internet Society, that holds the copyright for the purpose of replicating those standards and keeping them open a la copyleft. Internet standards are free for use, but they in fact are copyrighted by the Internet Society, and you would have to answer to them were you to try to proprietize derivative standards.
From Code to Content
Can the model formalized by Stallman for software be applied to other creative work? To address that question, I turn to a discussion of Creative Commons.
Creative Commons is a relatively new organization, founded to seed the lessons learned from the open-source (or “free software”) movement.9 Creative Commons starts with the conceptual understanding that copyright itself is a bundle of separable rights. Copyright need not mean one holds back all rights. It can mean that one holds back some rights while giving up others. Creative Commons aims to create a standardized—indeed, machine-readable—way for people to set elements of their works free categorically; that is, cleanly and clearly, but not necessarily wholly.
Creative Commons does have a human-readable as well as a machine-readable component; it is a basic “commons deed” that is attached to a work using Creative Commons templates. This deed explains in plain language what can and cannot be done with the work, either by consumers or by would-be authors who wish to create something new from it. There is also a lawyer-readable component, which is the legal code, the actual text of the license. The machine-readable component is the metadata, tags that make it so that computers can index, search, or display the work for others.
To keep the process simple, Creative Commons templates allow control by an author along five axes.10 First, one can introduce the work entirely into the public domain using language provided from Creative Commons to make it clear that it is so released. If someone wants to put a work, or an excerpt of a work, in their textbook, they need not worry about sending a permission letter to anybody because the work is declared public-domain material.
Another axis deals with attribution; there is a license signifying that others can do as they please with the work, but that they must also give the original author credit for having created it.
The third axis is “noncommercial”—meaning that, in general, one can copy or otherwise exploit the work, but only for noncommercial purposes. I am not sure Creative Commons has worked through exactly what the boundaries of noncommercial are, so this is likely to be developed further as the organization matures. The endpoints, at least, are clear: one could not put a work so licensed into an anthology and sell it, but if somebody wants to take the work and talk about it or even repost it in full on a Web log, that would be fine.
Note, by the way, that the noncommercial axis is very different from the conceptual orientation of the free software movement. Richard Stallman does not care if software (or its derivatives) gets sold for money, so long as, once it is sold, the person who buys it can make as many copies as he or she wants and can see how the code works.
A fourth axis among the Creative Commons license attributes is “no derivative works,” which means that a work can be copied, but it has to be copied exactly as it is found; one cannot incorporate it into any derivative work.
Finally, there is copyleft, which is to say—presuming that the author has not said “no derivative works”—that consumers are required to inherit the licensing terms, replicating them in any derivative work that they create.
Given this general plan, you can imagine Creative Commons as an organization functioning under a few different models:
A central conservancy. The idea would be for creators not only to adopt the licenses, but also store the work with Creative Commons. If people went to <CreativeCommons.org> and searched for “owl,” they could find pictures of owls that people have taken or rendered, with the respective license terms attached. Creative Commons has rejected this model for a number of reasons—such as the fact that they cannot necessarily build a functioning digital library given the kind of infrastructure that would be required to support it. Other reasons include the fact that Creative Commons could not necessarily verify or validate the works coming in—and perhaps the work would not really be authored by the person submitting it, thereby subjecting Creative Commons to possible liability for hosting it.
Another option would be a distributed conservancy. In this conception, the material is hosted somewhere on the Internet, or perhaps in a library as a physical object, but Creative Commons would maintain a central index of works under Creative Commons licenses. One could go to <CreativeCommons.org> and look for owls, find several owls that meet the description of the clip art one wants, and follow the links to the work. That might attenuate the legal liability that Creative Commons could face for contributors misrepresenting or simply being mistaken about what rights they can convey or release with regard to the material in question.
Finally, Creative Commons could build a completely distributed conservancy with no index at all. The main product of Creative Commons in this conception would be the licenses themselves. How would there then be an index? One would simply go to a search engine like Google, type in “owls” and add search terms matching metatags indicating the presence of Creative Commons licenses.
As far as I can tell, Creative Commons is somewhere between models (2) and (3) in this taxonomy.
My own sense so far is that Creative Commons is a spectacular project. It is interesting to note that it is not necessary or even likely that a work associated with Creative Commons will literally enter the public domain; the project is more complex and more flexible. Should we then ask whether the glass is half empty or half full? Creative Commons gives people an easy means—a computer-readable, searchable means—to assert what they desire with respect to content they create. This may result in more content being more clearly omitted from the public domain than clearly placed there, as candidates for Creative Commons licenses could be drawn from works that would otherwise be released entirely into the public domain, rather than works that would otherwise be copyrighted in the traditional way. Digital rights management systems could easily be placed on top of the machine-readable code, making the licenses self-enforcing. So you might even think of Creative Commons as converging with what the British Patent Office report called for: a greater understanding by people of the rights they have and can assert in the works that they create; an easy, simple, nonlawyerly way of expressing that desire; and, to the extent possible through machine execution, having it be so. One should note that this is a goal different—perhaps laudable, but different—from the goal of having as much material as possible enter into the public domain.
Another aspect of this discussion is important to note: the problem of derivative works is less a problem in relation to creative works than it is in software. In creating software, the purpose of using another’s source code is to create a new code that does something else. In this sense, one directly “builds on the shoulders of giants” with software—something one does more conceptually than literally with creative work. An author does not literally have to incorporate another’s poem into her own to have a successful or meaningful new work.
Finally, in the area of the scientific and technical data not subject, at least in the United States, to copyright protection, Creative Commons offers an opportunity to affirmatively and publicly catalogue the data, allowing the author to make clear the fact of his or her original production of them for attributive purposes. Scientific researchers thus can encourage maximum dissemination of their discoveries and methods, without sacrificing the fact of their own contributions to those discoveries.
What, then, is the real value of Creative Commons? First, it helps us to identify works intended for the public domain. Second, it helps people join the cultural melee. This is a battle over the description of rights, an assertion of copyright as an instrument, and not just an instrument for control. That part we understand quite well. Creative Commons helps to underscore the fact that a legitimate use of copyright is not simply to stop others from copying, but also to give permission—to imagine that something other than “All Rights Reserved” could be the phrase that follows one’s assertion of copyright ownership.
Will Creative Commons work? That will depend on the value of the work committed to the public domain, or at least to public use, under the Creative Commons system. Exactly what sorts of authors and work it will attract, we have no idea. This is one of those ideas so new that one really has to make a leap of faith to see if it is going to work. And then, if it does work, it will seem obvious that it was a great idea whose time had long since come.