The Illustrated Story of Copyright
© 2000 by Edward Samuels

Return to Contents

[74]

CHAPTER FOUR

________________________

 

The Computer

 

 

 

The Workings of the Computer

 

Probably no technology has had a more profound effect upon copyright, upon the creative process, or, for that matter, upon our lives, than the computer. Before getting into the intricacies of copyright, however, it will be helpful to spend just a little time exploring what computers are and how they work.

 

The Hardware

 

The real guts of most computers is the central processing unit, or CPU. (The underlined key phrases are illustrated in the diagram on this page.) Basically, this central processing unit is a very fast traffic cop. It receives information, in the form of electronic impulses, and reroutes it to other parts of the computer. These impulses are simply electronic on/off switches. Ultimately, all the information a computer receives, stores, manipulates, and sends is simply electronic impulses that are either “on”—there is an electrical charge stored in a particular place in the computer—or “off”—the relevant location in the computer does not have an electrical charge. The CPU does its magic by handling an amazingly large number of [75] operations in a very short period of time—about a billion operations per second on the most recent generation of computers.

 

            Another important function of the computer is the simple storage and retrieval of electronic impulses in what is called the memory unit of the computer. The memory unit stores the electronic impulses so that they can be retrieved by the central processing unit at a later time.

 

[75] Only $1,295!

This is a picture of a complete general purpose computer, here about twice actual size, from the cover of Scientific American  (May 1975). The computer was made by Teledyne Systems Company from chips manufactured by National Semiconductor Corporation. The two largest chips, in the center, were integrated circuits containing what at the time was the remarkable capacity of six thousand transistors each. These two chips controlled the four slightly smaller chips devoted to arithmetic and logic circuits, and the numerous other chips. The computer, mounted on a ceramic wafer and with six glass-insulated conducting layers, communicated with the outside world through 120 leads, 30 on each side. The cost of the computer at the time was $1,295.

            There are different types of memory in computers. Random access memory, or RAM, is basically short-term memory that’s stored in such a way that it’s immediately accessible to the central processing unit. In most computers, the electronic impulses in RAM must be constantly recharged. Think of RAM as live electricity flowing through the system. Turn off the computer, and the flow of electricity ceases—the information stored in RAM disappears. This form of memory is very good for information you want to access and process very quickly, but it isn’t very good for storing a word processing document until you need it next Thursday.  

 

            For that longer-term memory, you need some form of storage unit that will reliably retain the information well after the computer itself is turned off. One storage device is a hard drive, which basically records a copy of the electronic impulses on platters that spin within the computer. Other storage devices are magnetic tapes or floppy drives, magnetic disks or optical disks, or CD-ROMs, all of which store the information on a separate medium that can be removed from the computer and reinserted at a later time, or can be transferred to other computers to share the information encoded on them.

 

             Some information is so critical to the operation of the computer that you want it always to be accessible, and not normally altered. Such critical information [76] is usually contained in a permanent storage unit known as read-only memory, or ROM. Think of ROM as the hard-wired circuits that can’t normally be changed by the user of the computer. (You might notice that the line between computer memory and storage is not at all a fine one. For reasons that may be more historical than functional, read-only memory and random access memory are usually considered memory units, while hard drives and removable media are usually considered storage.)

 

            Other parts of the computer are the arithmetic unit, which can add, subtract, multiply, divide, and perform other advanced calculations; input devices, such as a keyboard, a mouse, a microphone, a midi (musical interface device), or various devices that allow the computer to recognize video images from a television receiver or video camera or video recorder; output devices, such as a computer monitor, a printer, or speakers; and connections to allow the computer to communicate with other computers through direct connections or over telephone lines. Once computers can access telephone or other long-distance lines, they can be linked via global networks like the Internet, so that practically any computer can access any other computer, wherever located.

 

          Computer Programs

 

At the end of this chapter, we’ll take a look at the profound effect computers have had on a broad range of creative works. For most of the chapter, however, we’re going to focus upon the protection of a new form of creative work: the computer program that controls all of the information inside a computer and makes it work its magic.

 

            What, exactly, is a computer program? In the words of the Copyright Act (as amended in 1980), a computer program is “a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result.” There are several different ways of categorizing computer programs. One traditional distinction is between object code  and source code. Object code is simply the program as it’s stored in the computer, the ones and zeroes, the ons and offs that tell the computer precisely what it’s supposed to do. This is the language that only a computer would love. Even a sophisticated computer programmer can’t really decipher all these ones and zeroes, and doesn’t usually write [77] programs directly in object code. Instead, the programmer usually writes in source code, a more abstract, or “higher,” programming language, such as Basic or C or PASCAL, and uses a computer program, a compiler, to convert the humanly intelligible source code into the computer intelligible object code.

 

            A simple programming example will illustrate the distinction. Let’s say I want to write a simple computer program that will ask the user for two numbers, and then respond by calculating their sum. Here’s how the program might look in an old version of Basic, a relatively straightforward programming language that unfortunately is hardly ever used anymore.

 

            10 Input A

            20 Input B

            30 Let C=A+B

            40 Print “The sum of your two numbers is ”; C

 

            Here’s what this program would accomplish on a computer that was set up to run Basic: at lines 10 and 20, the program would stop and wait for the user to input two numbers (that’s what the instruction “Input” is defined to do in Basic). It would then select a portion of its memory, give the memory locations the titles “A” and “B,” and store the two numbers in those locations. In line 30, the computer would go to the locations labeled “A” and “B,” take the numbers previously stored there, send them to the arithmetic unit for addition, and store the result in a portion of memory labeled “C.” In line 40, it would print out literally the words in quotation marks, followed by the number which is now stored in the portion of memory previously labeled “C.” On a computer screen, the running of the program would look something like this, with the underlined portions entered by the user:

 

            ? 12

            ? 23

            The sum of your two numbers is 35

 

            This Basic program is relatively easy for humans to understand and work with, because it’s written in source code. In fact, humans could work with this program, analyze it, and discuss it without ever entering it into a computer. As entered into any particular computer, [78] however, it’s reduced to computer instructions—or object code—that implement the above Basic program on a particular type of computer.

