After reading the National Information Infrastructure debate on intellectual property reform in the digital age, one could conclude that computers and copyrights have come to an impasse. The copyright code assumes that copying is difficult and expensive, hence authors are rewarded on a per copy basis. Digital technology makes such copying nearly costless. Why reward authors for their part in a costless transaction?
Some have proposed drastically curtailing electronic technology in order to protect future publishers. They want to put all forms of computer copying under the copyright code -- even reading a program from disk into RAM. They want to ban the electronic resale or renting of copyrighted material fearing that the piracy which has plagued software will plague movies and books when they enter cyberspace.
I believe it is possible to use this conflict as fuel for a new, stronger system of copyright protection. First we need to understand the function of intellectual property laws and the capabilities of digital technology.
In the past, copyright law has not been able to reward authors based on the benefits others gain from using their creations. The law has had to equate purchasing or viewing a copy of the work with using it because the cost of verifying and enforcing any other kind of contract has usually been too high. It was also assumed users were literate enough to interpret the replicas they purchased.
Digital documents, unlike written ones, can be quite easily manipulated, reproduced and transformed. Individuals can link numbers together into combinations that can be hard to understand. This fluidity contains the solution to the copyright dilemma.
In the computer world, context is everything. Every byte gets interpreted at some point whether it's in a text file scanned by a printer putting symbols on a page or it's in a program being read by the microprocessor. Passive data doesn't exist; everything's an instruction. But the context needed for proper interpretation is not always portable on the user's whim. It is impossible, for instance, to run Windows programs on a Macintosh without purchasing a special operating context.
We can adapt copyrights to the digital world if we realize that it takes a special form of literacy to use digital information. In other words, duplicating an author's digital work may be easy, but decrypting it is not. That requires extra, privately held information.
Using this contextual insight to create secure digital copyrights requires adding two more layers of interpretation to computer operating systems. Not only must data be encrypted, it must also have the power to communicate with its publisher for the unique and private key needed to decrypt itself. Only by using two layers can we prevent the theft of both the original document and the private information (context) needed to make the document useful.
I call this a system of 'telerights.'
A user would buy an encrypted copy of the document from the author or publisher. Each individual version would have a different key, so the user could make many duplicates but in essence only own the one 'copy' that was paid for. When the user wanted to use the document, it would contact its publisher for the key. If no other versions with the same key were in use, the publisher would send the key to the user's machine and the document would decrypt itself into an area of temporary memory like RAM. When the user was done, the document would delete the decrypted version.
Users would regain the fair use property rights in cyberspace they have come to expect in the printed domain. Users could make and pass around as many copies as they like because the publisher could guarantee a one to one correspondence between sales and use. A professor, for example, could lend out books to his students or sell used software -- two acts considered illegal under the current system of digital copyrights.
The old problem of piracy would be turned on its head. The user instead of the publisher would have to worry about theft. When someone stole his copy, they would steal his use of it as well. There would be no assurance the person you buy used information from would delete their old copies.
Because the bond of contractual trust between publisher and purchaser would be perpetually renewed, this would be a simple problem to guard against. The publisher would always know where and when a work was in use, so users could ask publishers to ban certain sites from using a particular version of a document (in the case of used data) or restrict use to a particular site or user ID (if we're dealing with virtual terminals). This is one way parents could control the flow of information going to their children. In any event, this wouldn't terribly onerous. A reseller could always purchase another version with a different key.
Stronger property rights will vastly expand the market for copyrighted material. Libraries would no longer have to purchase books that went unread. They could average out their risk in the same way some banks do with mortgage derivations by pooling their funds together and purchasing expensive or esoteric information for their common use. Corporations could purchase telerighted documents, then double encrypt them and rent them out with fixed duration keys. By republishing a document you could loan it to a friend without worrying about it being returned.
By knowing when, where and what parts of a collaborative work were in use, publishers would better know how to reward individual artists in a project. I could use parts of a telerighted document in my own publication and the original publisher would know better how to bill me for that use.
Furthermore, one could break up the public airwaves into multiple private channels. ABC could teleright its sitcoms and distribute them freely over the internet. No one would ever watch quite the same thing at quite the same time again, but network store and forward technology would vastly decrease distribution costs. By telerighting each copy, the networks and their advertisers would know exactly who watched their programs and how often.
Telerights combine networking technology with encryption, software meters and object oriented intelligent documents... items that have all been hot topics by themselves or in various subsets. To my knowledge no one has yet brought these concepts together under one roof, although a few efforts have come close.
Since before the start of microprocessor clocks, Ted Nelson has been pushing the concept of hypertext publishers. His ideas have inspired World Wide Web development and a host of other network computing innovations. By itself, though, his theoretical Xanadu server lacks cryptological security and fails to take advantage of high bandwidth networks, distributed computing and object-oriented data distribution.
The Copyright Office itself is creating an archive that will bind copyrighted documents together with information about their publishers. The Copyright Clearance Center has set up a World Wide Web site where users can negotiate rights with publishers and, in conjunction with Folio Corp., it is designing a CD-ROM LAN system so that universities and large businesses can be billed according to how they use information.
These projects target large government institutions and private corporations because they have a large technological base and both use and publish vast amounts of intellectual property. They are natural starting places for new systems of copyright protection.
These projects, though, won't keep up with the technology as it spreads into the home. In the future information won't be this easy to centralize. For one thing, CD-ROMs will be easy for individuals to copy and although piracy is rather easy to detect in large organizations, individual users won't be paralyzed by that same fear.
