can give the dollar value in the economy and its contribution to global trade. Furthermore, the dollar value of the contribution of the computer software industry to the economy is not so important as the fact that it is a critical component in every other sector of our economy. I can use several analogies to describe its importance, such as the "crown jewels," which Professor Oettinger objected to because he said crown jewels were not very productive and that the goose that laid the golden eggs was a better metaphor to describe the role of computers software in generating productivity. Adapso, in testimony before the Senate Judiciary Committee, has called it the brains of industry, and you could also call it the cog in the wheels of commerce. Whatever you call it, computer software is a technology which is highly utilitarian and our productivity as a nation and our competitive position in world markets is dependent upon it. So I think the better question to ask is, what is different about computer software that requires special attention in the law? Simply put, computer software consists of electrical impulses which open or close circuits, and are represented as zeros and ones. It appears in a digital form that disguises the nature of the signal, whether it be a book or a map or data. It can be easily distributed by networks where it's easily copied. The cost of copying is negligible compared to the high cost of development. Therefore, if the treasured asset is different and is difficult to protect other than through the legal system, I think we have to ask what can we do to improve the way the legal system treats computer software. Now, you have my written statement and I'm not going to read from it, but I would like to highlight three areas of concern which I think demonstrate some of the areas which Congress might address. I think one is the question of the intellectual framework of the use of the term "copy" to apply to computer software, in that "use" offers a better intellectual framework within which to analyze the utility of computer software. The second area that I would like to highlight is reverse engineering, which really constitutes a question of what is fair use of intellectual assets. The third area I would like to emphasize is that of the standardization of the user interface or the machine-to-machine interface, which is creating some of the tensions between various components of the computer industry. Now, use is really the measure of success for computer software, and I think the term "copy" is misleading, because in some instances, such as in a database, you may need only one copy for all uses, where people can access that one copy. Therefore, in order to measure the productivity in the economy or to provide fair compensation, it's much better to measure use. Also, it is necessary to make copies in order to use computer software. We make backup copies and archival copies. You have to copy from the disk into the machine memory in order to use it. A screen display is a copy and a printout is a copy. So copying is essential to computer software. The nature of computer software is that it is designed for multiple uses. Use presents a much better intellectual framework from which to determine when one becomes an infringer, so we need to determine what kinds of commercial uses are permitted for exploitation and what kinds of private uses or public uses should be authorized. The next area I would like to highlight is reverse engineering. This is a concept which is borrowed from the patent field and it's really a question which is similar to what constitutes fair use under the copyright rubric. When is it fair use to exploit another's intellectual productivity and when does it become misappropriation? Now, the reason why we have reverse engineering-and there are many people who are wedded to this concept. As you know, the right to reverse engineer was written into the Semiconductor Chip Act. We reverse engineer products to learn, for educational purposes, to improve upon the technology and to build a better mousetrap. But during the period of the monopoly that is granted by the patent laws, one cannot build the same mousetrap and market it during the 14 or 17 years after which that knowledge becomes part of the public domain and may be used by all comers. In computer software, there is no monopoly on the functionality or the ideas behind the development of a software, but only on the expression, which is protected within the copyright rubric. Therefore, there has developed a procedure called a clean room, in which one group of computer programmers sit on one side of a barrier and describe the functionality of a piece of software that they wish to replicate, and the group that has clean hands in the clean room write the code which can carry out that functionality. If they do it without using the original code, then they may copyright the new version and may market it. This process can be completed quite rapidly. Without the protection of patents, cloners, therefore, can produce a similar product without the large investment that's been made by the originator. There is a great reluctance, therefore, on the part of software developers to release their source code and the object code, which many programmers-such as Richard Stallman at the Free Ware Foundation insist upon. Access to the code may be necessary for users in order to be able to use that software productively and to apply it to the particular function that they wish to use, to maintain it and to improve upon it. This is one of the primary areas that Congress should address: What is fair use for what components of software? Perhaps a shorter period than copyright provides of something that is more like patent protection may provide time to recoup the capital investment, yet expedite the sharing of the basic knowledge underlying it. The third area I would like to highlight is the question of the standardization of the human or machine interface. This is a question of when knowledge should enter the public domain, or what my friend Harlan Cleveland calls the global commons. The history of the computer software industry has been one of incremental development with different software programmers using the building blocks to improve upon a previous version. There are a group of cases going through the courts which are called look and feel or structure, sequence and operation cases, in which things like the appearance of a garbage can on the screen might be protected under copyright, whereas it is something that facilitates the user interface and is something that all people who are writing programs for a particular function would like to use. Now, what's the difficulty in this? I think the courts