 

[78] Pre-pong.
In 1972, Magnavox released Odyssey, a “totally new dimension in television entertainment for the entire family.” One of the dozen games it could play was a Ping-Pong game that preceded the video arcade hit, Pong.

 

[78] Pac-Man.

[78] Super Mario Brothers. [78] Super Mario 64.

            Of course, computer programs get much more sophisticated than our example. Programs have been written that will allow word processing, graphic image processing, sound processing, and video processing on a sophisticated level. The cutting edge of the technology is usually reflected in the games. It was only in the early 1970s that the first commercial computer video game was made available to the [79] public. That game, Pong, was basically little more than a light blip that was moved around the screen to simulate a Ping-Pong game. What I think of as the second generation of computer games were games like Pac-Man, a program that could move simple imaginary characters around a relatively stationary two-dimensional grid. We might think of the third generation of games as the Mario Brothers type game, a still basically two-dimensional game, but with characters that moved around with much greater fluidity and sophistication: Mario walked across two-dimensional side-scrolling backgrounds, and interacted with those backgrounds. The fourth generation of games is represented by StarFox or Star Wars or Super Mario 64, in which the background scenery constantly changes so as to simulate a three-dimensional view of a virtual space. It’s hard to even imagine the amount of information—the hundreds of millions of operations per second—the computer has to process in each frame, at the rate of many frames per second, in order to create a realistic image that “puts” the viewer into the scene displayed on the computer monitor. The position of practically every pixel on the screen has to be recalculated, based upon the complicated three-dimensional contours of the “characters” and background of each scene.

 


How Do We Protect the Investment?

 

As the games and word processing and business programs have become more and more complicated, the investment in developing them has increased dramatically. Today, the development of a new computer program can cost millions of dollars and hundreds of thousands of programmer-hours. If you’re a developer who spends all that time and energy to bring your program to market, you certainly don’t want to have teenage kids send free copies of your latest version of MegaDoom to all their friends over the Internet, or have business executives distribute dozens or hundreds of copies of your latest version of MegaWord to all the employees over the company’s network, for free. You expect to get compensated, and you probably wouldn’t develop the software if you didn’t think you would be compensated.

 

            During the 1960s and 1970s, programmers began asking how they could protect their creations in the marketplace. One method that’s met [80] with some success is the development of physical copy protection systems. That is, you write the program so that it won’t work unless the user types in the proper password or serial number, or inserts a disk containing a legitimate copy of the program. The problem with using such physical protection is that many legitimate users are unhappy with the system. What happens if you forget the password, or if you don’t happen to have the original program disk when you want to run the program from your hard drive? Can you make backup copies to protect against the possibility that something goes wrong with your initial disk? What most software companies learned during the seventies and eighties is that legitimate users complained strenuously about overly burdensome copy protection systems; and so, most companies abandoned all but the most straightforward such systems.

 

[80] Can computer programs be patented?
Gary Benson and Arthur Tabbot sought a patent on a computer program that converted binary-coded decimal (BCD) numerals into pure binary numerals, and back again. Since computers use binary numbers, but people are better able to understand decimal numbers, or even binary-coded decimal numbers, such a program was absolutely essential to the running of most modern computers. If the patent were granted, it would be the computer program equivalent of a patent on transistors or on semiconductor chips.

            When their patent application was rejected by the patent office, Benson and Tabbot sued Robert Gottschalk, the acting commissioner of patents. The case went to the Supreme Court, which in 1972 held that the computer program was not patentable. As stated by Justice Douglas:

    A procedure for solving a given type of mathematical problem is known as an “algorithm.” The procedures set forth in the present claims are of that kind; that is to say, they are a generalized formulation for programs to solve mathematical problems of converting one form of numerical representation to another. . . .
       
            It is conceded that one may not patent an idea. But in practical effect that would be the result if the formula for converting BCD numerals to pure binary numerals were patented in this case. The mathematical formula involved here has no substantial practical application except in connection with a digital computer, which means that if the judgment below is affirmed, the patent would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself.

[81]      In recent years, the patent office has begun granting patents for computer programs, particularly ones that are part of methods of doing business. Some of these patents are controversial, and it remains to be seen how extensive such patent protection will be in the future.

             Some companies tried patent protection* of their programs, but there were several problems. The level of creativity required to get a patent is too high, and most computer programs don’t qualify. In any event, it usually takes about a year or two to get a patent, and in the software business, one year and you’re obsolete. And the Supreme Court, in a fascinating early case, held that a computer program—or at least the program in that case—was simply a mathematical algorithm, and not an “invention” subject to patent.

 

*Patent protection, see pp. 128 and 219.

 

            Some companies tried trade secret law.† If you only disclose your program to people who agree not to further disclose it, you can legally enforce their promises not to disclose. This worked for some computer setups, where the software supplier retained some control over the use of the software program. But it didn’t work very well for the new desktop computers, where copies of the software were sold to the users, and could be studied and sometimes deciphered by them.

 

†Trade secret protection, see p. 224.

