The effects of the Lotus v. Borland case Linda Shan INTRODUCTION Computer software is a unique way of expressing ideas, somewhere in between art and invention. This fact makes it difficult to determine what parts of programs can be "owned." Computer software has the unfortunate characteristic of being able to serve both as an expression of an idea and a functional piece of an idea. Copyrights protect the expression of an idea, not the idea itself. Issues concerning the copyrightability and infringement policy for computer software are being determined in the courts today. The outcome of these cases can have an enormous impact on the computer industry. The law is constantly changing. The direction in which it moves depends on the needs and demands of society as well as the opinions of the judges who shape the laws. One controversial case, Lotus v. Borland may help pave the way to a standard regarding intellectual property rights. Unfortunately, there are fundamental problems with the Lotus v. Borland ruling. To get a better understanding of the flaws in this case and to gain perspective on the unfavorable direction that this case leads us, it is necessary to first examine Lotus v. Borland in detail, looking at the reasons for the judge's decision, the flaws in his reasoning, and the ramifications of the ruling. BACKGROUND The copyright law was designed to protect an original work in a set medium. It covers the entire body of the work, and covers only the expression of an idea.It does not cover ideas, procedures, processes, systems methods of operation, concepts, or principles. It can be abridged by literal or non-literal copying. However, copyright infringement can also be legally avoided through independent creation. In other words, it takes both access and similarity to result in a copyright infringement. How does copyright apply to computer software? Well, there is the crux of the problem. Computer programs are in essence functional. Many times one cannot separate the expression from the idea, so it is difficult to determine what is copyrightable in software.According to the copyright law professors who wrote the amicus brief concerning the Lotus v. Borland case, "The highly functional nature of programs and the results they generate that computer programs makes programs an unusual subject matter for copyright protection." (Amicus Brief by copyright law professors, p. 23). The brief also states that "Since functional writings, by definition, contain not only abstract ideas, but also functional elements, such as processes, procedures, systems, or methods of operation, the scope of copyright protection available to such works tends to be narrower than for artistic and fanciful works."(Amicus Brief by copyright law professors, p. 40). The ultimate questions is the extent of the non-literal coverage of computer software. There is a delicate balance between what the courts determine to be "legal" and what benefits society and the computer software industry. Unfortunately, there is a significant divergence of opinion between the courts and the computer science community. To put the Lotus v. Borland case in perspective, it is first necessary to take a brief look at the history of the software copyright. In the court case, Whelan v. Jaslow, 1986, Whelan claimed that Jaslow infringed on a program written in EDL, even though the Jaslow program was created in Basic. This claim implies that the medium of expression is irrelevant, that only the expression of the idea is important. This is consistent with the non-literal aspects of copyright. In this case, the court decided that they "must determine whether the structure (or sequence and organization) of a computer program is predictable by copyright, or whether the protection of a copyright extends only as far as the literal computer code." (6.095 handout). Obviously, copyrights would be useless to the computer industry if they only protected the literal code. There are often many ways to implement the same idea, with the same resulting expression. The court believed that the deciding factor was where is the value of the program existed. The court ruled that there was an infringement because the programs were similar in many ways, such as functionality and structure. Computer Associates v. Altai differs with this ruling. In the Altai case, the court decided to reject the precedent set by the Whelan case, stating that it was "inadequate and inaccurate." This court ruled that the "structure, sequence, and organization" test was inaccurate in describing computer programs. The argument here was that programs behave, and copyright does not cover behavior. In addition, the words "structure", "sequence", and "organization" cannot be used interchangeably when speaking about computer programs. Changing any one of these could potentially change the function of the program. The court ruled on an "Abstraction-Filtration-Comparison" test for determining "substantial similarity." (6.095 handout). In the Lotus v. Paperback Software and Mosaic Software, 1990, Judge Keeton decided in favor of Lotus. Paperback and Mosaic both came out with spreadsheet pro grams that displayed extreme similarities to Lotus's 1-2-3. These cases have been dubbed the "look and feel" cases. The reason is that the courts basically determined that a company has the right to copyright the "look and feel" of its user-interface. Their success in this case prompted Lotus sue Borland as well. THE LOTUS V. BORLAND CASE RULING Lotus sued Borland over the similar user interface of the Quattro programs. Lotus argued that Borland copied the expressive components of Lotus 1-2-3, including it's "menu commands," "menu structure," "long prompts," "keystroke sequences," and "macro