are going in this direction because they are trying to provide equities and compensation which is not provided under the copyright law and because many of these forms of computer software may not be patented because they are more incremental than novel. But if programmers are required to make something that is different, that increases the diversity of the types of programs that appear for a particular user group. The users, on the other hand, find it far more comfortable to use a similar interface or a screen display. In migrating from one level of software to another, uses don't want to have to learn a whole new system. There's nothing more frustrating than having a can represent a storage capacity in one program and garbage in another, so that you save your data in one and throw it out in the other. There's a good example in the difference between the dictionaries of WordPerfect and PFS Write. In one program the number 1 skips to the next word and ignores it, and the other one puts the word into the dictionary. You inadvertently do something you didn't intend, because the different programs are pushed in the direction of increasing the diversity. The question is not whether copyright principles can be applied but whether they should be applied and how can we best encourage user-friendly, compatible human interfaces which enhance the productivity for that particular sector of the economy, whether it be word processing or financial management or medical care or whatever. Now, you will be told by eminent and well-intended lawyers and scholars that, if the system is not broken, why try to fix it? Well, the system may not be broken, but it certainly doesn't fit the new technologies of artificial intelligence, of databases, of laser disks, computer-assisted design, computer-assisted manufacturing, video digitizers and so on. The intellectual property laws are really bursting at the seams with new technologies with which we can find analogies, which almost but do not quite comfortably fit. It's like wearing the hand-me-down clothes of older brothers and sisters. And that may be appropriate for a third world country, which we seem to be determined to become. It is not appropriate for the leading economy of the world. The best answer I have found for not forging new laws, or modifying existing ones, is that our lawyers are comfortable with the old ones and that the Berne Convention might not apply if we called our laws something other than copyright. Well, I think that we who have the longest experience and largest investment in software should be leading the world with a better fitting legal regime, as we did through the leadership of this committee and the chairman in the protection of mask works or semiconductor chips, rather than browbeating our trading partners into accepting second or third best. In summary, computer software is vital to the health of the U.S. economy. It is a treasured asset in improving our competitive advantage in the global economy. Intellectual property laws do not comfortably fit computer software, leaving confusion and contradictions. Adjudications are slow and expensive. They add to the cost of U.S. products in the world market. A simplified, unified, expeditious and harmonized system, both domestically and internationally, would be highly therapeutic. We are highly innovative and pioneering people, and there are frontiers to be conquered in the law as well as in the technology. So I ask you the question: If Congress is not prepared to take the lead, who will? Mr. KASTENMEIER. I suppose the courts would. [Laughter.] Ms. BRANSCOMB. Well, the courts are supposed to refine the law and not make it. Mr. KASTENMEIER. I wish, as a matter of fact, there would be more reliance upon the Congress to settle intellectual property questions. I thank you for that statement. [The prepared statement of Ms. Branscomb follows:] PREPARED STATEment of Anne WELLS BRANSCOMB, ESQ., RESEARCH ASSOCIATE, PROGRAM ON INFORMATION RESOURCES POLICY, HARVARD UNIVERSITY, AND ADJUNCT PROFESSOR OF INTERNATIONAL LAW, FLETCHER SCHOOL OF LAW AND DIPLOMACY EXECUTIVE SUMMARY STATEMENT OF ANNE WELLS BRANSCOMB, ESQUIRE November 8, 1989 BEFORE THE SUBCOMMITTEE ON COURTS, INTELLECTUAL PROPERTY, AND ADMINISTRATION OF JUSTICE PROTECTING THE GOLDEN GEESE OF THE INFORMATION ECONOMY Computer software is essential to almost every sector of an information economy. Therefore, even if it were not also a growth industry contributing substantially to U.S. trade, it would deserve the special attention by Congress. There is a great deal of ferment within the computer software industry about the directions the courts are moving within the current framework of intellectual property law. There are several competing legal regimes into which some, if not all, of the computer software and its constituent parts fall. Attempts to fit round pegs into square holes through extensive litigation adds to the cost of doing business in both the domestic and international markets. Thus users, software providers, and the public coffers would all benefit from a unified, simplified, and harmonized system of legal protection for computer software. Areas of current ferment in the substantive law which deserve attention include (but are by no means exclusive): - clarification of existing intellectual property laws or enactment of a new statute for computer software protection; while patents may provide greater protection for truly novel computer programs, the major innovations in software development are incremental, whereas the courts are extending the coverage of software in a manner which lengthens the protection to a far longer period than would be available under the patent system. - - reconsidering the existing intellectual framework for computer software protection; "use" "access" or "service" as a theoretical framework may be far superior to "copy", "invention", or "secret". Copyright is cumbersome as, copying is the very essence of the productivity of computer software. - evaluating the right of "reverse engineering" as it applies to computer software development - redefining the balance between proprietary and public claims on intellectual assets which computer software represent - reconsideration of the effects of noncopyrightability of government originated software development reevaluation of the effect of the "first sale" doctrine" There are also elements in the administrative and judicial procedures which can be improved. These include: - the unification of administration of the intellectual property laws which are currently divided between Congress and the Department of Commerce and the states. - bolstering judicial capability with professional expertise in computer software. better access to |