 

            As we saw in some detail in chapter 1, copyright is an effective method of protecting creative works, [81] designed to encourage the making of such works by granting certain exclusive rights in them. Why not bring computer programs within the scope of copyright protection? After all, copyright has been around for a long time, and it seems to work. It’s easy to get a copyright, because you don’t have to go through any particular qualifying applications. The threshold of copyrightability is pretty low. All you have to do is create something original—that is, that you figured out yourself, and didn’t simply copy from someone else.

 

            There are some conceptual problems in using copyright as the form of legal protection to cover computer programs. Under the “works of utility” doctrine,* functional works are not the proper subject of copyright protection. Under that doctrine, how can copyright protect a computer program, the purpose of which is to run a machine? Computer programs are simply different from most creative works: while copyrightable works are generally designed for human communication, isn’t communication with a machine something of a different order?

 

*Works of utility, see p. 185.

 

            Some commentators suggested that computer programs didn’t fit neatly into any of the existing forms of legal protection of creative works—patent, copyright, trade secret, or anything else. What we needed, these commentators suggested, was some new form of protection designed specifically to protect computer programs.

 

[82] The argument for copyright protection: the Commission majority (1978).

    The cost of developing computer programs is far greater than the cost of their duplication. Consequently, computer programs . . . are likely to be disseminated only if:

                (1) The creator can recover all of its costs plus a fair profit on the first sale of the work, thus leaving it unconcerned about the later publication of the work; or
                (2) The creator can spread its costs over multiple copies of the work with some form of protection against unauthorized duplication of the work; or
                (3) The creator’s costs are borne by another, as, for example, when the government or a foundation offers prizes or awards; or
                (4) The creator is indifferent to cost and donates the work to the public.
               The consequence of the first possibility would be that the price of virtually any program would be so high that there would necessarily be a drastic reduction in the number of programs marketed. In this country, possibilities three and four occur but rarely outside of academic and government-sponsored research. Computer programs are the product of great intellectual effort and their utility is unquestionable. The Commission is, therefore, satisfied that some form of protection is necessary to encourage the creation and broad distribution of computer programs in a competitive market.
               
    The conclusion of the Commission is that the continued availability of copyright protection for computer programs is desirable. This availability is in keeping with nearly two centuries’ development of [83] American copyright doctrine during which the universe of works protectible by statutory copyright has expanded along with the imagination, communications media, and technical capabilities of society.  

[84] The argument against copyright protection: John Hersey's dissent (1978). 

    In the early stages of its development, the basic ideas and methods to be contained in a computer program are set down in written forms, and these will presumably be copyrightable with no change in the 1976 Act. But the program itself, in its mature and usable form, is a machine control element, a mechanical device, which on Constitutional grounds and for reasons of social policy ought not to be copyrighted.
                The view here is that the investment of creative effort in the devising of computer programs does warrant certain modes of protection for the resulting devices, but that these modes already exist, or are about to be brought into being, under other laws besides copyright; that the need for copyright protection of the machine phase of computer programs, quite apart from whether it is fitting, has not been demonstrated to this Commission; and that the social and economic effects of permitting copyright to stand alongside these other forms of protection would be, on balance, negative.
               
    The heart of the argument lies in what flows from the distinction raised above, between the written and mechanical forms of computer programs: Admitting these devices to copyright would mark the first time copyright had ever covered a means of communication, not with the human mind and senses, but with machines.  


           
To cut a long story short, all of this, by now, is actually ancient history. For the most part, we’ve opted to treat computer programs as creative works protectable under the copyright laws. This might not be obvious from the Copyright Act itself, since the Act does not list computer programs as one of the eight categories of works covered by copyright. But one of the categories of works is “literary works,” and literary works are defined to include works “expressed in words, numbers, or other verbal or numerical symbols or indicia.” During the congressional discussions leading up to the 1976 Copyright Act, Congress concluded that computer programs were within the definition of “literary works,” even though computer programs clearly had features that were unlike those of any literary works that had previously existed. So, whether or not it seems intuitive, and whether or not the statute seems clear, [82] computer programs are “literary works,” and, as such, they are protected by copyright.

 

[85] Another argument against copyright protection: Office of Technology Assessment (1986). 

    Although the copyright law adopts a uniform approach to protected works, not all types of information-based products are the same, nor can they be treated as if they were. A list of stock and bond prices, for example, differs from the musical score or a motion picture, and both of these are distinct from a computer program. In the case of stock prices, the value is in the information itself—the number of shares traded and the daily fluctuation in prices. The value of a musical score, in contrast, lies in the way it sounds to an audience—the appeal of its melody, rhythm, and harmony. And computer programs are valued for what they do—their effectiveness at performing a given task in a computer.
               This analysis has identified three types of copyright works:
    works of art, works of fact, and works of function. . . .  It is these differences that pose problems for the uniform application of copyright principles to all three categories. . . . In theory . . . copyright law cannot be successfully applied to computer programs. On the basis of the recommendations of the CONTU Commission, and without legislative debate, Congress determined that computer programs could be copyrighted as “literary works” under Section 102 of the 1976 Copyright Act. Although the issue of whether computer programs could or should be either copyrighted or patented was the subject of considerable legal controversy, it is now dormant. . . . [T]he courts have resolved these questions in favor of copyright protection for computer programs.