language." The U.S. District court ruled that the claims made by Lotus were copyrightable, and that the Quattro programs put out by Borland were similar enough to infringe on the copyright that Lotus had on the 1-2-3 interface. The outcome of this case caused quite an uproar in the copyright law and computer software industry communities. Not only do they disagree with the judge's ruling, they also believe that such a ruling could lead to disaster for the computer software industry. FLAWS IN THE RULING Both parties, copyright law professors and the computer software industry, believe that the reasoning behind Judge Keeton's ruling was flawed. According to the amicus brief of copyright professors concerning the Lotus v. Borland case, the Computer Associates v. Altai ruling embodies the best method for determining the copyrightability of software. With this method one would first look at the hierarchy of abstraction for the entire body of work. The next step would be to filter out the things that are not original. The Lotus case would have ended here with the court determining the unoriginal nature of the spreadsheet user interface. Step three in this analysis would be to consider the similarity between the copyrightable material and the work of the defendant. On the other hand, the method used by Lotus v. Borland case first examines the hierarchal nature of abstraction for each piece of the program. Step two involves deciding the copyrightability based solely on the functionality of the code. Step three of this test would be to consider the creativity of the product. However, the court ruled in this case that the fact that the expression was intertwined with the functionality of the work did not make it uncopyrightable. Copyright law explicitly states functional works cannot be protected under copyright. The test for creativity used by Judge Keeton was not originality, but rather, whether or not there were viable alternatives to the work. This is an illogical assumption, for something that can be implemented differently does not imply that every different method of implementation is creative or copyrightable. COMPUTER LANGUAGES, COPYRIGHTABLE? The macro language of Lotus 1-2-3 should not be copyrightable. A good example that contradicts the court's ruling is the "short-hand system" case. In this case, the shorthand system and the vocabulary of the system were shown to be beyond the scope of protection provided by copyright. This precedent conflicts with Judge Keeton's decision to uphold Lotus' copyright privileges concerning the macro language. The main threat that Judge Keeton's ruling poses to the computer community is that it implies that computer languages are copyrightable. Although it seems ridiculous that such an idea could be considered, the Lotus v. Borland case paves the way for copy righting computer languages. The macro language that was shown to be covered under copyright. The macro language enabled the user to define keystroke sequences by input ting a sequence into a spreadsheet call and assigning a label to the call. Judge Keeton cited that "the fact that a form of expression takes on functional character does not remove it from the protection of copyright." (Lotus v. Borland ruling, p. 22). Basing his ruling on this belief, Judge Keeton ruled that Borland's use of the macro language infringed upon Lotus's copyright privileges. According to the Amicus Brief by computer scientists, the Court used the words "procedure" and "operation" at least fourteen times in describing the macro language. This indicates that there was a serious flaw in the court's reasoning. Either Keeton disregarded Section 102(b) of the Copyright Act, or he found something in the case that would overshadow this line of reasoning. The reasoning that the court gave for determining the copyrightability of the macro language was the innovation and creativity behind the idea for a macro language. The innovation behind an idea does not necessarily make an idea copyrightable. Although the value of the program does lie within its function, it does not indicate that this is reason enough to prove copyrightability. The implications of such a ruling are hazardous to the computer software industry. Computer languages are used by programmers not only to communicate with each other, but to also communicate with the machine. The language is clearly the means of expression, not the expression itself. The purpose of the copyright law was to protect the expression of an idea. For instance, it is undisputed that the English language itself, meaning English words, are not copyrightable. However, a creative sequence of words that express an idea, such as a poem or novel, is copyrightable. It is expressly stated in the Copyright Act that any "idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is embodied" is not copyrightable.(Amicus Brief by computer scientist). The same general idea applies to computer programs. Programming languages are functional. They serve as the medium for expression, not as the expression itself. If computer languages could be copyrighted, this would clearly threaten the industry's ability to have a software compatibility the free use of common languages. The amicus brief provided against the copyrightability of computer languages gives some great examples of the dangers of copyrighting computer languages. For instance, computer programs are generally written in easy-to-program, high-level languages as "source code" which translated into a low-level machine language as "object code", via the means of a compiler. This makes programming much easier and more powerful for programmers. However, under the Lotus v. Borland ruling, the "source code" would be owned by the supplier