Apple v. Microsoft. On August 7, 1992, Judge Vaughn Walker ordered partial summary judgment against Apple Computer, Inc. in the copyright case brought by Apple against Microsoft Corporation. Apple claimed that Microsoft’s Windows user interface infringed Apple’s copyright. With apologies to T. S. Eliot, the case, which has received intense scrutiny from the computer industry since it was filed in March 1988, ended “not with a bang but a whimper.”
For those who enjoy following the tactical sparring of industry giants quick review of the case’s background is in order. When Apple first filed suit Microsoft responded with a contractual (as opposed to a copyright ) defense. It argued that all of the elements in Windows had been licensed to Microsoft in a 1985 agreement.
The 1985 agreement proved a powerful shield for Microsoft. The judge ordered Apple to prepare a list of what Apple claimed were infringing features in Windows. Apple’s list had 189 elements. The judge ruled that 179 of these elements had been licensed to Microsoft under the 1985 agreement. Apple was left with only 10 features that it could complain were improperly copied.
After the contract issues were resolved Microsoft moved for summary judgment: it told the judge that it was apparent, as a matter of law, that the 10 remaining features were not protected by copyright law, and therefore there was no need for a trial.
In his August 7th opinion Judge Walker carefully parsed the remaining 10 features of the Macintosh interface and held almost none of them to be protectible by copyright law. Almost every element was found to be an idea, to flow naturally from an idea, to be unoriginal to Apple (copyright protection is limited to original expression), or to be the only, or one of very few, ways to express an idea.
The court’s insistence that the Apple interface be “analytically dissected” to determine which elements were protected by copyright law forced Apple to make arguments that were nothing short of embarrassing. For example, Apple claimed that it was entitled to copyright protection for rectangularly shaped windows, muted tones (such as black, white and grey) and overlapping (as opposed to tiled) windows. The court rightly concluded that, taken in isolation, these are ideas, not copyrightable expression.
Apple realized early on that analytic dissection of the Macintosh interface would not lead to a finding of copyright infringement against Microsoft. It did everything possible to persuade the court to view the programs in their entireties and compare their “look and feel,” rather than to look at each feature in isolation. However, the judge refused to take this approach, and once Apple’s software was subjected to analytic dissection, the outcome was a forgone conclusion.
Although this case involved two of the largest, most interesting players in the microcomputer/software industry, and the most creative, profitable software interfaces in the industry’s history, the applicability of the case in other contexts is limited by the 1985 license agreement between Microsoft and Apple. That agreement, which was unique to this case, determined much of the outcome. It was next to impossible for Apple to successfully argue an overall “look and feel” test when 95% of the interface had first been eviscerated by the license agreement.
Nevertheless, Judge Walker’s methodology is independent of the contractual issues. Defendants can rely on the decision to argue that before their programs and plaintiffs’ programs are compared for substantial similarity, the plaintiffs’ programs must be dissected, and each element excluded if it is not protected by copyright law. This approach is very similar, if not identical, to the approach taken by the Second Circuit inComputer Associates International, Inc. v. Altai, Inc. , which was discussed in detail in our most recent issue.
Lotus v. Borland. Depending on the observer, Lotus’ campaign to protect the command sequence and menu choices of the Lotus 1-2-3 user interface may be admired or condemned. However, one thing is beyond dispute: Lotus’ courtroom strategy over the last five years has been extraordinarily effective in achieving Lotus’ legal objectives.
Lotus began in early 1987 by filing suit in federal court in Massachusetts against Paperback Software and Mosaic Software, two so-called Lotus “clone” developers. These companies had published programs which copied the Lotus 1-2-3 command sequence virtually word for word. The cases were assigned to Judge Robert Keeton.
The issue in these cases was not whether the defendants had copied the Lotus command sequence, but whether what they had copied wasprotected by copyright law. Judge Keeton ruled against Paperback Software following a trial, holding that the Lotus 1-2-3 command sequence was protectible by copyright law and that Paperback Software had infringed Lotus’ copyright. Soon afterwards Paperback — the larger of the two defendants and the primary target of Lotus’ suit — agreed to settle the case (rather than appeal) by taking its product off the market and paying Lotus damages. Judge Keeton ruled against Mosaic following a separate trial, and shortly thereafter Mosaic filed for bankruptcy.
Defeating these two companies was only the first step in Lotus’ grand strategy. Within a few days after Judge Keeton’s decision against Paperback, Lotus filed suit in federal court in Massachusetts against a more important competitor, Borland International. Borland, the publisher of the popular Quattro spreadsheet program, had developed a user interface completely unlike Lotus 1-2-3. However, Borland gave Quattro users the option of emulating Lotus 1-2-3, and it was this aspect of the program which was the subject of Lotus’ suit.
Borland, expecting that this suit might come on the heels of a victory by Lotus in the Paperback litigation, had filed a preemptive declaratory judgment action in federal court in California immediately upon learning of the Paperback decision. That action, although first-filed, was dismissed in favor of the Massachusetts action. In a procedural coup de grace , Lotus was successful in assigning the Massachusetts action to the one judge in the United States whom it knew, without a doubt, viewed the Lotus 1-2-3 interface to be protected by copyright law: Judge Keeton.
After these strategic victories by Lotus, the outcome in the trial court was all but assured. The only surprise is that it took as long as it did. Borland tried, unsuccessfully, to persuade Judge Keeton that his decision in thePaperback action was incorrect. After initially denying a Lotus summary judgment motion in March 1992, Judge Keeton issued partial summary judgment against Borland on July 31, 1992, holding that Borland had illegally copied large parts of the Lotus 1-2-3 command structure.
The result of this decision is that Lotus has effectively prevailed on its claim of copyright infringement in the trial court. On the first business day following the decision, Borland announced that it would not include the Lotus-emulation portion of Quattro in future shipments of its program, in effect giving Lotus the injunctive relief it was seeking in its lawsuit.
It remains to be seen whether Borland will put itself through a damages trial and attempt to obtain victory on appeal, or try to settle the damages claims, thereby giving Lotus the benefit of a second favorable District Court decision in this circuit. We believe it is in Borland’s best interests to avoid the uncertainty of a damages trial, and in Lotus’ best interest to avoid the risk of a reversal in the Court of Appeals. Therefore the most likely outcome is a settlement in the District Court.
Because it involved literal copying of the Lotus 1-2-3 command sequence, as opposed to non-literal copying (that is, structural or organizational copying), the Lotus decision is less significant than either Apple orComputer Associates , which was discussed in detail in our last issue. Borland decided to copy the Lotus command structure long before the law was clear that user interfaces of software programs would be protected by copyright law. Only the most naive software developer would make the same mistake today. Therefore, in the future most cases involving infringement of computer software program will involve non-literal copying of either source code or a user interface, as was the case in Computer Associates and Apple .
Nevertheless, Lotus stands clearly for the proposition that — in the absence of some clear defense, such as non-originality or very limited forms of expression — wholesale copying of a complex, copyrighted command sequence creates a serious risk of liability.
Postscript: As our later articles discuss, the Lotus/Borland case did not settle, and was ultimately reversed by the U.S. Court of Appeals for the First Circuit. The case is now on appeal to the U.S. Supreme Court.
As to the Apple/Microsoft suit, Judge Walker’s decision was upheld on appeal.