Patentable Subject Matter After Bilski

For a claim to be clearly abstract then, it must in Judge Rader's view “manifestly override” the statutory categories of ' 101, or it must be entirely and completely capable ' as claimed ' of operation by the human mind. In short, it appears that under Judge Rader's approach, any recitation of physical technology lets the claim escape this narrow category.

While Judge Rader's framework is favorable for finding patent-eligible subject matter, it needs further development. For example, Judge Rader states that inventions requiring “complex” programming are likely to be patent eligible, but does not define either quantitatively or qualitatively the level of complexity required to meet this test. If complex programming is intended to indicate eligible subject matter instead of a pure mental method or an abstract idea, the presence of any programming alone regardless of its “complexity” should suffice, since even the simplest program must be executed by a computer. Requiring “complexity” may eventually lead to an undesirable blurring of the novelty and non-obviousness requirements with the patent-eligible subject matter requirement.

Nonetheless, Judge Rader's expansive reading of ' 101 produces an end result that is in line with the social contract between a patentee and the public, i.e., a patentee only gets a patent for eligible inventions that have been adequately disclosed to the public. A claimed abstract idea that may slip through Judge Rader's coarse eligibility filter would not end up being patented because it would likely fail the requirements of enablement, written description, definiteness, novelty, and non-obviousness.

Judge Dyk's Approach

Judge Dyk, in Cybersource Corp. v. Retail Decisions, Inc., 654 F.3d 1366 (Fed. Cir. 2011) and Dealertrack Inc. v. Huber, __ F.3d __ (Fed. Cir. 2012), has ruled that a method that may be performed purely mentally is ineligible subject matter. In coming to that determination, the court may ignore the express language of the claim, and attempt to look at some “underlying” invention. Dealertrack, __ F.3d at __; Cybersource, 654 F.3d at 1374-1375. Under this approach, even if claims for such methods recite a particular technology to perform certain method steps, the presence of such may not render the method patent eligible unless the technology adds “meaningful limitations” to the claim ' a phrase that itself has not yet been defined or elaborated by the courts for 40 years since it was first introduced in Gottschalk v. Benson, 409 U.S. 63 (1972). Additionally, in Cybersource, 654 F.3d at 1374-1375, Judge Dyk held that a claim for a computer readable medium including instructions for performing such methods is also ineligible subject matter,
essentially invalidating more than 15 years of standard claim drafting practice. The court did carve out an exception for computer readable media storing instructions for performing a method that cannot be mentally performed. Id.

Judge Dyk's justification for creating this new exception stands on shaky grounds, and the court's application of its rules to identify abstract ideas does not provide adequate guidance to inventors and patent counsel. Judge Dyk acknowledges the Federal Circuit's ruling in In re Alappat, 33 F.3d 1526 (Fed. Cir. 1994) (en banc), that a general purpose computer becomes a particular machine when it is programmed through software to perform a particular function. Id. However, Judge Dyk ignores the rest of the Alappat court's analysis. The Alappat court held that a claimed apparatus that performs a mathematical algorithm is patent eligible because the claim as a whole recites an apparatus and is therefore not a disembodied algorithm. In re Alappat, 33 F.3d at 1543-1544. Similarly, a claim for an apparatus that implements a method ' even one that can be performed mentally ' must be taken as an eligible apparatus and not a disembodied mental method.

In addition to ignoring the analysis in Alappat, the creation of this new exception leads to unintended consequences of potentially excluding broad fields of technologies from eligible subject matter. An inventor of a new method of analyzing any information ' whether image data, network signals, or medical test results ' must necessarily first think through the process, step by step, before reducing it to a machine implementation. Judge Dyk's approach can be used then to attack the method claims for these processes for being purely mental, even though the claims clearly recite that the processes are implemented by machines and are applied to real-world data with real-world effects.

Beyond the Four Corners

The analytical framework in Cybersource and Dealertrack makes matters worse as it goes beyond the four corners of the claims at issue to identify the inventions and determine their eligibility. Although these rulings acknowledge that a claimed invention is described by the claim, these rulings ignore the claim language and deduce the claims' “simplest forms” or the “underlying inventions” and then analyze these forms for patent eligibility. This analytical framework revives the previously overruled jurisprudence that deduced the “gist of the invention” or the “spirit of the invention” to analyze a claim. See, e.g., Bausch & Lomb v. Barnes-Hind/Hydrocurve, Inc., 796 F.2d 443, 447-49 (Fed. Cir. 1986), cert. denied, 484 U.S. 823 (1987) (“The district court [improperly] distill[ed] the invention down to a 'gist' or 'core,' a superficial mode of analysis that disregards elements of the whole ' . [and] express limitations [in the claims]“). Indeed, ignoring the actual words of the claims violates the “bedrock principle” of patent law that “we look to the words of the claims themselves ' to define the scope of the patented invention.” Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996). Revival of this previously rejected framework is fraught with dangers as it: 1) does not properly analyze the scope of the claimed invention, 2) glosses over various limitations in the claims, and 3) makes it easier to avoid analyzing the nuances of a claimed invention.

