Software patentability: restoring the transatlantic balance
At the heart of the debate over whether computer programs should be patentable is a philosophical question: to what extent do particular patents for computer programs seek to monopolise abstract ideas? Despite these philosophical origins, the patentability of computer programs has very significant practical consequences in the technological era.
The UK’s approach, as set out in section 1(2)(c) of the Patents Act 1977, expressly excludes computer programs from patentability to the extent that a patent application relates to a computer program “as such”. By contrast, US patent law contains no specific exclusion for computer programs and, focusing simply on whether an application involves patent-eligible subject matter, allows at least in theory for computer program patents.
Hence there is seemingly a fundamental difference in the two approaches and, for this reason, the US has traditionally been regarded as more amenable to patenting inventions covering computer programs and business methods.
UK approach
The starting point for determining whether a computer program falls within the exclusion is the English Court of Appeal’s four-stage test in Aerotel v Telco Holdings (2006).
Aerotel: the four-stage test
- Properly construe the claim;
- Identify the actual contribution;
- Does the contribution relate to a computer program “as such” (ie, excluded matter)?; and
- Is the contribution technical?
If the contribution of an invention solely comprises excluded matter, then the invention is not patentable. The Court of Appeal clarified in Symbian v Comptroller General of Patents (2008) that the key question is whether the invention reveals a technical contribution to the state of the art. This is subject to the rider that purely excluded subject matter does not count as a technical contribution even if it is novel and inventive.
In order to help answer whether a claimed invention reveals a technical contribution, the courts have developed a set of five ‘signposts’ to find whether the application discloses a relevant technical contribution to the art; these were refined by the Court of Appeal in HTC Europe v Apple (2013).
Does an invention reveal a technical contribution? The five ‘signposts’:
- The claimed invention has a technical effect on a process carried out outside the computer;
- The claimed invention operates at the level of the architecture of the computer, ie, the effect is produced irrespective of the data being processed or the applications being run;
- The claimed invention results in the computer being made to operate in a new way;
- The claimed invention makes a computer a better computer in the sense of running more efficiently and effectively as a computer; and
- The claimed invention overcomes the perceived problem as opposed to merely circumventing it.
These signposts are by no means conclusive and recent case law demonstrates that determining whether a computer program is patentable remains a challenge even with the assistance of these signposts.
A useful illustration is the decision in Lantana v The Comptroller General of Patents, Designs and Trade Marks (2014), where the Court of Appeal upheld a rejection of a patent application covering a method of extracting and transferring files from a remote computer to a local computer using email.
The contribution of the invention was avoiding the need for a continuous connection to the remote computer, thereby reducing the risks of being hacked or the connection dropping. The Court of Appeal applied the five signposts in turn before ultimately concluding that the invention avoided the technical problem, but it did so by conventional means.
Accordingly, the contribution to the art consisted solely in the software itself, which was excluded subject matter. Thus the contribution was not technical in nature. The Court of Appeal confirmed that this conclusion would be reached even though the invention taken as a whole was novel and inventive.
The Lantana decision therefore demonstrates the courts’ intense scrutiny of patent applications involving computer-implemented methods. Applicants must be able to show a clear technical contribution to the art which goes beyond the computer program itself—otherwise the application is liable to fail.
The US view
Unlike the UK’s express exclusion, patent law in the US draws no ‘bright line’ rule concerning the patentability of software. In the US, patentable inventions are broadly defined under 35 USC §101 as “any new and useful process, machine, manufacture, or composition of matters, or any new and useful improvement thereof”.
The statute does not enumerate specific exclusions, but US courts have long held that §101 contains an implicit exception for “laws of nature, natural phenomena, and abstract ideas” (Alice v CLS Bank, 2004). So, in the context of software patents, lawyers and inventors must grapple with the questions: Is an application patentable under the §101 test? If so, is it nevertheless excluded by implication?
A brief review of the evolution of software patents in the US is instructive. In 1972, the US Supreme Court held that a programming method—essentially a numerical algorithm—claimed as a process under §101 was patent-ineligible because “the patent would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself” (Gottschalk v Benson, 1972). The decision indicated that, absent new legislation, the court would read existing patent laws strictly to exclude such software.
In 1981, the court adopted a more accommodating interpretation of the patent laws, finding that abstract mathematical formulae are not patentable, but holding that a process incorporating a mathematical algorithm could be patented where the invention as a whole performs a function that the patent laws were designed to protect—eg, “transforming or reducing an article to a different state or thing” (Diamond v Diehr, 1981).
Following Diehr, lower courts grappled with the boundaries of software patentability, but by the late 1990s software patents had become quite popular, in part owing to a 1995 determination of the commissioner of patents and trademarks that “computer programs embodied in a tangible medium, such as floppy diskettes, are patentable subject matter under 35 USC §101” (In re Beauregard, 1995).
“THE US APPROACH POST ALICE BEARS AN INCREASING RESEMBLANCE TO THE UK APPROACH AS THEY BOTH INVOLVE LOOKING FOR SOMETHING ADDITIONAL TO THE COMPUTER PROGRAM ITSELF.”
So, software patents generally could be obtained if drafted as “manufacture” claims under §101 for computer-readable media. Other court decisions further liberalised the patentability of software.
In more recent years, that general trend has reversed. In one notable decision, Bilski v Kappos in 2010, the Supreme Court held that the petitioner’s process claim concerning a mathematical formula to hedge risk was a patent-ineligible abstract idea because hedging is a fundamental economic concept and “allowing petitioners to patent risk hedging would pre-empt use of this approach in all fields”. The Supreme Court’s 2014 decision in Alice further called into question the patentability of software claims.
Post-Alice patentability
The Alice ruling is widely regarded as a major retreat from the more liberal approach to software patentability in the US. The invention in question—involving the use of a computer system as a third-party intermediary to manage risk in financial transactions—effectively amounted to a computerised escrow system.
The Supreme Court unanimously invalidated the patent, holding that the abstract idea of an “intermediated settlement” (eg, escrow) is patent-ineligible and the method claims setting forth its “generic computer implementation” failed to transform the abstract idea of intermediated settlement into a patent-eligible invention.
The court affirmed the two-step framework set forth in Mayo v Prometheus (2012), a case concerning whether a medical treatment method rested essentially on a law of nature and thus was patent-ineligible. The Mayo test asks first whether the claim at issue is directed to a patent-ineligible concept, such as an abstract idea or law of nature. If no, then the claim is patent-eligible and may be patented subject to other provisions of the patent laws.
If yes, then the court must consider the second Mayo prong: whether “the elements of each claim both individually and as an ordered combination ... transform the nature of the claim into a patent-eligible application” [internal quotation marks omitted]. The court described the second prong of the Mayo test as a “search for an ‘inventive concept’” to ensure that the patent in practice would amount to significantly more than a patent upon the ineligible concept itself.
Mayo two-step test applied in Alice
Step 1: Are the claims at issue directed to a patent-ineligible concept, eg, laws of nature, natural phenomena or abstract ideas?
