The Federal Circuit recently issued an en banc decision which will significantly impact the patentability of mathematical algorithms and computer software. In re Alappat, 31 USPQ2d 1545 (Fed. Cir. 1994).
The Alappat decision addresses two issues: 1) the propriety of a decision by the PTO Board of Patent Appeals and Interferences (Board) sustaining an Examiner's rejection of claims in Alappat's patent application as being unpatentable under 35 USC 101; and 2) how that Board decision was made.
Alappat's invention dealt with displaying graphic information on a video display. Originally, Alappat appealed the Examiner's § 101 mathematical algorithm rejection to the Board. The Board reversed the Examiner. The Examiner requested the Board to reconsider its decision. In a highly unusual move, a new Board panel was convened to rehear the case. The new panel included the PTO Commissioner and four other high level PTO officials as well as the original panel members. The expanded Board panel sustained the Examiner's rejection, thus reversing the original panel.
Many viewed the Commissioner as improperly overruling the original panel decision by "packing" the new panel with selected members. This created much discussion and concern among members of the Board and in the patent bar. In the first part of the Alappat decision, the Federal Circuit examined the Commissioner's legal basis for changing the composition of a Board of Appeals panel in this way and found such change to be within his authority.
The second part of the Alappat decision dealt with the substance of the mathematical algorithm rejection. Under U.S. law, mathematical algorithms per se are generally not patentable subject matter. For many years the PTO has routinely used the "Freeman-Walker-Abele" test for patentability of mathematical algorithms. Under the "Freeman-Walker-Abele" test, a claim that recites a mathematical algorithm is nevertheless patentable if the algorithm is applied to physical elements or process steps, as long as its application is circumscribed by more than a field of use limitation or non-essential post solution activity.
The expanded Board panel found: 1) the Alappat claims to recite a mathematical algorithm and 2) the means-plus-function elements in Alappat's claims to be construed as including all means that performed the recited function. Without reference to the specification to interpret means-plus-function elements, the expanded panel found that the claims did not recite any physical structure. Therefore, because the claimed subject matter lacked any physical structure, the expanded panel found the claims to cover all possible ways of implementing the algorithm, and to be unpatentable.
The Federal Circuit found that the expanded panel failed to read the means-plus-function elements to include the structures disclosed in the specification, citing In re Donaldson, 29 USPQ2d 1845 (see our June 1994 Special Report). A correct interpretation of the mean-plus-function elements in light of the specification showed the claims to recite a statutory apparatus under § 101.
The court's analysis did not stop there, however. The Court further stated that mathematical algorithms were not a type of subject matter expressly prohibited by § 101; rather, unpatentable types of mathematical algorithms represent laws of nature, natural phenomena and abstract ideas. The proper test for determining whether a claim is unpatentable, the court said, is "whether the claimed subject matter as a whole is a disembodied mathematical concept. . . which in essence represents nothing more than a 'law of nature,' 'natural phenomenon,' or 'abstract idea.'" The court said that a claim directed to a combination of interrelated elements' recites a specific machine, not a disembodied concept.
This theme has been continued in an even more recent case, In re Warmerdam, No. 93-1294 (Fed. Cir. August 11, 1994). The invention in Warmerdam was primarily a method of collision avoidance, employing data structures, for a robot. When the Federal Circuit analyzed the method claims within the "law of nature, natural phenomenon, abstract idea" framework, they were found to be invalid as non-statutory abstract ideas.
The Federal Circuit did not discuss the Freeman-Walter-Abele test for patentability of mathematical algorithms. This silence may signal a substantial change in the law. The "law of nature, natural phenomenon, abstract idea" test may be increasingly applied in future § 101 analysis, particularly of computer based inventions involving algorithms and/or data structures.