Receive e-mail updates for new posts!

Responding to the New Analytic Framework of Ex parte Gutta during the Examination of Software Applications

By: David Moore of Rahman LLC

As noted in a previous article on FoundPersuasive, in Ex parte Gutta, the Board of Patent Appeals and Interferences (“BPAI”) delivered its precedential opinion affecting examination of software-related patent applications appearing before the United States Patent and Trademark Office (“USPTO”). Ex parte Gutta, No. 2008-4366 (B.P.A.I., Dec. 21, 2009). Ex parte Gutta is of a particular interest to those practicing before the USPTO, because the facts before the BPAI were distinguishable from the facts presented to the Supreme Court in the recently argued In re Bilski. See In re Bilski, 545 F.3d 943, (Fed. Cir. 2008). In particular, Ex parte Gutta signifies a new analytical framework for patent applications where an algorithm is recited in an apparatus and article of manufacture claim. Id. at 2 (BPAI, Dec. 21, 2009). This new analytical framework will have a direct bearing on patent professionals prosecuting patent applications before the USPTO, as Examiners rely on Ex parte Gutta to reject claims as allegedly reciting non-statutory subject matter. Applicants are not without recourse, however, during the prosecution of pending applications and preparation of new applications, as discussed in further detail below.

The BPAI introduced its new analytical framework as a two-part test to determine whether an apparatus and article of manufacture claim, which includes an algorithm, is patent-eligible. Ex parte Gutta, No. 2008-4366 slip op. at 15 (B.P.A.I., Dec. 21, 2009). In the first element of its test, the BPAI asked whether a claim is sufficiently limited to a tangible practical application, in which the mathematical algorithm is applied, that results in a real-world use (e.g., “not a mere field-of-use label having no significance”). Id. In the second element of its test, the BPAI asked whether a claim is sufficiently limited so as to not encompass substantially all practical applications of the mathematical algorithm either “in all fields” of use of the algorithm or even in “only one field.” Id. Each of these inquiries must be satisfied in order for an invention to be eligible for patent protection. Id. Through the creation of its two-part test, the BPAI further defined the bounds of patent-eligible subject matter for apparatus and article of manufacture claims that implement an algorithm.

The apparatus claim before the BPAI recited a “system for identifying one or more mean items for a plurality of items”, where the system included “a memory for storing computer readable code” and “a processor operatively coupled to the memory”. Id. at 5. The processor in this apparatus was configured to “compute a variance of the symbolic values of the plurality of items relative to each of the items” and “select the at least one mean item having a symbolic value that minimizes the variance”. Id. at 5-6. During its application of the two-part test, the BPAI indicated the claimed apparatus neglected to define a particular or physical item or article and did not define a tangible practical application that would use the algorithm. Id. at 20. The BPAI also indicated the apparatus claimed the algorithm itself, because the claim’s scope covered all practical applications of the algorithm. Throughout its opinion, the BPAI did not significantly distinguish between the apparatus claims and articles of manufacture claims when applying its new analytical framework. Consequently, persuasive responses to non-statutory subject matter rejections will likely follow a similar form for both statutory classes of claims.

Throughout its analysis, the BPAI appears to focus on whether the scope of what is being claimed by the applicant describes a real-world device (part one of its test) and whether there exists other possible devices that may implement the algorithmic portion of the claimed device without infringing the recited claim (part two of its test). In discussing the first part of its test, the BPAI found the recited “items” were “not limited to either a particular or physical item or article” and were merely abstract ideas. Id. at 20. Such abstractions, therefore, could not represent a real-world device. Claiming particular items (instead of mere abstractions) that the algorithm operates upon suggests a real-world device. For example, the BPAI suggested that if the claimed apparatus referred to “television programs” or “channels” or “titles” – instead of just “items” – it may have recited a real-world device and satisfied the first part of the test. Id.

Although the BPAI acknowledged the claimed invention included physical components of a real-world device – namely the memory and the processor – the BPAI found these elements of the apparatus to be inadequate structural recitations. The rationale in finding such structural recitation inadequate was anchored by the generic descriptions of those components provided by the specification. Thus, memory was described in the specification as merely RAM and/or ROM and the processor was merely described as a CPU – moreover, the claims were not limited in any way to these broad descriptions found in the specification. Thus, the BPAI found that there was no evidence another apparatus could have implemented the recited algorithm without infringing on the claimed apparatus. Id. at 21. In other words, the claimed apparatus pre-empted substantially all practical applications of the recited algorithm. Moreover, restricting the apparatus to a field-of-use (e.g., television programming) would still pre-empt substantially all practical applications of the recited algorithm because only the claimed “processor” and “memory” would enable a real-world device to perform the recited algorithm within a field-of-use. Id.

In its analysis of the recited structure, the BPAI failed to conceive or contemplate a possible apparatus, which was available at the time of the invention but was not claimed. The BPAI did suggest, however, that the failure of the claimed apparatus to tie the processor’s execution of the algorithm’s operations (e.g., compute and select) to the memory resulted in merely claiming a processor with the ability to perform the recited operations. Id. at 22. Conceivably, a claim that does tie the processor’s execution of an algorithm’s operation would distinguish over Ex parte Gutta. Furthermore, when a general-purpose processor relies on a memory to feed it instructions to execute an algorithm, such a processor would not pre-empt a purpose-built processor that executes the recited algorithm.

As noted above, several small measures may be taken while preparing new applications to avoid potential rejections in light of Ex parte Gutta. For example, avoid reciting abstractions in the claims and instead recite particular or physical entities. Unfortunately, the degree of particularity or the degree of physicality sufficient to overcome the first part of its test was not discussed by the BPAI. In dicta, however, the BPAI does suggest a fair amount of latitude. For example, “television programming” may have a sufficient degree of particularity. In addition, future prosecution of a new application may benefit from language in the specification that describes multiple device categories that may implement an algorithm, and each independent claim focuses on only one such device category.

Concerning pending applications, the same advice described above may also apply. One notable exception occurs, however, when an amendment to overcome a rejection is not explicitly described in the specification. In such circumstances, avoiding a subsequent new matter rejection may depend on arguments and/or evidence showing what was known to those skilled in the art at the time of invention. Consequently, in applications that do not describe particular items or device categories in their specification, any amendments used to persuade an Examiner that the claimed apparatus recites patent-eligible subject matter would likely fail to persuade an Examiner that a such added language provides a patentable distinction over prior art.

As demonstrated by the BPAI in Ex parte Gutta, securing patent rights for innovations that include an algorithmic component continues to be a hot-button issue in the patent community. Although the BPAI did not explicitly preclude all patent rights for algorithms in its opinion, the opinion does suggest more specificity and more practicality when claiming an apparatus or article of manufacture utilizing an algorithm. Consequently, innovations that incorporate an algorithm into an apparatus or article of manufacture continue to be viable, and valuable, investments in technology capable of patent protection.

David E. Moore is a registered patent attorney and practices intellectual property law in the Greater Washington, D.C. area. He may be reached at david.moore@me.com.