 

            Even as Congress passed the 1976 Copyright Act, it established the National Commission on New Technological Uses of Copyrighted Works (CONTU) to analyze, among other things, the appropriateness of the copyright approach. That Commission concluded that computer programs were and should be copyrightable, and recommended a minor amendment to the Copyright Act that Congress adopted in 1980.* The amendment created a “limitation” on the copyrightability of computer programs by specifically allowing users to make archival, or backup, copies of programs, and to make minor adaptations necessary to get programs to run on their particular machines. These exceptions would hardly have been necessary unless computer programs were otherwise already covered by the Act. So, while the statute still does not say in so many words that computer programs are copyrightable, the 1980 amendment, by carving out a limited exception, implicitly confirms that they are. To be sure, there was a well-reasoned dissent to the CONTU report, as well as periodic negative comments by various commentators over the years. But by now it’s a done deal: computer programs are copyrightable.

 

*1980 computer amendment, see p. 88.

 

            The only remaining question is not whether computer programs should be protected by copyright, but to what extent they should be protected. This is not an easy question, since computer programs are not obviously or easily amenable to analysis under traditional copyright principles.

 

             Let’s turn now to two of the leading cases in which the courts have proven themselves able to resolve the tough issues. We’ll look later at the ways in which Congress has felt it necessary to change the existing laws in order to accommodate the particular problems raised by protecting computer programs.

 

[83]

The Computer Copyright Cases

 

          Apple Computer, Inc. v. Franklin Computer Corp.

 

[83] The Apple prototype—using a Mostek 6502 microprocessor, four kilobytes of memory, and a cassette recorder for storage of programs and data—can now be viewed in the Smithsonian Institution.

[83] The first commercial Apple computer, with its own keyboard, monitor, disk drives and power supply.

One of the great success stories of the computer industry is that of Apple Computer, Inc., which developed the Apple II, one of the first more or less user-friendly desktop computers. The computer was assembled in 1976 by a couple of enterprising kids in their parents’ garage using standard parts that were readily available, designed around the new, cheap 6502 chip manufactured by Mostek. Within months, dozens of companies were developing software to run on Apple’s new machine. The big breakthrough for Apple came with the development of the first truly successful spreadsheet program, VisiCalc, in 1979. In its first year, VisiCalc was available only on the Apple; and it was so successful that it resulted in the sale of many Apple computers to businesses and individuals who wanted or needed to have VisiCalc.

 

[86] Two kids in a garage.
The two “kids” did have jobs. Steve Wozniak (affectionately known to his friends as “Woz”) was an engineer at Hewlett-Packard, and Steve Jobs worked part-time at Atari.  

                Woz had designed a single-board computer around the Motorola 6800 microprocessor. While the 6800 was fairly powerful, it was also fairly expensive at $175. In 1976, a small company called MOS Technology announced a pin-for-pin replacement for the 6800 called the 6502. Although it performed many of the same functions as the 6800, it cost only $25.
                Woz purchased a 6502 chip and immediately began work on a programming language—BASIC—for the new chip. Once the programming was complete, Woz redesigned his original 6800-based computer to accommodate the new 6502 chip. He and Jobs named the machine the Apple computer. Working from Jobs’s garage, Woz designed circuitry to connect a video monitor and keyboard to the computer. . . .  By the end of 1976, Wozniak had designed and built a much-improved computer, the Apple II.  The Apple II was a single-board computer like the Apple I, but the Apple II went several steps farther. The Apple II had the BASIC programming language built in, and it also had the ability to display text and graphics in color.
                                                                            -Les Freed


           
Shortly after Apple introduced its computer, Franklin Computer Corporation tried to duplicate Apple’s achievement. It designed its computer, the Franklin Ace 100, around the same 6502 chip that Apple had used. Franklin realized early on that it would be very difficult to compete with Apple and other existing computer manufacturers. When first introduced, a new computer would have virtually no software that it could run, while the existing computer manufacturers had the advantage of hundreds of programs already up and running. Franklin’s solution was to make its computer “Apple-compatible.” That way, Franklin could plug into the existing software and peripherals that were already designed to run with Apple computers. To achieve compatibility, Franklin thought it was not enough to simply use the same computer chip that Apple used; it also had to duplicate some of the Apple operating [84] system. That is, in developing the software to run its machine, Franklin copied various portions of the computer programs Apple used to run its machine.

 

            Without yet getting to the legal arguments, Franklin’s practical assessment may have been correct. It might be possible to write an operating system that would make Franklin’s computer look very much like an Apple. But unless it were precisely the same, it would be hard to guarantee that a program that ran on an Apple would run on a Franklin. For example, a program like VisiCalc, written to run on the Apple operating system, might be written on the assumption that certain specific operating instructions, like telling the printer to print a copy, or telling the monitor to move a pointer, were controlled by instructions at particular locations in the Apple operating system. These are known as “entry points,” the precise locations within an operating system where specific instructions can be found. If Franklin wrote its own operating system, then some of the critical entry points would be at different locations. So if you tried to run VisiCalc on a Franklin, and VisiCalc tried to make a call to the location where it expected printer instructions, and those instructions weren’t there, then your computer would crash.

 

            In order to achieve Apple compatibility, Franklin simply copied parts of Apple’s operating system. And, no surprise, Apple sued Franklin for copyright infringement of its computer programs. Apple’s argument was straightforward: Computer programs are copyrightable; copying and distributing are exclusive rights of copyright; so Franklin’s copying and distributing* of Apple’s computer programs are a copyright infringement.

 

*Copying and distributing, see p. 167.

