Advances in computers and data networks inspire visions of a future ``information economy'' in which everyone will have access to gigabytes of all kinds of information anywhere and anytime. But information has always been a notoriously difficult commodity to deal with, and, in some ways, computers and high-speed networks make the problems of buying, selling, and distributing information goods worse rather than better.
To start with, the very abundance of digital data exacerbates the most fundamental constraint on information commerce---the limits of human comprehension. As Nobel laureate economist Herbert A. Simon puts it: ``What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention, and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.'' Technology for producing and distributing information is useless without some way to locate, filter, organize and summarize it. A new profession of ``information managers'' will have to combine the skills of computer scientists, librarians, publishers and database experts to help us discover and manage information. These human agents will work with software agents that specialize in manipulating information---offspring of indexing programs such as Archie, Veronica and various ``World Wide Web crawlers'' that aid Internet navigators today.
The evolution of the Internet itself poses serious problems. Now that the Internet has been privatized, several companies are competing to provide the backbones that will carry traffic between local networks, but workable business models for interconnection---who pays how much for each packet transmitted, for example---have yet to be developed. If interconnection standards are developed that make it cheap and easy to transmit information across independent networks, competition will flourish. If technical or economic factors make interconnection difficult, so that transmitting data across multiple networks is expensive or too slow, the largest suppliers can offer a significant performance advantage; they may be able to use this edge to drive out competitors and monopolize the market.
Similar problems arise at the level of the information goods themselves. There is a growing need for open standards for formats used to represent text, images, video and other collections of data, so that one producer's data will be accessible to another's software. As with physical links, it is not yet clear how to make sure companies have the right economic incentives to negotiate widely usable standards.
In addition to standards for the distribution and manipulation of information, we must develop standards for networked economic transactions: the actual exchange of money for digital goods. There are already more than a dozen proposals for ways to conduct secure financial transactions on the Internet. Some of them, such as the DigiCash system, involve complex encryption techniques; others, such as that used by First Virtual, are much simpler. Many of these protocols are implemented entirely in software; others enlist specialized hardware to support electronic transactions. ``Smart'' credit cards with chips embedded in them can perform a variety of authentication and accounting tasks.
Even when the financial infrastructure becomes widely available, there is still the question of how digital commodities will be priced. Will date be rented or sold? Will articles be bundled together, as is done today in magazine and newspapers, or will consumers purchase information on an article-by-article basis? Will users subscribe to information services, or will they be able to buy data spontaneously? How will payment be divided among the various parties involved in the transaction, such as authors, publishers, libraries, on-line services and so on? Not one of these questions has a definitive answer, and it is likely that many market experiments will fail before viable solutions emerge.
The shared nature of information technology makes it critical to address issues of standardization and interoperability sooner rather than later. Each consumer's willingness to use a particular piece of technology---such as the Internet---depends strongly on the number of other users. New communication tools, such as fax machines, VCRs and the Internet itself, have typically started out with long periods of relatively low use followed by exponential growth; this means that changes are much cheaper and easier to make in the early stages. Furthermore, once a particular technology has penetrated a significant portion of the market, it may be very difficult to dislodge. Fortunes in the computer industry have been made and lost from the recognition that people do not want to switch to a new piece of hardware or software---even if it is demonstrably superior---because they will lose both the time they have invested in old ways and the ability to share data easily with others. If buyers, sellers and distributors of information goods make the wrong choices now, repairing the damage later could be very costly.
All this discussion about managing, distributing and trading in information is overshadowed by the more fundamental issue of how much data authors and publishers will be willing to make available in electronic form. If intellectual property protection is too lax, there may be inadequate incentives to produce new electronic works; conversely, if protection is too strict, it may impede the free flow and fair use of information. A compromise position must be found somewhere between those who suggest that all information should be free and those who advocate laws against the electronic equivalent of browsing at a magazine rack.
I believe that extending existing copyright and patent law to apply to digital technologies can only be a stopgap measure. Law appropriate for the paper-based technology of the 18th century will not be adequate to cope with the digital technology of the 21st century; already the proliferation of litigation over software patents and even over the shape of computer-screen trash cans makes the need for wholesale revisions apparent.
Computer scientists have been investigating various forms of copy protection that could be used to enforce whatever legal rules may be put into place. Although such protection often inconveniences users and requires additional hardware and software, ubiquitous network access and more powerful machines may eventually allow for unobtrusive and effective protection. File servers, for example, can track who owes how much to whom for the use of particular information, and documents can be subtly encoded so that investigators can trace the provenance of illicit copies.
Faced with such a daunting list of problems, one might be led to question whether a viable information economy will ever take shape, but I believe there are grounds for optimism. During the 1980s, 28,000 for-profit information libraries sprang up in the U.S. alone. Every week more than 50 million people visit these facilities, where they can rent 100 gigabytes of information for only two or three dollars a day.
Although these video rental stores faced many of the same problems of standards, intellectual-property protection, and pricing that the Internet faces today, the industry grew from nothing to $10 billion a year in only a decade. Ten years from now we may find the economic institutions of the information economy a similarly unremarkable part of our day-to-day life.