Early in the development of the art of software writing, the originators feared traditional forms of protection such as patents and copyrights would not give them sufficient protection and relied upon trade secret law. It soon became clear, however, that this was not the best solution to the problem since to establish trade secret protection, one must take some steps to impose a confidential relationship with those who have access to the secret. In a free flowing industry such as the computer software business, this is difficult.

Attention, therefore, has returned to the traditional protection of patents and copyrights. As discussed above, many computer-related inventions are patentable. However, patent protection cannot be obtained for inventions in this field that do not meet the current test for patentable subject matter. Furthermore, to be patentable the program must be “not obvious”. This can be a difficult requirement to fulfill. Furthermore, patent protection requires a rather prolonged examination by the Patent Office before any rights arise and is thus not the ideal way of dealing with copyists who may be extremely quick off the mark.

A basic principle laid down by the Supreme Court in the case of Baker v. Selden in 1879 is there is no infringement of copyright in using or copying something if such copying or use is the only way for putting a particular idea into practice. [13] Copyright protection is available only for a particular expression of an idea, not for the idea itself. (The case related to a book explaining a new system of bookkeeping and including certain blank forms). In that case, it was held that for the position to be otherwise, there would be a grant of a monopoly “when no examination of its novelty had ever been officially made which would be a surprise and a fraud upon the public.” The protection of ideas was to be a matter of Letters Patent not copyright. Indeed Section 102(b) of the Copyright Act 1976 specifically provides that:

When a computer program is written out on a piece of paper, it is quite clear that as long as the requirements noted above are complied with, copyright exists in this work from that moment in the same way as it would in respect of any other literary work. The first problem the courts had to grapple with was whether copyright law could be extended to cover computer programs that exist merely in magnetic or electric form or as specific circuits etched on to a silicon chip.

The United States Copyright Office has provided for the registration of computer programs since 1964. However, such registrations were accepted on the basis it was the duty of the Copyright Office to give applicants the benefit as to whether computer programs were protectable by law. Indeed, prior to the Copyright Act of 1976, court decisions and the opinions of the leading writers in the field were widely split on the question of whether computer programs were entitled to copyright protection and, if so, to what extent protection was to be given.

The 1976 Act itself was not very helpful in resolving these doubts, although as noted above, Section 102 states that copyright protection exists in works that are fixed in any present or future tangible medium even if it can be reproduced only by means of a machine or other device. On the other hand, Section 117 of the 1976 Act as originally enacted specifically stated that it was not the intention of the Act to change the pre-existing law in respect to protection of computer programs. The Act, however, did not indicate what the pre-existing law was. This problem was finally resolved in 1980 by the passage of the Computer Software Copyright Act 1980 which repealed original Section 117. This repeal removed any doubt as to whether the broad definition of Section 102 extends to software. Furthermore, a new Section 117 was enacted which provided that the owner of a computer program had a limited right to copy or adopt that program if this was necessary to ensure that the program could be used in a particular computer. [15] Indeed, the new Act even provided a definition of a computer program as “a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result.”

The first generation of cases to come before the courts mostly involved actions for copyright infringement in ROMs (Read Only Memory). Three main lines of argument were used by defendants to try to resist actions brought by the originators of a computer program which have been incorporated into the defendants ROMs. The first was that ROMs are utilitarian (the sub-argument of this being that even if not all such ROM’s are utilitarian, at least those whose function is to work the computer itself are). The second was the program as it appears on a ROM is in object code form and therefore not intelligible to human beings, it falls outside the definition of “copy” in the Copyright Act. The final line of argument used was it is against public policy to give computer manufacturers copyright protection for their software because such protection would hinder competition in the development of new computers.

The courts have not been impressed by arguments that computer software is purely utilitarian when they are presented in the context of direct copying of a particular program. [16] For example, in an early case the Third Circuit Court of Appeals rejected the argument that the use of a computer program encoded in a ROM was utilitarian. It also rejected an argument that since the coding was in object code it was not a copy intelligible to humans and, therefore, outside the Copyright Act protection. [17] The latter point was established by analyzing the definition of “copy” in the Copyright Act. Section 101 of the Act defines copies as:

The court held that since Congress had chosen to define “copy” broadly in this manner, it would be perverse to seek to interpret this broad language in a manner, which would severely limit the copyrightability of computer programs, by restricting this protection to the text and not to duplication of the program on a fixed silicon chip. The court further rejected an argument that ROMs were utilitarian objects and, thus, not subject to copyright protection by observing that ROM’s were not the objects which had to meet the test of copyrightability. Only the original program had to be copyrightable. Once copyright existed in that program, the only question remaining was whether a ROM constituted a copy of it. As noted above, the court had no difficulty in deciding that this was the case. A similar conclusion was drawn by the District Court in the Northern District of Illinois. [18]

One might have expected defendants to be more effective in utilizing the idea-expression dichotomy running through the United States copyright law when the defendant has done something more than copy the source or object code expression of the copyright owner’s program. With a few exceptions, however, the courts have still tended to favor the copyright owner. One case where the argument was successful, however, was in Q-Co. Industries, Inc. v. Hoffman.[19] In that case, the court denied a preliminary injunction to prevent the defendants selling a program that was inspired by one that was protected by copyright where the alleged infringer had taken a program designed to run on one type of hardware and had expended substantial effort to make it run on different hardware. The court noted that although the general structure of the two programs in question are very similar, they were written in different languages and employed wholly distinct algorithms. It concluded that the similarities “can be more closely analogized to the concept of wheels for the car rather than the intricacies of a particular suspension system ... it was the idea that was used rather than its expression. Therefore, copyright infringement has not been established.”

As can be seen from the cases discussed above, consideration of the meaning of “copy” was of importance in determining whether there was copyright protection for a given product, in copyright law unlike patent law the issues of what constitutes protectable subject matter and what constitutes infringement are often intertwined. Furthermore one should bear in mind the fact that unlike the situation with patents, there is no Court of Appeals with nation-wide jurisdiction for copyright cases and so different views may be taken in different circuits.

Once the basic issue of whether copyright protection could exist at all for computer software had been resolved the next issue, which is still with us, was to determine what exactly might be covered by copyright, particularly if the alleged infringer had not copied code line by line but had taken something less specific. The first group of cases to tackle this issue considered whether the “look and feel” of the two programs is the same and if the defendant had access to the plaintiffs program, then copyright infringement was likely to have occurred. Thus, if there was substantial copying of details of a program designed to fulfill a particular function on one type of hardware, in producing software to fulfill the same function on different hardware, the mere fact that substantial effort was involved in making the conversion or even that different programming languages were used might not avoid a finding of copyright infringement. This is particularly the case if what is shown on the screen is the same at each stage of running both the copyrighted and the allegedly infringing program or if similar errors appear in both programs. [20]