 

            Franklin made several counterarguments. It could hardly argue that computer programs weren’t copyrightable, since that, at least, was clear from the 1976 Act and its legislative history. Instead, Franklin argued, more specifically, that these computer programs weren’t copyrightable for three somewhat overlapping reasons: (1) they existed in [85] the Apple computer in object code, and object code, as part of a machine, was a “work of utility,”* and therefore not copyrightable; (2) some of the programs were embedded in the ROM of the Apple computer, and ROM, as part of a machine, was not copyrightable; and (3) the programs were part of the operating system, and, as such, a functional part of the computer; as part of the “process, system, or method of operation,”† under section 102 of the Act, it was not copyrightable. Although the district court was sympathetic to Franklin’s arguments, the Third Circuit Court of Appeals ruled in favor of Apple on the copyrightability issues. The programs were copyrightable even though they were in object code, even though they were embedded in ROM, and even though they were part of an operating system. The court held that there was no basis for making the distinctions Franklin had urged: computer programs were copyrightable no matter what form they might take. Franklin ultimately settled the case for a large cash payment, undertook to develop a separate operating system, and then proceeded to fail in the marketplace.

 

*Works of utility, see p. 185.

†Process, system, or method, see p. 188.

 

            So what is a company like Franklin to do? It’s theoretically entitled to make a new computer using the same chip Apple used. But if the computer can’t be made Apple-compatible, then there will be no software to run on the new machine, and it will have a difficult time getting out of the starting gate. Apple thus gets an indirect benefit from a lot of programs that are not written by Apple at all, but by third parties to run on Apple machines. Well, the court was unsympathetic to this argument, stating that “Franklin may wish to achieve total compatibility with independently developed application programs written for the Apple II, but that is a commercial and competitive objective” which did not enter into the consideration of whether Apple was entitled to copyright in its operating system.

 

[86]

          Apple Computer, Inc. v. Microsoft Corp.

 

[87] The original Macintosh desktop, showing icons representing folders, applications, and documents, and windows showing and controlling open applications. 

[87] Microsoft Windows 3.0.

Apple’s next breakthrough was the development of the Macintosh computer in 1984. The Mac, as it came to be known, represented the first major successful use of a graphical user interface to run a desktop computer. Instead of typing in archaic lines of code that were gibberish to the uninitiated, the idea of the Macintosh was that major computer functions were represented by images, or icons. The user could run the computer by simply clicking on pictures. For example, to open a particular application, you didn’t have to type in the name of the program you wanted to run, but, using a pointing device known as a mouse, you just moved a cursor to a picture of the program and clicked twice. To open a file, or a work in progress, you just double-clicked on a visual representation of the file. The images were laid out on the screen in ways that allowed for their easy manipulation. You could store files in folders, and folders in folders, to organize your programs and data. You could click on “windows” that contained different groups of programs and data, and move the images from one window to another. You could open windows to work on them, and close them to clear them away from the screen while you worked on something else.

  

[88] The first graphical user interface.
Many of the principles of the graphical user interface had already been worked out by the Xerox Corporation on its Star computer. For various reasons, the Star computer was not itself successful, but it was the inspiration for both the Macintosh and Windows programs that have come to dominate the personal computer market.

            One of the people who admired this new graphical way of running computers was Bill Gates, the person who in his early twenties had developed the DOS operating system that ran the IBM computer, as well as most other desktop computers. In 1983, Gates’s company, Microsoft, developed the Microsoft Interface Manager as a graphical overlay to the existing DOS operating system. The problem was that the program didn’t work very well, particularly when compared with the Macintosh operating system that was introduced the following year. In 1985, Microsoft signed a licensing agreement with Apple, under which Microsoft was [87] allowed to incorporate certain features of the Macintosh interface into its new program. Although it took several more years to get it right,

 

            Windows 3.0, released in 1990, finally seemed to answer most of the problems with earlier versions of the program. Windows 3.0 was an immediate success, and eventually became the most successful computer program ever written. It made the awkward DOS operating system almost manageable. It also made Microsoft look like a monopoly, leading to one of the largest antitrust lawsuits since the breakup of AT&T—but that’s the subject of a book on antitrust.*

 

*Antitrust, see p. 222.

 

            Apple, by this time regretting its licensing agreement, claimed that the new version of Windows was not included within the earlier arrangement, and sued Microsoft for copyright infringement. This suit, however, was completely different from its earlier suit against Franklin. While Franklin had clearly copied portions of Apple’s actual computer program—its operating system—Microsoft did not copy any of the actual code from the Apple operating system. Microsoft basically wrote a completely different computer program that nonetheless copied the effect, or the “feel,” of a Macintosh.

 

            Microsoft raised several defenses, based upon some of the basic copyright principles we’ll discuss in Part Two. It argued that what it had taken was merely the “idea” of Apple’s graphical user interface, not any of the actual expression,† or code. It had taken some of the functional elements, but the functional elements, under the works of utility doctrine,‡ were not protected by copyright. Microsoft also argued that certain of the features in the Macintosh operating system were not original to Apple, but in fact owed their origin to earlier prototypes, particularly the Xerox Corporation’s Star computer. And many of the features, Microsoft argued, were covered by its earlier licensing agreement with Apple.

 

†Idea-expression, see p. 188.

‡Works of utility, see p. 185.

•Originality, see p. 127.

 

[88]      In a series of decisions in the federal district court in California, Apple basically got trounced. The court concluded that most of the elements of Apple’s graphical user interface were not protected by copyright, either because they were not original with Apple, because they were functional, or because they were covered by the original licensing agreement.

 

            There are other computer copyright cases, but they are basically consistent with the Apple cases. Copying of computer code generally constitutes a copyright infringement, but copying of the “idea” of a computer program does not. Of course, the problem in many cases is in deciding what constitutes “idea” and what constitutes “expression”—and on this point, courts sometimes disagree. Yet, despite the early doubts and the not-so-perfect match, courts have found that the basic principles of copyright—originality, works of utility, the idea-expression distinction, substantial similarity, and fair use*—also work just fine in determining the copyrightability of computer programs.

