IP monitor: Claim construction revisited: Positive changes for applicants seeking to protect computer-related inventions

A recent Federal Court of Canada decision found the current test  the Canadian Intellectual Property Office (CIPO) uses for assessing whether a computer-implemented invention (in this case, a method for selecting and weighing investment portfolio assets for an “anti-benchmark portfolio”) is patentable is not the proper test to apply.

This decision is a positive development for patent applicants as the Federal Court has indicated that CIPO’s restrictive problem-solution approach is akin to using the “substance of the invention” approach discredited by the Supreme Court. Software-related / computer-related inventions (or computer-implemented inventions) have encountered challenges during examination at the Canadian patent office (CIPO), particularly due to CIPO’s interpretation¹ of jurisprudence (Federal Court and the Federal Court of Appeal decisions in Amazon FC², Amazon FCA³, and Genencor⁴).  

The main issue is whether computer-implemented inventions as claimed are patentable under the definition of “invention,” under Section 2 of the Patent Act. The term invention is defined as “any new and useful art, process, machine, manufacture of composition of matter, or any new and useful improvement in any art, process, machine, manufacture of composition of matter.” Furthermore, Section 27(8) of the Patent Act prohibits granting a patent for a “mere scientific principle or abstract theorem.” However, while “a disembodied scientific idea is not per se patentable, it will be patentable if it has a method of practical application.”⁵

Two practice notices were issued by CIPO that adopted a stricter claim interpretation approach that followed a European-style problem-solution identification that did not consider an inventor’s intent as to which elements of the invention are essential. CIPO claimed the approach followed the purposive construction principles set forth by the Supreme Court, but in practice, the guidance was used by examiners to first “re-interpret” claims to establish the examiner’s purposive construction, removing claim terms relating to computer implementation and then issuing a finding of non-patentable subject matter. 

In practice, this new guidance has made it much more difficult to obtain allowances for computer-implemented inventions, as certain claim features were read out of the claims. Many applicants have noted CIPO’s approach significantly deviated from the principles of purposive construction but were unsuccessful before the Patent Appeal Board. The Patent Appeal Board had maintained that the problem-solution approach was in accordance with Canadian jurisprudence in several recent commissioner’s decisions, rebutting applicant arguments that the Practice Notices had no basis in law.

In Yves Choueifaty v Attorney General of Canada⁶, the commissioner’s reasons for refusal of a computer-implemented method for selecting and weighing investment portfolio assets were set aside in view of a fresh assessment in accordance with the principles of the purposive construction as set forth by the Supreme Court. It was expressly noted that the problem-solution approach taken by the commissioner in the initial review was the wrong test. 

This decision is a major victory for protecting computer-implemented inventions, as it reiterates the importance of the inventor’s intention and encourages a claim interpretation that is more sympathetic to the accomplishment of the inventor’s purpose expressed or implicit in the claims. The problematic problem-solution and associated claim re-interpretation approach taken by CIPO has been expressly disavowed by the courts, and applicants may have more latitude in advancing arguments for patentable subject matter.

In terms of practical drafting points, applicants should emphasize the technical nature and essentiality of computing components and computerized implementation in addressing practical, non-trivial challenges. The description should make it abundantly clear that the computer components giving physical effect to the claimed embodiments are intended to be essential. 

Background

In Amazon FC and Amazon FCA, the Federal Court and the Federal Court of Appeal reversed the commissioner’s decision on Patent Application No. 2,246,933, directed using cookies during internet shopping to store information used to streamline transactions in subsequent visits to a website (“one-click purchase”), ultimately finding that the claimed embodiments were indeed directed to patentable subject matter. 

