John Kong | June 19, 2014
Ex Parte Cutlip
June 2, 2014
Panel: Lorin, Mohanty and Hoffman.
After the debacle of three March 2013 PTAB decisions by a five judge PTAB panel relying on a strange American Heritage dictionary definition of “processor” as being software, this PTAB decision sets the record straight about a “processor” as clearly being structure.
The “Processor” Mess at the USPTO:
Several Federal Circuit decisions have held that a “circuit” is a structural term. See, e.g., Linear Tech. Corp. v Impala Linear Corp. (Fed Cir 2004); and MIT v Abacus Software (Fed Cir 2006). One Federal Circuit decision even found that a “computing unit” sufficiently connotes structure because of claimed connections to other structures. Inventio AG v. ThyssenKrupp (Fed Cir 2011). However, no Federal Circuit decision has yet specifically addressed whether a “processor” is a structural term.
In re Alappat (Fed Cir 1994)(en banc) stated that “a computer operating pursuant to software may represent patentable subject matter…” and “a computer, like a rasterizer, is apparatus not mathematics.” The implication is that a computer includes a processor, and this computer/processor is considered to be an “apparatus.” The court in WMS Gaming v Int’l Game Tech. (Fed Cir 1999) identified the corresponding structure for a means-plus-function (MPF) claim element to be the microprocessor of a computer carrying out the disclosed algorithm for the claimed function. The implication is that a “processor,” which is sometimes used interchangeably with a “microprocessor,” is structure. In Microprocessor Enhancement Corp. v Texas Instr. (Fed Cir 2008), the court held that a claim directed entirely to a “pipelined processor” is not indefinite for using functional language. The court’s analysis of this “processor” claim as an apparatus claim implies that the court recognized a “processor” as a structural element. In Typhoon Touch v. Dell (Fed Cir 2011), the court held that “a processor…for executing said data collection application” requires that the recited function be performed. However, in all of these cases, the court did not specifically address the question as to whether the term “processor” itself sufficiently connotes structure to a POSA in the context of a §101 or §112, sixth paragraph issue. Over
Over the past year, practitioners may have experienced an increase in rejections under 35 USC §101 of claims reciting “a processor” as being directed to software, which is per se non-patentable subject matter. In the same vein, examiners may have treated “processor” as a nonce word, just like the word “means,” and interpreted the claimed element under 35 USC §112(f) or sixth paragraph (pre-AIA) as a means-plus-function (MPF) element.
These trends may have been foreshadowed by the questions presented to the public by the USPTO at the beginning of 2013. In January, 2013, the USPTO asked, “should claims that recite a computer for performing certain functions or configured to perform certain functions be treated as invoking 35 U.S.C. 112(f) although the elements are not set forth in conventional means-plus- function format?” The general response was that there’s a strong presumption that MPF is not being triggered if “means for” is not used.
Nevertheless, in March 2013, a five Judge PTAB panel issued three decisions which held that a “processor” is just software and interpreted the processor plus function limitation as a MPF element: Ex parte Erol (Appeal 2011-001143, US app. s/n 11/461,109); Ex parte Lakkala (Appeal 2011-001526, US app. s/n 10/949,568); and Ex parte Smith (Appeal 2012-007631, US app. s/n 12/579,383).
Ex parte Erol involved a system claim comprising “a processor adapted to:
determine…a first object descriptor…
determine a set of … objects from digital media content;
generate an object descriptor…;
identify at least one object descriptor…; and
perform an action in response to identifying the at least one object descriptor that matches the first object descriptor.”
Ex parte Lakkala involved a device claim comprising a memory storing a program and “a processor … configured with the program to:
control creation of metadata…;
control collection of content data…; and
control adding of the created metadata to the content data set.”
Ex parte Smith involved a system claim comprising a memory and “a processor in communication with the memory, … programmed to:
receive…a first review of an asset;
store the first review…;
receive…a second review of the assert;
store the second review…; and
generate an opinion timeline for the asset for the user associated with the user identifier.”
In all three decisions, this PTAB panel asserted that the term “processor” was not defined in the specification. In the Smith case, the specification stated that the “computer system 600 includes a processor 602 (e.g., a central processing unit (CPU), a graphics processing unit (GPU) or both…” In the Lakkala case, the specification stated that “[u]ser device 100a may also include a CPU 200 and associated memory 205 containing programming for controlling, in accordance with the present invention, data processing and transfer operations among the various elements of device 100a via a data transfer bus 250.” Apparently, the specification reference to a CPU is not a definition of a “processor.”
Accordingly, the PTAB panel proceeded to rely on a 2006 American Heritage dictionary definition of a “processor” as: a) a computer, b) a central processing unit, OR c) a program that translates another program into a form acceptable by the computer being used. Because of this strange definition of a “processor” as being a “program,” the panel held that a person of ordinary skill in the art (POSA) would not recognize “processor” as the name of a sufficiently definite structure for the claimed function(s).
This seems to be an unreasonable application of “broadest reasonable interpretation.” To date, I have not yet met anyone of “ordinary skill in the art” who has given me a definition of a “processor” as being “a program that translates another program into a form acceptable by the computer being used.” Actually, that appears to be a definition for a compiler – not a processor.
Unfortunately, the damage has been done. Because a “processor” is supposedly not recognized by a POSA as connoting sufficiently definite structure, and is supposedly a program (software), §101 rejections are being made for claims reciting a “processor” as being directed to non-patentable software per se.