 

*Originality, see p. 127; works of utility, p. 185; idea-expression, p. 188;

       substantial similarity, p. 151; fair use, p. 190.

 

[80] Narrowing the scope of protection.
A leading case has held that, in determining whether one computer program infringes another, the test is not the usual one of “substantial similarity,” described in chapter 7, but a special three part test. First, in the so-called “abstraction” step, a court is supposed to “dissect the allegedly copied program’s structure and isolate each level of abstraction contained within it.” Second, in the “filtration” step, a court is supposed to “examine the structural components at each level of abstraction” to determine whether they are uncopyrightable as idea, “required by factors external to the program itself,” or “taken from the public domain.” (This filtration is supposed to reduce the program to the “golden nugget” of protectable expression.) And third, at the “comparison” step, the court is supposed to compare the infringed program to the “golden nugget” that survives the above analysis, to determine if there is any substantial similarity.
            This test, announced in the case of Computer Associates v. Altai, is about as abstract and metaphysical as the law gets. Even the court admitted that “[t]o be frank, the exact contours of copyright protection for non-literal program structure are not completely clear.” Yet the court clearly contemplates [89] that application of its test will result in a narrow range of protection for computer programs. As explained by the court, “If the test we have outlined results in narrowing the scope of protection [for computer programs], as we expect it will, that result flows from applying, in accordance with Congressional intent, long-standing principles of copyright law to computer programs.” The net result is that computer programs are protected under copyright, but not as liberally as for other types of copyrighted works. 


The Computer Copyright Amendments

 

Even while the courts were tackling some of the tough issues raised by the decision to include computer programs within the scope of copyright, Congress was also fixing some of the problems that it thought should not have to await judicial resolution.

 

          The Exception for Archival and Adaptive Copies
 

In 1980 Congress passed an amendment to the Copyright Act that created a limited exception to the rights of copyright in computer programs.† The amendment specifically allows purchasers of computer programs to make backup copies for archival purposes, and [89] to make adaptations necessary to get the program to run on a particular machine. Of course, if the computer program comes with some sort of physical copy protection system, it might not be possible for the user to make such copies; but the making of such copies, for the purposes specified in the amendment, would not otherwise violate the copyright law.  

 

†1980 amendment, see p. 82.

 

[89] A limited exception for computer programs.

    Notwithstanding the provisions of section 106, it is not an infringement for the owner of a copy of a computer program to make or authorize the making of another copy or adaptation of that computer program provided:
               (1) that such a new copy or adaptation is created as an essential step in the utilization of the computer program in conjunction with a machine and that it is used in no other manner, or
               (2) that such new copy or adaptation is for archival purposes only and that all archival copies are destroyed in the event that continued possession of the computer program should cease to be rightful.
                                                                    -Copyright Act

[91] Reasons for the exception (subsection 1).         

    Because of a lack of complete standardization among programming languages and hardware in the computer industry, one who rightfully acquires a copy of a program frequently cannot use it without adapting it to that limited extent which will allow its use in the possessor’s computer. The copyright law, which grants to copyright proprietors the exclusive right to prepare translations, transformations and adaptations of their work, should no more prevent such use than it should prevent rightful possessors from loading programs into their computers. Thus a right to make those changes necessary to enable the use for which it was both sold and purchased should be provided.
             -Final Report of the National Commission
               on New Technological Uses
               of Copyrighted Works

 

          The Special Protection for Computer Chips

 

Back in 1984, before it was clear that copyright would or could be made to effectively protect computer programs, the computer industry convinced Congress to pass a statute creating exclusive rights in computer semiconductor chips. A semiconductor chip is defined in the statute as a product “having two or more layers of metallic, insulating, or semiconductor material, deposited or otherwise placed on, or etched away or otherwise removed from, a piece of semiconductor material in accordance with a predetermined pattern.” These chips are usually made of silicon crystal, and the patterns are etched into the chips in layers so that they interact like miniature electronic circuits, connecting miniature transistors (that is, circuits that control other circuits). The chips are made up of as many as a dozen or more layers: each layer is designed by making a large scale “mask work,” which functions something like a photographic negative in that it allows the making of copies. The images produced from the mask works are miniaturized and embedded one on top of the other to make the three-dimensional semiconductor chip. It’s these chips that form the guts of virtually all computers manufactured today. Their main feature is that they are tiny: a sophisticated chip like the Intel Pentium CPU chip measures only 1.5 square inches, yet contains 3.1 million transistors.

 

[90] The first microprocessor chip, the Intel 4004, containing 2250 transistors on a single chip.


[Photo of engineer hunched over image,
looking through magnifying glass.]

[90] An engineer working on a mask work.

[90] “Mask Works.”
These mask works are more than just the blueprints for a semiconductor chip: they are the “negatives” that are used to actually make the chip. Engineers shine light through the mask works as they build up the layers that make up the chip. Shown here are the mask works from the Intel 4004 chip.

[90] The final product: the 4004 chip (detail). [The image shown here is the complete chip, not just a detail.]

 

            The problem, as far as the computer industry is concerned, is that it’s relatively easy—or at least in 1984 it was relatively easy—for someone [91] to take a semiconductor chip, effectively strip away the layers, and use the chip as the model for making a limitless number of copies. So, what takes months or years, and maybe millions of dollars to develop, can be copied and mass-produced in a matter of weeks or days. Congress understood immediately that such practices, if widespread, would destroy the computer industry. Its response was the Semiconductor Chip Protection Act of 1984, which granted exclusive rights to the developers of semiconductor chip products.