Similar issues have arisen for computer-implemented inventions in other patent offices around the world, as applying this exclusionary approach is difficult in practice as there is no “bright-line test.” As noted by the US Supreme Court, one should “tread carefully in construing this exclusionary principle lest it swallow all of patent law. […] At some level, “all inventions … embody, use, reflect, rest upon, or apply laws of nature, natural phenomena, or abstract ideas.”⁷

In the two practice notices, Patent Notices 2013-02 and 2013-03, CIPO established a claim interpretation approach allegedly based on the “purposive constructive” approach set forth by the Supreme Court in Free World Trust⁸  and Whirlpool⁹.

The approach set out in the Patent Notices required a European-style problem-solution identification, and provided CIPO with a wide ambit to consider certain claim elements to be non-essential / superfluous. Furthermore, the Patent Notices noted that identifying the problem and the solution were an “integrated exercise,” whereby how the solution is described can help inform the problem, and vice versa.

It is important to note that Patent Notice 2013-02 further relied upon a statement in Genencor¹⁰ noting that the claims of a pending patent examination are different than those of granted patents, and justify a different approach to purposive construction for the Patent Office.

In practice, examiners utilized the Patent Notices to effectively re-write claims to remove the computer components from computer-implemented inventions, asserting they were non-essential. CIPO then argued the claims, as re-written by the examiner, were not patentable in respect of the definition of the term “invention.” 

This approach was frustrating to practitioners and applicants as the approach varied significantly between examiners in application, and adhered poorly to the principles of purposive construction set forth by the Supreme Court.

A number of applicants objected to CIPO’s approach, but were ultimately unsuccessful both in examination and on appeal to the Patent Appeal Board. The success rate on appeal to the Patent Appeal Board for patentable subject matter for computer-implemented inventions has been very low.¹¹ 

Yves Choueifaty v Attorney General of Canada

In Yves Choueifaty v Attorney General of Canada¹², the commissioner similarly refused Patent Application No. 2,635,393, directed to “Methods and Systems for Provision of an Anti-Benchmark Portfolio,” a computer-implemented method for selecting and weighing investment portfolio assets. 

The commissioner adopted the problem-solution approach as described above, and found that, for a first set of proposed claims, the essential elements of the claims were “directed to a scheme or rules involving mere calculations” for weighing securities, finding no discernible physical effect to satisfy the definition of ‟invention,” stating that the claims’ essential elements are only the rules and steps of an abstract algorithm. For a second set of proposed claims, the commissioner found that a computer was not an essential element and refused the claims for similar reasons as the first set of proposed claims.

The Federal Court found the commissioner applied the wrong test when interpreting the claims, and stated that using the problem-solution approach to claims construction is akin to using the “substance of the invention” approach discredited by the Supreme Court of Canada in Free World Trust. Furthermore, the Federal Court found the problem-solution approach fails to respond, as taught in Whirlpool, to the issue of the inventor’s intention.

The Federal Court further found that Genencor is no longer good law, and the commissioner’s position that the Whirlpool test is not applicable to examiners is not correct. Rather, the Federal Court noted that the commissioner is required to employ the purposive construction test set out in Whirlpool and Free World Trust, because the commissioner’s job, like a judge at trial, is to determine validity.

The decision set aside the commissioner’s decision.

The appellant submitted that the commissioner mischaracterized the purpose (or solution) of the claimed invention to be simply creating a new financial portfolio, and that another purpose of the invention was to improve computer processing.  The applicant asked the court to issue a declaration that the ‘393 Application discloses an invention.  The Federal Court noted it was not prepared to do so, noting that aspects of the invention require closer examination.

Accordingly, the commissioner has been tasked with conducting a fresh assessment.

Note that the review of the present decision invoked a recent change in the law governing standard of review. Whenever a patent office decision is appealed in court, the court has to determine what level of review of the decision is called for. For example, the courts could possibly review for “reasonableness” rather than “correctness.”¹³ The Supreme Court in twin decisions¹⁴ issued in 2019 laid out that any appeal that has basis in statute would be subject to the appellate standard of review where questions of law are reviewed for correctness. This is consistent with a view that CIPO guidelines touching on questions of law may be challenged and reviewed for correctness at the Federal Court.