On the consumer side of today's market we have 'interactive TV' -- a decidedly different sort of venture since most homes cannot manage digital information. This system gets around the piracy issue by directly broadcasting information on demand to home televisions. Because people lack the technology to copy the information entering the home in any useful way, many scholars have proposed using on demand broadcast as the model for the future distribution of intellectual property.
I dislike the on demand system because it denies individuals the ability to own their own copies of information. Even if you get around the problem of future users being able to record information transmitted into their home, under telerights it will always be cheaper to resend a key instead of the whole document. 'Interactive TV' is a poor model for the future because it concentrates too much on the later half of the term -- the television, a very dumb device.
These projects make important technical and social contributions to the protection of intellectual property. In all fairness to them, even though the technology exists, the infrastructure needed to make a system like telerights work is not yet in place. I think getting it in place will be the most difficult part of the whole affair.
I am assuming, of course, that the problems telerights share with other internet projects can be solved -- namely, that everyone will get a personal, high bandwidth network connection and generic security issues will be laid to rest. If it turns out to be technically impossible to produce secure cryptographic codes or prevent the counterfeiting of network addresses, my project and a host of others will go down the drain.
I'm optimistic these technological issues will be settled, so that leaves us with the question of how to standardize teleright technology and make it widespread. This will take some coordination among computer makers, operating system designers, phone companies, network providers, software concerns and, of course, the government.
The government does not need to alter existing copyright laws; telerights are merely new contractual forms. It does need to clarify the regulatory environment and encourage companies to agree on and accept a standard. Since the government is a large producer and consumer of intellectual property, it can facilitate the acceptance of a standard by using public monies to purchase first generation systems.
The contemporary regulatory breakdown of the old content/carrier distinction makes all this problematic.
On the consumer end, there is the privacy issue. Any company that both maintains other people's teleright accounts and publishes its own documents will be tempted to use for financial gain private information about other companies' customers.
It would perhaps be best to assign day to day teleright accounting to a government agency or semi-public utility. After all, when you transfer title on land, it's publicly registered at the local courthouse. Telerights aren't much different. Government is the ultimate enforcer of property rights and is also the best place to ensure anonymity in the collection of economic statistics.
On the corporate end, without clearly defined lines in the market companies will be tempted to use the standardization process to compete against one another. It will take longer to establish a winner if rival standards emerge. In the interim, old copyright laws will continue to be inadequate in the digital world so a delay in the standardization process might make attractive solutions that look less palatable today.
The issue of encryption itself is sticky because there are already two established and ideologically opposed groups fighting about it. The government must be coaxed into relaxing its objections to strong encryption and Clipper opponents must learn to accept a key escrow standard to which the government has warranted access.
Bobby Inman has proposed using the Federal Reserve as one of the depositories for cryptological keys for a very good reason. In an information economy, information is money. Just as the government supervises banks to make sure that they do not launder money, counterfeit currency, falsify bank sheets or expose themselves or their customers to abnormal risks, so too must the government regulate telerights. One doesn't need to look at the Great Depression to remember the benefits of regulation. The '80s S&L debacle will suffice.
The government must also encourage software and computer companies to accept some level of professionalization. With the proper tools and knowledge it will be possible to trap keys or decrypted documents stored in temporary memory. These tools and skills must be tightly regulated and those sections of the operating system must be shut off from amateur tampering. This may cause angst among some programmers but for most of us this should not be a burden. It does, after all, take a license (and the proper employer) to tamper with phone boxes and electric meters. Part of the digital frontier must be closed and civilized if society is to settle the land and make it productive.
Although it may be possible to make telerights work without encryption, it would involve a more onerous regulation of programming. One would have to vastly restrict low level media access to make unencrypted telerights work because it would be easy to pull raw information off the disk with a sector editor. With encryption, the restrictions are narrower and easier to enforce because the data is coded wherever it is stored in permanent form. Only certain sections of the runtime environment need to be restricted.
Telerights will not eliminate all forms of electronic piracy, but by lowering transaction costs and raising the risk and expense of counterfeiting, the most widespread individual and institutional forms of piracy will be severely curtailed.
It won't prevent the pirating of physical, non-digital materials like books but it doesn't have to. Information goes digital because it is more useful in that form. Hypertext indices cannot be printed out and scanned back in nor can people make videotapes of movies playing on their screen and expect to retain the original fidelity. A digital document is a non-linear object that always plays back in a linear frame.
Given the low costs of electronically distributing materials and the strength of the teleright protection system, many would-be pirates may opt instead to sub-license and re-publish materials both in physical and electronic form.
Users do have to give up some 'privacy' to make telerights work, but their behavior is not turned into a public document. Like a number of other encrypted communication schemes, telerights breaks up public broadcast channels into a multitude of two-way private channels that span the boundaries of public and private, sprawling across an indeterminate number of nodes.
Users will chose give up some privacy for the added convenience and the lower costs of information. They will also gain a measure back since they will be able to encrypt information about themselves and better control who has access to it.
We are, to use the old Chinese pejorative, living in interesting times. Why the Chinese have historically found this undesirable I do not know. Their word for 'crisis' means both 'danger' and 'opportunity' and our western capitalism has always been fueled by such critical change.
When it comes to social conflicts over the development of the internet, we shouldn't forget the creative part of Schumpeter's "creative destruction." We just need to follow Carver Mead's advice to "listen to the technology"... in this case, at least to our own cliches.
If the future is one of distributed intelligence where everyone will be a publisher, then we should distribute the responsibility for enforcing copyrights by making our documents intelligent.
We must stop thinking of copyright stamps as passive marks. Their presence has always made a very active social statement and should continue to do so in the future.
Wade Riddick is a graduate student and National Science Foundation Fellow in the Government Department at the University of Texas at Austin. His email address is riddick@jeeves.la.utexas.edu.