Why Copyright Software?|
Most people think of copyright law as applying to creative works like books, movies, paintings, photographs, sculptures and other works of art and/or literature; however, copyright law also protects technical works like software, technical specifications, and database schemas. As explained in our previous article here, a work must be registered with the U.S. Copyright Office before the copyright holder can sue for infringement. To register the copyright in a software application, the applicant must file the source code--that is, the human readable code that software programmers write and that is compiled into object or machine-readable code--with the U.S. Copyright Office. (For an article discussing the confidentiality of copyright deposits, please click here.) A copyright registration for software prevents someone else from copying your source code, but what else does it protect?
In 1986, the U.S. Court of Appeals for the Third Circuit held that a federal copyright registration for software protects the "concept and feel" (also referred to as "look and feel") of the program even where there is no copying of the source or object code. Whelan Assocs. Inc. v. Jaslow Dental Lab., Inc., 225 U.S.P.Q. 156 (E.D. Penn. 1985), aff'd, 230 U.S.P.Q. 481 (3d Cir. 1986). This holding does not extend as far as patent protection, which protects the functionality of the software (i.e., what it does), but it does go further than simply protecting the code that is written by programmers. Source code is analogous to the words written by a novelist--in the latter case, copyright law protects the way the writer expresses her story, but it does not protect the underlying story plot. In response to Whelan and similar cases, and out of concern that copyright law was being applied too broadly to software programs, courts began to analyze computer programs as if they were plays, the question being where to draw the line between an underlying idea and the expression of that idea.
In this vein, courts articulated and followed an abstraction-filtration-comparison test for determining to what extent software is protected by copyright. Under the abstraction step, the court must determine where to draw the line between source code (literal instructions) and functionality (the ultimate abstraction). To this author's knowledge, no court has articulated a bright-line test for settling this issue.
Under the filtration step, non-protectable elements of the software--such as those governed by the merger and scenes à faire doctrines--are filtered out. The merger doctrine applies when an idea cannot be separated from the expression of that idea. The scenes à faire doctrine provides that certain elements that are common to a particular genre of work are not copyrightable. Both the merger and scenes à faire doctrines have been relied upon by courts, often interchangeably, in excluding from copyrightability certain aspects of computer programs.
Finally, under the comparison step, the court compares what is left of the copyrighted software after the abstraction and filtration steps to the allegedly infringing work to determine whether the copyrightable aspects of the software have been copied.
If this all sounds overly complicated and less than certain, it is. The fact is that the scope of copyright protection for software is highly fact-specific, and the law in this area has fluctuated over time. To maximize the remedies available for infringement, however, the software program should be registered with the U.S. Copyright Office prior to first publication or, at the latest, within three months of the date of first publication or deployment. The bottom line is that a copyright registration for software protects more than just literal copying of the source code, but it does not protect the underlying functionality. Where along this spectrum a particular program falls will be dependent on the originality of the software, the court, and good advocacy.