In addition, the term “processor” was treated as a nonce word in these three 2013 PTAB decisions, just like the word “means” in a MPF claim element. Apparently, this PTAB panel answered the USPTO’s own question in the affirmative – treating a computer/processor claim element as merely triggering MPF interpretation, even though the processor claim element is not drafted in conventional MPF format. And then, the PTAB panel easily disposed of these cases under the Aristocrat rule, finding the claims indefinite under 35 USC §112, second paragraph (pre-AIA) for failing to have a specification disclosure for the required algorithm for at least one of the claimed functions of the computer-implemented MPF claim element.
Hence the tri-partite dilemma of a “processor” claim: (1) “processor” is merely software and a claim simply reciting a processor is non-patentable software per se; (2) alternatively, it is interpreted under 35 USC §112(f) or sixth paragraph (pre-AIA); and (3) it is indefinite under 35 USC §112(b) or second paragraph (pre-AIA) if there is no disclosed algorithm for performing the claimed function of the MPF “processor” claim element.
Ex Parte Cutlip to the Rescue:
A PTAB panel issued a decision on June 2, 2014 in Ex Parte Cutlip (Appeal 2011-0011658, US app. s/n 11/345,010) that is directly contrary to the previous three March 2013 decisions.
Ex Parte Cutlip involved a system claim comprising:
a processor and a memory configured to provide computer program instructions to the processor;
a module to read ontological data from a data source…;
a module to generate a first ontological model…;
a module to read performance characteristics…;
a module to read real time or near real time knowledge information…; and
a module to generate a modified ontological model…
The invention is directed to functional features recited in the claimed “modules” (software units executing on a processor). The Examiner rejected the claim under §101 as being directed to software. However, this PTAB panel reversed the Examiner’s rejection saying “the claimed processor and memory very clearly refer to structural elements which Appellants’ Specification discloses are used to implement the functionality recited by the claimed modules.” The referenced specification disclosure stated:
These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. (Paragraph ).
In comparison to the Smith and Lakkala cases, Cutlip’s specification does not define a “processor” and doesn’t even mention any CPU. But, this PTAB panel did not assert that the claimed “processor” is undefined and did not rely on the American Heritage dictionary definition of a “processor” as being software. Quite the contrary, this PTAB panel found that “the claimed processor and memory very clearly refer to structural elements.” This is a welcome relief from the Erol, Lakkala, and Smith decisions. Ex Parte Cutlip is sure to be cited to overcome §101 rejections as support for the argument that a processor is very clearly a structural element and is not just software per se. Likewise, MPF interpretation does not apply because a “processor” is very clearly a structural element, as indicated in Ex Parte Cutlip.
Although the Erol, Lakkala, and Smith PTAB decisions still exist, applicants now have at least a fighting chance by citing this Cutlip PTAB decision. Nevertheless, if an appeal is anticipated, it wouldn’t hurt to submit evidence (e.g., a declaration, dictionary definitions, and/or other testimony) indicating that a POSA would understand a “processor” as being a hardware device – and not just software. New applications may also benefit from specification descriptions of a processor being a hardware device programmed with computer program instructions to produce a machine, special purpose computer, etc.
Although not raised in this appeal, another question is whether the “module” elements should be interpreted as MPF elements. The Cutlip specification states that “an embodiment combining software and hardware aspects that may all generally be referred to herein as a ‘circuit,’ ‘module’ or ‘system.’” (Paragraph ). This disclosure ties the claimed term “module” to a combination of software and hardware. It could be argued that the inventor is acting as his own lexicographer in defining a “module” as connoting structure by the combination with hardware. If so, this is not a MPF claim element. However, what specific hardware is being identified? A POSA might interpret the referenced “hardware” as a computer, processor, circuit, ASIC, or some other logic device that can run the programming provided by software. As described in the specification, the modules are essentially the functions/acts/computer program instructions executing on the processor or some other programmable data processing apparatus. Without a specific identification of a type of hardware, “module” could be argued to be serving the same purpose as a nonce word, like “means.” If so, this is a MPF element.
Another question not raised in this appeal is whether the inclusion of structure, i.e., the processor and memory, is enough to save the claims from a rejection under §101 as being directed to the non-patentable subject matter of an abstract idea. This is an extension of the issue in the Alice Corp v CLS Bank case just decided by the Supreme Court. Does the claim recite mere generic computer implementation of generating an ontological model? If so, it is no more than the abstract idea of doing so. There must be some additional feature, such as improving the functionality of the computer in doing the generation of the ontological model, or effecting an improvement in some other technology or technical field, that would transform an otherwise abstract idea into a patent-eligible invention. The mere inclusion of a computer or a processor is insufficient to convert an abstract idea into patentable subject matter. Does the claim recite some specific way the processor must be programmed to read and/or generate ontological models? Although this question is debatable, treating the “module” as a MPF element may be one way to overcome the abstract idea problem. If the “module” elements are interpreted as MPF elements, meaningful limitations may exist in the algorithms disclosed to perform the various functions.
Of course, the applicant’s concern would then focus on the scope of protection for a claim that is limited to the disclosed algorithms. But that would be a different issue from the abstract idea issue. And, the scope of protection issue depends on whether the specification includes alternative/different levels of detail describing the algorithm.