For example, the claim in Dealertrack recites a method of managing a credit application that includes the steps of receiving credit application data from a credit application terminal, selectively forwarding the credit application data to remote funding terminals, and forwarding the funding decision regarding the credit application to the credit application terminal. The claim also recites a list of four criteria from which a criterion may be selected to selectively forward the credit application data. The Dealertrack court recites this claim in its “simplest form” as receiving data from one source, selectively forwarding the data based on a criterion, and forwarding reply data to the first source. According to the Dealertrack court, these steps recite an abstract idea that pre-empts the clearinghouse concept.

There are three problems here. First, the actually claimed method does not pre-empt the clearinghouse concept. For example, the claim does not cover methods for matching loan applicants with a particular lender where the lenders bid for qualifying loans without ever receiving any application data or before an applicant has even applied for the loan.

Second, there is no clear rule as to how far the court should go in identifying the simplest form: Under the Dealertrack approach, any claim can be simplified to such an extent as to be deemed “abstract.” Excising limitations from the claim is no less dangerous than incorporating them from the specification. As the Supreme Court warned in McCarty v. Lehigh Valley R.R. Co., 160 U.S. 110, 116 (1895), “if we once begin to include elements not mentioned in the claim, in order to limit such claim ' , we should never know where to stop.”

Third, Judge Dyk's analytical framework takes a restrictive reading of ' 101 and creates the risk of excluding deserving inventions from patent protection. Unlike Judge Rader's approach, set forth in Research Corp. and Ultramercial, that can rely on other requirements of the Patent Act to minimize the effect of the approach's deficiencies, Judge Dyk's framework does not have similar safeguards. Judge Dyk's approach puts a chilling effect on inventors and patent counsel to file applications on inventions that may ultimately be rejected or invalidated as being “abstract.” As such, these inventions ' regardless of their merits ' may never get pursued, protected, and commercialized. By contrast, Judge Rader's approach provides a comparative bright-line test and thus does not discourage such patent applications.

Practices for Consideration By Patentees

As discussed above, both the Rader and Dyk schools of thought need further development and refinement for producing a consistent framework for eligible subject matter jurisprudence. In particular, Judge Dyk's expansion of the pure mental method exception contradicts previous jurisprudence and leads to unintended consequences. Companies and inventors cannot adopt a wait and see approach to learn which school of thought will prevail, as they must now take steps to protect their intellectual property. To minimize the risk of claims being found abstract, patentees should consider the following practices in pursuing patents for their inventions.

Most importantly, a patent specification should expressly describe as much technical and physical structure as possible for implementing the invention, especially for software and business method inventions. Beyond the typical “white box” system diagram and several flowcharts, further levels of detail should be included, such as database schemas, data access methods, procedural interfaces, and the like. In some cases, portions of computer code (real or pseudo) should be included to explain complex algorithms or operations. The specification should indicate how individual modules receive signals or data from other modules, what particular information received signals represent, the processing performed by the modules on the received information, the output signals representing output information produced by these modules and where the output information is sent (e.g., to a display, a database, or a particular module). Such description and claims are likely to make it harder to find claims as disembodied abstract ideas instead of a particular machine or a machine-implemented method.

It may also be helpful to identify specific fields of technology and describe how the invention has application in such fields. While it is generally desirable to claim a particular method broadly, a different approach would be to have multiple independent claims targeted to specific applications.

Additionally, if the claimed invention solves a particular problem or provides additional improvements over the prior art, like providing a faster, more reliable, more accurate, or lower-cost solution, then these or similar tangible benefits should be explicitly described, and where possible, quantitative data given of the benefits. Such description is likely to indicate the invention's contribution to improving existing technology and thus provide courts with one more reason to find the claimed inventions patent eligible.

Patentees with high-tech inventions (i.e., inventions in hardware, software or business method fields) should consider explicitly stating that the claimed method and apparatus do not cover a pure mental method performed solely by the human mind. Disclaiming pure mental methods should preclude Judge Dyk's analysis, and does not practically curtail the claims' scope because mental implementations of high-tech methods do not have commercial value.

