O'Reilly: What do you think about the efforts of some companies, particularly BMG, to try to come up with a solution that is more accommodating to Napster?
Lessig: I think this should be encouraged and I think a wide range of alternatives should be encouraged, but these alternatives should not all turn on the power of the industry to say yes or no. Encouraging the development of companies outside of the traditional media is an important way to make sure that the potential for innovation here is realized.
Look at, for example, the MP3.com case. All MP3.com was attempting to do was allow people to listen to their music more easily. It wasn't facilitating copying of anybody else's music. It was just facilitating my ability to listen to my CDs whether I'm sitting at home or work. Now, that was a pretty creative way to attempt to make the music more valuable for the consumer. There is no showing that there is an increase in the amount of music that's being stolen because of this; in fact, they were taking important steps to make sure people didn't wrongfully copy music here. But the legal system's response was to say, "absolutely willful violation of the copyright laws," and we're going to punish you as strongly as we can - something like a $100 million fine against this company - for their experimentation with different modes of facilitating distribution.
Now, if that's the legal backdrop against which innovation in this context has to occur, then you're going to see much less innovation. The one area of development where this backdrop is going to have the greatest significance is in P2P-like architectures, where you are essentially creating less easily controllable structures for facilitating the exchange of data. In a free society one would think we're not worried about that, but the point is this is anathema to the interests of the copyright industry.
O'Reilly: What you're saying effectively is that the balance is tilting too far in favor of copyright holders. Are there other cases where you know the copyright industry is trying to extend its reach at the expense of rights like fair use, and so forth?
Lessig: Well, sure. You remember when this debate about copyright in cyberspace began, it was the copyright industry racing around saying that cyberspace was going to kill us, that copyright is going to be impossible to enforce, that we need lots of new laws. And they got some new laws, including the Digital Millennium Copyright Act, which attempted to strengthen their legal rights.
Now, from the very beginning there have been people, and I have been in this chorus, who've been saying, "Wait a minute, to the extent that the architecture of cyberspace weakens the position of copyright interests, they have a handy remedy and that is the development of code that facilitates the better control of their data than exists right now."
And there were people like Mark Stefik at Xerox PARC who were sketching architectures for facilitating the control of data with trusted systems, so that copyright interests would be able to distribute their content without fearing that the content would be stolen by other people. The reason that's significant is that these trusted systems were not just efforts at making sure that the copyright interests would have the same rights that they had under copyright law, but in fact could be developed to give copyright holders more rights than they had under traditional law.
So the copyright holder could guarantee that his work would be used exactly as the copyright holder wanted, as opposed to the traditional legal structure, which is that the copyright holder has some control over the use of the work but never perfect control.
That's where the real battle is going on, as code plus law combines to reduce the rights that consumers used to have under the old copyright. The best example of this is the DeCSS case - in fact, I'm writing a brief right this minute in this case - where the Court of Appeals is going to be confronted with the question of Congress' law - the anti-circumvention provision of the Digital Millennium Copyright Act, which says that if you develop tools for cracking a copyright protection scheme, then you've violated the law. The 2nd Circuit is going to have to ask whether that law is constitutional because, in effect, it says you cannot crack copyright protection schemes even if your purpose is to allow you to have fair use of the underlying material.
So, in this case the claim is that the particular use that was facilitated by this code - DeCSS was designed to crack the encryption code CSS on DVD discs - was a completely fair use under traditional copyright law. But by banning the code, the law is effectively eliminating the fair-use protection that traditionally exists in copyright law.
Richard Koman: Can I ask you to give a traditional definition of fair use?
Lessig: Sure. The traditional idea of fair use - and the law has been extremely vague in defining this - is that the copyright owners do not have the right to perfectly control how you use their copyrighted material. You have the right under fair use to make limited free use of copyrighted material for purposes of criticism, for purposes of display, for purposes of personal consumption, and so on.
In the context of software, this has been an extremely important issue about whether you have the right to reverse-engineer software to figure out how to make it interoperable with another piece of software that you might have. The software industry has tried to say you have no right to reverse-engineer in order to make it interoperable. But the courts have been pretty good in saying that reverse-engineering is a use of the copyrighted work that should be protected, regardless of the interests of the copyright holder to protect themselves against reverse-engineering.
So, this means that you can Xerox a chapter out of a book or make a copy from an article in a journal, if you want to take it home and read it. But it doesn't allow you to, for example, copy the whole book and hand out a copy to all of your friends - that's not fair use - or for a professor to copy all of the articles he needs for his class and hand them out instead of buying the case book that is necessary for the class. The law tries to draw a distinction between uses that facilitate the distribution of this information and uses that undercut the market that the original work was designed for. That is the objective of fair use.
What is happening in the digital realm is that fair use is being coded away both through technology and through the licenses that you agree to when you sign up for some of these technologies.
A wonderful example of this is the electronic book of "Alice in Wonderland" -- this is a public domain work, first of all - and on the very first page a list of credits include the list of rights that you have purchased by getting this book (and, in fact, it was free, so you didn't purchase anything at all), it said, "You have no right to read this aloud." So basically you're not allowed to read "Alice in Wonderland" to your child at bed. That was a right you had waived or given away by virtue of getting this book from that person or from that particular company.
What's happening in digital books generally is that a whole bunch of rights that you would effectively have with ordinary books - like I could loan it to my friend, I could destroy it, I could copy a chapter out of it, I could read it to my children, I could sell it somebody else - all of those rights are erased in the digital context because these shrink wrap licenses and the code built into these books makes it impossible for you legally to give it to a friend, or to sell it to somebody afterward or to copy a chapter out of it or in this case, to read it to your child.
So what they are doing is using contracting code to restrict the rights that you used to have. The reason they can do this is that copyright law has always permitted some amount of contracting in addition to the rights granted by copyright. The fact is people didn't waste their time entering into those contracts before because they were essentially unenforceable.
You could, in principle, write whatever you want into the shrink wrap license selling the book, but what are they going to do? You can't give this to a friend, how are they going to police that? So because it is impossible to police, there is no reason to require it. But now the technology makes it so that you can begin to police it, so the copyright interest says, "We've always been able to add these restrictions. Now we're adding these restrictions and they should be as enforceable as they were before."
But the pushback to that is the context has changed so significantly that to permit you unrestricted ability to add conditions onto the distribution of this type of work is to make it so that we no longer have any balance between protection and free use or fair access. That kind of ambiguity that used to exist in the law creates the opportunity for these rights holders to expand their rights as fast and broadly as they can before the rest of us have a chance to catch up.