of the computer's processor and they would also own the "object code". There fore, by this same line of reasoning, Lotus may not own the rights to 1-2-3, the company that wrote the language would. (Amicus brief of computer scientist). This circular reasoning is inconsistent and hinders the development of a strong computer software industry. THE "LOOK AND FEEL" CONTROVERSY Another issue that concerns the computer science community and the copyright professors is the copyrightability of the "look and feel" of a software package. In Baker v. Selden, a case concerning ledger sheets in accounting books, the functional items of the presentation of information was ruled uncopyrightable. Selden sued Baker because its ledger sheets were similar in selection and arrangement of columns and headings. The key issue here was whether someone else could "make or use similar ruled lines and headings, or ruled lines and headings made and arranged on substantially the same systems without violating the copyright." (Amicus Brief of copyright law professors, p. 5). The Supreme Court decided that Selden did not have the right to the system, no more so than "the copy right on a book about the composition or use of medicines would give its author an exclusive rights to the medicinal compounds or uses of them discussed in the book." (Amicus Brief of copyright law professors, p.6). Following this line of reasoning, the elements of the Lotus 1-2-3 user-interface, such as "menu commands", "menu structure", "long prompts", and "keystroke sequences" cannot be copyrighted, contrary to Judge Keeton's ruling. Copyrighting the "look and feel" of a software package is detrimental to society. Certain user-interfaces become popular because they are easy to use and functional. The user becomes accustomed to the "look and feel" of the software package and this makes the use of computer comfortable. According to Pamela Samuelson, user interfaces are "the most visible aspects of software products and the key to their usability."(Samuelson, "Why the look and feel of software user interfaces should not be protected by copyright law.", p. 563). The legal term, or rather the recognized term that describes "look and feel" is "total concept and feel." It is obvious that such the "total concept and feel" of a product should not be copyrightable, after all, Copyright law prohibits the protection of concepts. One possible reason that the "look and feel" of a package may have been considered as copyrightable by the courts is that many judges do not recognize the functionality of these interfaces. They often mistake these interfaces for being purely artistic expressions. For instance, in the Broderbund case, the judge stated that the interface was "artistic, explicitly denying that the interface had any functional character." (Samuelson, p. 565). User interfaces function much as the steering wheel of a car does. Although there are many ways of implementing the control of the motion of a car, this is the most intuitive, and therefore, became the standard. No one would consider copyrighting such an idea simply because it is functional and not implemented through text. However, since computer software's medium is text subjects it to the review of copyrightability. The fact that the "look and feel" could be copyrighted, forcing each computer company to come out with a unique user interface would be as disastrous as if each car manufacturing company made cars had controls that functioned differently. People would become accustomed to one type of car, and it would be difficult for them to switch to someone else's car. Similarly, the software would be incompatible. People would become dependent on a particular user interface, and being creatures of habit, would continue to only use the software packages put out by that company. This would make it exceedingly difficult for new software companies to enter the industry. CONCLUSION The law is constantly evolving. This change is based on the needs of the people. Lawmakers cannot control the interpretation of the law. This is something that is worked out with time and experience. Even then, the law is never perfect. Technology is growing at a phenomenal rate, and the law is running to catch up. There must be a balance between the concerns of the computer software industry, the end-user, and the copyright lawyers. Perhaps there should be more computer scientists involved in the court rulings concerning software copyrights. For example, in the Computer Associates v. Altai case, Randy Davis, a computer scientist from MIT, pointed out that the structure, sequence, and organization of a computer program were being interpreted incorrectly by the courts due to its lack of knowledge concerning computer programming. Maybe Keeton was unaware of the ramifications of his decision. After all, the final goal of the law is to benefit society. The Lotus v. Borland case not only is inconsistent with the existing copyright laws, it is also detrimental to the development of convenient, compatible computer software. BIBLIOGRAPHY Ableson, Hal, 6.095 "Ethics and Law on the Electronic Frontier" class handout. Judge Keeton's ruling in Lotus Development Corporation v. Borland International US District Court, District of Massachusetts, July 31, 1992. Amicus Curiae Brief by the copyright law professors in the appeal of the Lotus v. Borland decision, filed in the US Court of Appeals for the First Circuit. Amicus Curiae Brief by computer science professors in the appeal of the Lotus v. Borland decision, filed in the US Court of Appeals for the First Circuit. Samuelson, Pamela, "Why The Look and Feel of Software User Interfaces Should not be Protected by Copyright Law." Communications of the ACM, May 1989.