 

[92] Separate protection for mask works and semiconductor chips.

    The Committee decided that the formidable philosophical, constitutional, legal and technical problems associated with any attempt to place protection for mask works or semiconductor chip designs under the copyright law could be avoided entirely by creating a sui generis form of protection, apart from and independent of the copyright laws. The new form of legal protection would avoid the possible distortion of the copyright law and would establish a more appropriate and efficacious form of protection for mask works. Rather than risk confusion and uncertainty in, and distortion of, existing copyright law as a result of attempting to modify fundamental copyright principles to suit the unusual nature of chip design, the Committee concludes that a new body of statutory and decisional law should be developed. It should be specifically applicable to mask works alone, and could be based on many copyright principles, and other intellectual property concepts; it could draw by analogy on this statutory and case law framework to the extent clearly applicable to mask works and semiconductor chip protection, but should not be restricted by the limitations of existing copyright law.
                                                    -House Committee Report

            The new law was based loosely upon copyright principles. For example, only original works were protected, borrowing the lesser standard of originality* from copyright, rather than the higher standard of novelty from patent. Some features were different from copyright. For example, the period of protection was only ten years, much less than in the case of copyrighted works generally.† Although the act was codified in chapter 9 of title 17 of the U.S. Code, right after the Copyright Act that comprised chapters 1 through 8, it was decided that semiconductor chip products should be treated separately from copyright. There were at least two reasons for doing this: first, it was thought that separate treatment would avoid possible distortions of basic copyright principles to make them fit the new type of work. And second, if such protection were included under American copyright law, it might trigger U.S. treaty obligations‡ to give equal protection to foreign creators of such works even though their own countries didn’t grant similar protection to United States citizens. By keeping the law separate, the U.S. could withhold protection for foreigners, or could negotiate to get other countries to specifically pass reciprocal laws.

 

*Originality, see p. 127.

†Duration of copyright, see p. 205.

‡International treaties, see p. 241.

 

            Curiously, the Semiconductor Chip Protection Act has turned out not to be all that important. Registrations for mask works were below expectations, and in the first ten years after its passage, there was only one reported case brought under the act. Why should this be? To some extent, the evolving technology of semiconductor chips may have made it more difficult for others to make copies anyway. Or maybe it’s just that copyright protection of computer programs has developed so effectively [92] and so rapidly that semiconductor chip protection just isn’t that attractive anymore. If you manufactured such a product, would you be more interested in the ten-year protection of your chip, or the ninety-five-year protection of the contents of the chip? (Actually, there’s no need to choose; both types of protection are granted.)

 

            In any event, while the Semiconductor Chip Protection Act was initially praised as a model for “sui generis” statutes—separate statutes for particular types of creative works—such special protection has turned out not to be nearly as effective as the protection afforded under the more general principles of copyright.

 

          The Computer Software Rental Amendment, 1990

 

The “first sale” doctrine* provides that, once a purchaser has bought a copy of a copyrighted work, the purchaser normally may resell, rent, or otherwise dispose of that particular copy of the work, so long as the purchaser doesn’t make further copies. This caused problems in the music industry, because, it was thought, the rental of records or CDs inevitably led to the home copying of those works. Congress responded to the problem in 1984 by making an exception to the “first sale” doctrine for CDs and sound recordings:† while purchasers of such works could resell them, purchasers were prohibited from renting them.

 

*First sale doctrine, see p. 167.

†1984 audio rental amendment, see p. 48.

 

            In 1990, the computer software industry was successful in convincing Congress that it should be afforded similar treatment to sound recordings. If companies were allowed to purchase computer programs and rent them to the public, then the public would very likely use such rentals to make copies of the computer programs that would displace sales. For example, if you could rent a copy of Adobe Photoshop for, let’s say, $20, and it costs about $500 to buy a copy, then you might be awfully tempted to simply make a copy from the one that you’d rented. To prevent this, Congress [93] amended section 109 of the Act to extend the exception to cover computer programs as well as sound recordings. “[N]either the owner of a particular phonorecord nor any person in possession of a particular copy of a computer program . . . may, for the purposes of direct or indirect commercial advantage, dispose of, or authorize the disposal of, the possession of that phonorecord or computer program . . . by rental, lease, or lending . . . .”

 

            But, you say, your kids rent copies of the latest version of Super Mario or Final Fantasy, or other Nintendo or Playstation video games, from the local video rental store. That’s because of an exception to the exception. The exclusive rental right for computer programs created by section 109 does not apply to “a computer program embodied in or used in conjunction with a limited purpose computer that is designed for playing video games. . . .” What does that mean? It means that a video store can rent video games that run on dedicated systems, like the Nintendo system, from which they cannot normally be copied. If your kids rent such games, they can’t easily make copies of them, and so the rental does not lead to a copy that displaces a sale.

 

Look What I Can Do Now!

 

So far in this chapter, we’ve been looking at how copyright has come to embrace and protect computer programs, the new type of creative work that makes computers do what they do. But computers have raised a much broader issue for copyright.

 

            As we’ve seen in the previous chapters, copyright protects books and magazines, music and records, plays and movies, and now, computer programs. We'll explore many other types of protected works in chapter six. With the advent of the computer, however, all of these works are being digitized to make them accessible to the computer. We create our literary works on computers, using word processors to manipulate the words and letters as part of the creative process, and to store vast amounts of information on new media such as disks and CD-ROMS. Using image scanners and digital cameras, or plugging our video recorders and camcorders into our computers, we capture pictorial and photographic images, even moving images, in pixels that can be displayed and manipulated on a computer screen and stored in ever smaller amounts of space. Using [95] musical interface devices (midis), we “sample” music electronically, reducing sound itself to pure information that can be captured, stored, manipulated, and copied with remarkable consistency and fidelity.