When appealing the examiner's subject matter rejections or defending claims during litigation, patentees should consider providing separate arguments for the method and apparatus claims. Otherwise, the courts may conveniently analyze only the method claims and invalidate both the method claims and apparatus claims based on such analysis. Separately arguing for apparatus claims forces a court to focus on the actual claim language, and thus makes it more difficult to distill the “gist of the invention” as a prelude to patent eligibility.

As the Federal Circuit continues the debate regarding eligible subject matter, further guidance is likely to be provided. Moreover, the Supreme Court may also provide some guidance on this issue in the near future when the Court rules on related but different subject matter issues in Prometheus Labs. v. Mayo Collaborative Services, which was argued on Dec. 7, 2011. Given the complexities of this debate, the issue is unlikely to be resolved in the near future. To weather the continuing storm of patent eligibility, patentees should shore up their patent applications just as they would their homes: with a strong technical framework and solid factual foundation.

Robert R. Sachs is an Intellectual Property Group partner in the San Francisco office of Fenwick & West LLP. His practice emphasizes strategic counseling for patent portfolio development. Puneet Sarna is an associate in the firm's Intellectual Property Group. His practice focuses on patent portfolio development.

The Supreme Court in Bilski v. Kappos , 130 S. Ct. 3218 (2010), affirmed the long-standing maxim that laws of nature, natural phenomena, and abstract ideas are the only three exceptions to broad categories of patent-eligible subject matter. The Supreme Court, however, did not define what constitutes an abstract idea and left this task to the Federal Circuit. Two schools of thought in the Federal Circuit, one led by Judge Randall R. Rader and the other by Judge Timothy B. Dyk, have been competing for development and application of a framework to identify an abstract idea. Both of these camps, however, have avoided tackling this task head on. Instead, each camp resorts to a variant of Justice Potter Stewart's famous “I know it when I see it” test. Judge Rader knows a claim is not abstract when he sees “hard” technology recited in the claim. Conversely, Judge Dyk knows a claim is abstract when he can imagine the claim being performed mentally. As a result of these two contrary approaches, a patent applicant is left with inadequate guidance for drafting its claims and can only take steps to minimize the likelihood of a court finding a claim abstract.

Judge Rader's Approach

Judge Rader, in Research Corp. v. Microsoft , 627 F.3d 859 (Fed. Cir. 2010), and in Ultramercial v. Hulu , 657 F.3d 1323 (Fed. Cir. 2011), approaches the problem with a definite bent toward an expansive reading of 35 U.S.C. ' 101, using a framework that gives due recognition to both modern technology and to the core assumption that inventors create ' and patent attorneys attempt to claim ' commercially valuable and useful inventions. Like the Supreme Court though, Judge Rader does not offer a definition of what is “abstract,” but instead employs a functional test: If the claim recites any recognizable technology, it is not abstract. This approach follows Judge Rader's view that ' 101 is at best a “coarse eligibility filter,” and thus a subject matter eligibility test should be as inclusive as possible. Thus, in both Research Corp. and Ultramercial, Judge Rader found patentable subject matter because the claims clearly recited specific technologies, such as computers, printers, display devices, and implicated “intricate and complex” computer programming. Further, Judge Rader pays respect to actual words of the claim. If the claim recites a computer system operating in a particular way, Judge Rader accepts this as a real and particular computer, and thus real technology.

For a claim to be clearly abstract then, it must in Judge Rader's view “manifestly override” the statutory categories of ' 101, or it must be entirely and completely capable ' as claimed ' of operation by the human mind. In short, it appears that under Judge Rader's approach, any recitation of physical technology lets the claim escape this narrow category.

While Judge Rader's framework is favorable for finding patent-eligible subject matter, it needs further development. For example, Judge Rader states that inventions requiring “complex” programming are likely to be patent eligible, but does not define either quantitatively or qualitatively the level of complexity required to meet this test. If complex programming is intended to indicate eligible subject matter instead of a pure mental method or an abstract idea, the presence of any programming alone regardless of its “complexity” should suffice, since even the simplest program must be executed by a computer. Requiring “complexity” may eventually lead to an undesirable blurring of the novelty and non-obviousness requirements with the patent-eligible subject matter requirement.

Nonetheless, Judge Rader's expansive reading of ' 101 produces an end result that is in line with the social contract between a patentee and the public, i.e., a patentee only gets a patent for eligible inventions that have been adequately disclosed to the public. A claimed abstract idea that may slip through Judge Rader's coarse eligibility filter would not end up being patented because it would likely fail the requirements of enablement, written description, definiteness, novelty, and non-obviousness.

