Protecting computer programs as compilations.

AuthorWilkins, Jon S.
PositionCase Note

Writing software requires both technical expertise and artistic creativity. A competent programmer must have a firm grasp of the intricacies of hardware organization, programming languages, and other basic principles of computer science. This base of knowledge supplies the constituent elements of software. Designing a program, however, involves more than competent engineering. Selecting, arranging, and coordinating program elements to construct a complete software work is a subjective process that requires imagination and creativity. Just as a composer creates a symphony by selecting and arranging underlying elements--from the specific notes played to broad choices of theme and orchestral composition--a programmer designs software by selecting and integrating programming techniques, problem-solving algorithms, data structures, and other software elements. Although programming also entails the more mechanical process of translating design into functioning code, this necessary step does not detract from the significant creative expression embodied in the design and overall structure of software.

Unfortunately, the leading test of software copyright infringement employed by courts fails to consider selection and arrangement expression in program structure. In Computer Associates International v. Altai, Inc.,(1) the Second Circuit established a three-part test for deciding software infringement cases.(2) Known as "abstraction-filtration-comparison" or "successive filtration"' this test requires courts to dissect programs and then evaluate, the copyright protection afforded each part. While this approach rests on a sound doctrinal foundation and provides a useful analytical framework for evaluating complex software, it fails to account for the creative authorship required to design program structure.

Most traditional programs contain such design-level expression as a significant creative element. Underprotection of these programs alone offers sufficient reason to reevaluate abstraction-filtration-comparison. The software industry's increasing reliance on a new technology known as object-oriented programming suggests even more strongly, however, that Computer Associates should be modified. Object-oriented programmers create large programs out of preexisting software building blocks, an approach that offers many advantages but tends to limit possibilities for creative expression to the selection and arrangement of software "objects." Although selection and arrangement expression is but one kind of expression present in traditional software, in many cases it will be the only protectable element present in an object-oriented program. Courts must consider this kind of expression in order to address the next generation of software litigation.

Copyright law's protection of works of "compilation" focuses on identifying precisely the selection and arrangement expression ignored by Computer Associates. The Supreme Court most recently addressed works of compilation in Feist Publications, Inc. v. Rural Telephone Service Co.,(3) and various lower courts have developed the principles elaborated in Feist. This Note argues that courts should apply these principles to protect design-level expression in software. Part I provides a brief overview of relevant copyright and software issues. Part 11 explores the appropriateness of compilation protection for computer program design. Part Ill develops in more detail how compilation doctrine applies at each phase of abstraction-filtration-comparison. Part IV demonstrates the urgency of this doctrinal modification by discussing the importance of compilation expression in object-oriented software.

  1. OVERVIEW: PUTTING ABSTRACTION-FILTRATION-COMPARISON IN COPYRIGHT CONTEXT

    1. Basic Copyright Principles

      Cases of nonliteral(4) infringement of software focus on one issue: whether the copying alleged is legally sufficient to constitute infringement.(5) Not all copying of a copyrighted work is illegal. For example, copying public domain elements, or copying only a small portion of a copyrighted work, does not infringe a copyright.(6) Determining whether copying constitutes infringement turns instead on a comparison of the copyrighted and accused works that requires a court to determine "the extent of similarity which will constitute a substantial and hence infringing similarity."(7)

      The issue underlying substantial similarity analysis is copyright law's distinction between "ideas" and "expression": Original expression is protectable, ideas are not.(8) The example of Romeo and Juliet demonstrates this principle's basic operation. A work that copies the play's dialogue and stage directions line-for-line clearly would infringe Shakespeare's copyright because it would replicate his original expression.9 A play that merely employs similar general ideas--for example, another play with a plot involving young lovers whose families bitterly oppose their relationship--would not infringe, because the only similarity between the two works would be at the level of ideas. As a matter of law, similarity only as to ideas does not reach the "substantial" level of similarity required for infringement.(10)

      Although the distinction between specific dialogue and general themes in this example seems clear, protected expression and unprotected ideas in fact exist along a continuum. Consider a comparison of Romeo and Juliet and West Side Story. The literal expression of the Broadway production differs from Shakespeare's play. Yet the similarity in plot, characters, and sequence of events extends far beyond merely sharing a common theme. At some point along this continuum, a court must draw the line between protected expression and unprotected ideas.(11)

      In a famous passage from Nichols v. Universal Pictures Corp.,(12) Learned Hand articulates the "abstractions" test used by courts as the general approach to this problem:

      Upon any work . . . a great number of patterns of increasing

      generality will fit equally well, as more and more of the incident is

      left out. The last may perhaps be no more than the most general

      statement of what the [work] is about, and at times might consist only

      of its title; but there is a point in this series of abstractions where they

      are no longer protected, since otherwise the [author] could prevent the

      use of his "ideas," to which, apart from their expression, his property

      is never extended.(13) For example, a highly general statement of Romeo and Juliet is "a play about star-crossed lovers." More specific levels of abstraction include plot sequence or character development. The most specific level would be the words of a particular dialogue. After dissecting a work into general levels of abstraction, courts distinguish between unprotectable idea and protectable expression on a case-by-case basis.(14) Similarity of expression constitutes infringement; similarity of ideas does not.

      However, even case-by-case analysis cannot always cleanly dissociate protectable expression from unprotectable ideas. One copyright axiom is that a copyright holder should not obtain an effective monopoly over an idea by virtue of receiving protection for her particular expression. Courts accordingly deny protection if only a few ways to express an idea exist, under the theory that the idea and expression have "merged." For example, in Baker v. Selden, the Court denied copyright protection for the printed forms associated with a new kind of accounting system, reasoning that copyright protection for the forms would effectively give the author a patentlike monopoly over the system itself.(15)

      Similarly, courts deny copyright protection to expression portraying standard elements of a work when allowing protection would have the effect of precluding others from creating works of the same genre. For example, in Hoehling v. Universal City Studios, Inc.(16) the Second Circuit refused to extend protection to an author's descriptions of certain aspects of German life around the time of World War II, noting that "it is virtually impossible to write about a particular historical era or fictional theme without employing certain `stock' or standard literary devices."(17) This scenes a faire doctrine, like merger, supplements the basic idea/expression dichotomy by ensuring that protecting a work's expression does not have the unintended effect of also protecting ideas.

      Once a court has used these doctrines to identify a work's protectable elements, substantial similarity has a second, more intuitive meaning: how much protectable material must be copied to make the accused work substantially similar. The degree of similarity required to become "substantial" significantly varies by type of work.(18) For example, the content of a factual compilation such as a phone directory must be virtually identical to that of the copyrighted work in order to constitute infringement.(19) In contrast, one Second Circuit decision says that copying even a single line of a poem may constitute infringement.(20) Determining substantial similarity ultimately turns on a court's ad hoc value judgment of whether the copying is qualitatively significant to the copyrighted work,(21) a standard that obviously offers only general guidance to courts and litigants.(22)

      Although these copyright principles apply to computer programs just as to musical works or plays, most judges and juries have virtually no familiarity with software. Thus, while even a copyright neophyte can apply the idea/expression dichotomy to Romeo and Juliet, a court faced with the task of separating ideas from expression in a computer program requires a conceptual tool that relates traditional infringement analysis to software works. The Second Circuit's abstraction-filtration-comparison test provides such a framework and therefore represents a basically sound approach to deciding software infringement cases.

    2. Abstraction-Filtration-Comparison: Applying Traditional Principles to...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT