•  
  •  
 

Abstract

This article examines the issues surrounding software copyright protection and the problem of inconsistent court analysis across the country. After reviewing the problems particular to software copyright, the author recommends a new course of court analysis that would bring consistent analysis and decisions to this growing area of law. Courts have begun to examine the scope of protection the Copyright Act grants to computer software beyond the literal text of the software's programming language. Two classifications for computer programs have developed: literal, which focuses on the actual text of the program, and non-literal, which is concerned with organization, structure and presentation of the program. However, the courts have failed to develop a uniform copyright analysis to accompany the non-literal software classification. Instead, the decisions regarding potential non-literal copyright infringement are decided on an ad hoc basis. Absent a clearly defined uniform analysis, non-literal computer program components will not be fully protected by American copyright law. The answer to this problem lies in standardizing a combination of tests that some individual courts have applied to determine copyrightability and substantial similarity for computer programs. In these types of copyright cases, the idea/expression dichotomy must be settled in order to determine non-literal copyrightability. Therefore, the courts must use a two step process: they must extract and discern abstract components, and they must filter out unprotected components. The first phase of the copyrightability test, which was originally developed by Judge Learned Hand, determines copyrightability. The court uses expert testimony which explains the development of the allegedly infringed program, and then defines various levels of non-literal components within the program. Each non-literal component will contain some ideas, expressions and combinations of both. The second phase of the copyrightability test, the filtration phase, requires the court to sift through the components ascertained from the first phase to separate unprotectable elements from those deserving protection. This phase is accomplished using the numerous tests developed in the case history. The test to determine substantial similarity involves a three-pronged analysis. First, expert testimony is used to ascertain the level of similarity in the two programs in question. Second, the credibility of the expert testimony is weighed and qualified accordingly. Finally, an infringement will be found if, by a preponderance of the evidence, a computer-literate person would find that the program in question was copied from the original work. This bifurcated analysis is supported by the text and theory of copyright law and will produce better informed and more uniformly computer-literate court opinions. This uniformity in itself will provide more certain and comprehensive protection for non-literal computer program components.