Judge Dyk's Approach

Judge Dyk, in Cybersource Corp. v. Retail Decisions, Inc. , 654 F.3d 1366 (Fed. Cir. 2011) and Dealertrack Inc. v. Huber , __ F.3d __ (Fed. Cir. 2012), has ruled that a method that may be performed purely mentally is ineligible subject matter. In coming to that determination, the court may ignore the express language of the claim, and attempt to look at some “underlying” invention. Dealertrack, __ F.3d at __; Cybersource, 654 F.3d at 1374-1375. Under this approach, even if claims for such methods recite a particular technology to perform certain method steps, the presence of such may not render the method patent eligible unless the technology adds “meaningful limitations” to the claim ' a phrase that itself has not yet been defined or elaborated by the courts for 40 years since it was first introduced in Gottschalk v. Benson , 409 U.S. 63 (1972). Additionally, in Cybersource, 654 F.3d at 1374-1375, Judge Dyk held that a claim for a computer readable medium including instructions for performing such methods is also ineligible subject matter,
essentially invalidating more than 15 years of standard claim drafting practice. The court did carve out an exception for computer readable media storing instructions for performing a method that cannot be mentally performed. Id.

Judge Dyk's justification for creating this new exception stands on shaky grounds, and the court's application of its rules to identify abstract ideas does not provide adequate guidance to inventors and patent counsel. Judge Dyk acknowledges the Federal Circuit's ruling in In re Alappat, 33 F.3d 1526 (Fed. Cir. 1994) (en banc), that a general purpose computer becomes a particular machine when it is programmed through software to perform a particular function. Id. However, Judge Dyk ignores the rest of the Alappat court's analysis. The Alappat court held that a claimed apparatus that performs a mathematical algorithm is patent eligible because the claim as a whole recites an apparatus and is therefore not a disembodied algorithm. In re Alappat, 33 F.3d at 1543-1544. Similarly, a claim for an apparatus that implements a method ' even one that can be performed mentally ' must be taken as an eligible apparatus and not a disembodied mental method.

In addition to ignoring the analysis in Alappat, the creation of this new exception leads to unintended consequences of potentially excluding broad fields of technologies from eligible subject matter. An inventor of a new method of analyzing any information ' whether image data, network signals, or medical test results ' must necessarily first think through the process, step by step, before reducing it to a machine implementation. Judge Dyk's approach can be used then to attack the method claims for these processes for being purely mental, even though the claims clearly recite that the processes are implemented by machines and are applied to real-world data with real-world effects.

Beyond the Four Corners

The analytical framework in Cybersource and Dealertrack makes matters worse as it goes beyond the four corners of the claims at issue to identify the inventions and determine their eligibility. Although these rulings acknowledge that a claimed invention is described by the claim, these rulings ignore the claim language and deduce the claims' “simplest forms” or the “underlying inventions” and then analyze these forms for patent eligibility. This analytical framework revives the previously overruled jurisprudence that deduced the “gist of the invention” or the “spirit of the invention” to analyze a claim. See, e.g., Bausch & Lomb v. Barnes-Hind/Hydrocurve, Inc. , 796 F.2d 443, 447-49 (Fed. Cir. 1986), cert. denied , 484 U.S. 823 (1987) (“The district court [improperly] distill[ed] the invention down to a 'gist' or 'core,' a superficial mode of analysis that disregards elements of the whole ' . [and] express limitations [in the claims]“). Indeed, ignoring the actual words of the claims violates the “bedrock principle” of patent law that “we look to the words of the claims themselves ' to define the scope of the patented invention.” Vitronics Corp. v. Conceptronic, Inc. , 90 F.3d 1576, 1582 (Fed. Cir. 1996). Revival of this previously rejected framework is fraught with dangers as it: 1) does not properly analyze the scope of the claimed invention, 2) glosses over various limitations in the claims, and 3) makes it easier to avoid analyzing the nuances of a claimed invention.

For example, the claim in Dealertrack recites a method of managing a credit application that includes the steps of receiving credit application data from a credit application terminal, selectively forwarding the credit application data to remote funding terminals, and forwarding the funding decision regarding the credit application to the credit application terminal. The claim also recites a list of four criteria from which a criterion may be selected to selectively forward the credit application data. The Dealertrack court recites this claim in its “simplest form” as receiving data from one source, selectively forwarding the data based on a criterion, and forwarding reply data to the first source. According to the Dealertrack court, these steps recite an abstract idea that pre-empts the clearinghouse concept.