 

            It’s the digital revolution. If you could look inside the memory of a computer, you’d see all the electrical impulses, stored in a binary code of on/off switches. If you saw these electrical or magnetic impulses, you couldn’t immediately tell whether what you saw stored in any part of the computer was words, images, sounds, or computer programs. But the computer itself keeps track of what type of information is stored where, and can easily transform the information back into the words, images, and sounds that make sense to us as humans.

  

            With the computer, we can create works, copy them, distribute them, perform them, and transform them. As we’ll see in chapter 7, these are precisely the exclusive rights that are supposed to be protected by copyright. Clearly something has got to change in how we think about these creative works.

 

            We’ll get back to some of these basic questions. But first, we’ll look at the Internet, the popular new way of distributing the vast amounts of information we’ve accumulated and digitized.

 

 [94]  My favorite early CD-ROMs.
Below are some of my favorite early CD-ROMs, illustrated at less than half actual size. They each contain up to 650 megabytes of information. The first contains the complete works of Shakespeare, with so much room left over that it includes the entire works twice—once in Elizabethan English and once in contemporary English.
            The second contains a map of virtually every street in the United States, viewable at varying scales. I was able to find the very street where I grew up in Paragould, Arkansas! Street maps are now available basically for free on the Internet.
            The third CD-ROM contains the complete Microsoft Encarta multimedia encyclopedia from a few years ago, one of many encyclopedias now on CD-ROM that contain the equivalent of dozens of volumes of their book counterparts. (The more recent versions are on two or more disks, because of extensive use of interactive media that take up more space than does text.)

 

 

[95] A favorite graphics program.
Using Adobe Photoshop, I scan old family pictures into my computer, touch them up or alter them, and then print them out in improved form. In this picture of my grandmother (third from left) and three of her sisters, I have enlarged and enhanced the image. You may also notice that the car has disappeared with a few waves of the magic Adobe wand. (It’s debatable whether cutting out my grandfather, and the car—from which my father can date the picture precisely—is an improvement, but the point is that you can do what you want.) If you don’t have the equipment for doing this yourself, your local photography store will be glad to correct pictures, even erasing old friends, putting together new ones, or, as advertised in one local store, enlarging the size of the fish you catch.

 

Conclusion

 

It’s interesting to contrast the legislative response to new technologies in the context of music, television, and computer programs. As we saw earlier, Congress adopted highly technical regulatory schemes to handle some of the big copyright issues raised by digital technology in the music industry. Relatively few new laws were required for motion pictures and television. In this chapter, we’ve seen that although Congress has adopted some amendments to handle specific problems [97] in administering copyright in computer programs, for the most part it’s the courts that are deciding the major computer copyright issues, and they’re doing so on traditional copyright principles, not the legislation directed specifically at protecting computer programs. This may seem surprising, given the number of commentators who had originally argued that computer programs were just too new, too different from traditional notions of copyright, to expect a workable solution to be developed.

[97] International context.
Article 10 of the international Agreement on Trade-Related Aspects of Intellectual Property (1994) provides that “Computer programs, whether in source or object code, shall be protected as literary works under the Berne Convention.” For more on the 1994 agreement, see p. 244.

            

            Yet a reasonable balance has been struck, and it seems to be working fairly well. Indeed, the United States has been so successful in developing standards for protecting computer programs that it’s even begun exporting the concept. While many European and other countries had originally balked at the notion of treating computer programs as within the subject matter of copyright, the United States has recently prevailed upon most of the rest of the world to protect computer programs under copyright.

 

[96] Another favorite program.

Another favorite program is Kai’s Power Goo, with which a user can manipulate pictures in all sorts of fascinating ways—again, all because the information is stored in digital form that can be readily altered. With apologies to the artist, here is a relatively mild distortion of the Mona Lisa. Many wilder variations are available using the tools pictured here—grow/shrink, move, smear, smudge, nudge, mirror toggle, smooth, and ungoo which undoes any of these effects. Using another set of tools, the user can bulge, twirl, rotate, stretch, spike, or “static” the image, or unwind any of these effects. Don’t ask what these tools do; finding out is half the fun. Using the sliding switch at the left, the user can ease the effects in and out of the image. Or the user can record a succession of different images at the bottom, and the program creates a smooth “goovie” moving from one distortion to another. All of this is of course good clean fun, since Leonardo da Vinci is dead and the Mona Lisa  is in the public domain. But what if the artist is alive and the original work is still copyrighted? Stay tuned for further developments.

 

 

 

[96] 3-D rendering program. Using Ray Dream Studio, my daughter, Claire, constructs three-dimensional models in which the instructions for reproducing characters, objects, and background are stored as basic information. By relatively easy manipulation, the characters can be moved around, seen from different perspectives, or strung together over time to make animated movies. Shown here is “Trish,” a cartoon cat.

   
        
It’s my own opinion that the treatment of computer programs, leaving the tough issues to be developed by the courts based upon general copyright principles, has led to a more satisfying and nuanced resolution. As we face the copyright issues raised by other new technologies, such as the Internet, I would argue that the preferable approach is not to run to Congress for quick fixes to every tough question, but, if possible, to allow a little time for the courts to work out the issues.

 

Return to Contents

Go to Chapter 5: The Internet

Permission, Limitations, and Format