There are three problems here. First, the actually claimed method does not pre-empt the clearinghouse concept. For example, the claim does not cover methods for matching loan applicants with a particular lender where the lenders bid for qualifying loans without ever receiving any application data or before an applicant has even applied for the loan.

Second, there is no clear rule as to how far the court should go in identifying the simplest form: Under the Dealertrack approach, any claim can be simplified to such an extent as to be deemed “abstract.” Excising limitations from the claim is no less dangerous than incorporating them from the specification. As the Supreme Court warned in McCarty v. Lehigh Valley R.R. Co. , 160 U.S. 110, 116 (1895), “if we once begin to include elements not mentioned in the claim, in order to limit such claim ' , we should never know where to stop.”

Third, Judge Dyk's analytical framework takes a restrictive reading of ' 101 and creates the risk of excluding deserving inventions from patent protection. Unlike Judge Rader's approach, set forth in Research Corp. and Ultramercial, that can rely on other requirements of the Patent Act to minimize the effect of the approach's deficiencies, Judge Dyk's framework does not have similar safeguards. Judge Dyk's approach puts a chilling effect on inventors and patent counsel to file applications on inventions that may ultimately be rejected or invalidated as being “abstract.” As such, these inventions ' regardless of their merits ' may never get pursued, protected, and commercialized. By contrast, Judge Rader's approach provides a comparative bright-line test and thus does not discourage such patent applications.

Practices for Consideration By Patentees

As discussed above, both the Rader and Dyk schools of thought need further development and refinement for producing a consistent framework for eligible subject matter jurisprudence. In particular, Judge Dyk's expansion of the pure mental method exception contradicts previous jurisprudence and leads to unintended consequences. Companies and inventors cannot adopt a wait and see approach to learn which school of thought will prevail, as they must now take steps to protect their intellectual property. To minimize the risk of claims being found abstract, patentees should consider the following practices in pursuing patents for their inventions.

Most importantly, a patent specification should expressly describe as much technical and physical structure as possible for implementing the invention, especially for software and business method inventions. Beyond the typical “white box” system diagram and several flowcharts, further levels of detail should be included, such as database schemas, data access methods, procedural interfaces, and the like. In some cases, portions of computer code (real or pseudo) should be included to explain complex algorithms or operations. The specification should indicate how individual modules receive signals or data from other modules, what particular information received signals represent, the processing performed by the modules on the received information, the output signals representing output information produced by these modules and where the output information is sent (e.g., to a display, a database, or a particular module). Such description and claims are likely to make it harder to find claims as disembodied abstract ideas instead of a particular machine or a machine-implemented method.

It may also be helpful to identify specific fields of technology and describe how the invention has application in such fields. While it is generally desirable to claim a particular method broadly, a different approach would be to have multiple independent claims targeted to specific applications.

Additionally, if the claimed invention solves a particular problem or provides additional improvements over the prior art, like providing a faster, more reliable, more accurate, or lower-cost solution, then these or similar tangible benefits should be explicitly described, and where possible, quantitative data given of the benefits. Such description is likely to indicate the invention's contribution to improving existing technology and thus provide courts with one more reason to find the claimed inventions patent eligible.

Patentees with high-tech inventions (i.e., inventions in hardware, software or business method fields) should consider explicitly stating that the claimed method and apparatus do not cover a pure mental method performed solely by the human mind. Disclaiming pure mental methods should preclude Judge Dyk's analysis, and does not practically curtail the claims' scope because mental implementations of high-tech methods do not have commercial value.

When appealing the examiner's subject matter rejections or defending claims during litigation, patentees should consider providing separate arguments for the method and apparatus claims. Otherwise, the courts may conveniently analyze only the method claims and invalidate both the method claims and apparatus claims based on such analysis. Separately arguing for apparatus claims forces a court to focus on the actual claim language, and thus makes it more difficult to distill the “gist of the invention” as a prelude to patent eligibility.

As the Federal Circuit continues the debate regarding eligible subject matter, further guidance is likely to be provided. Moreover, the Supreme Court may also provide some guidance on this issue in the near future when the Court rules on related but different subject matter issues in Prometheus Labs. v. Mayo Collaborative Services, which was argued on Dec. 7, 2011. Given the complexities of this debate, the issue is unlikely to be resolved in the near future. To weather the continuing storm of patent eligibility, patentees should shore up their patent applications just as they would their homes: with a strong technical framework and solid factual foundation.

Robert R. Sachs is an Intellectual Property Group partner in the San Francisco office of Fenwick & West LLP. His practice emphasizes strategic counseling for patent portfolio development. Puneet Sarna is an associate in the firm's Intellectual Property Group. His practice focuses on patent portfolio development.