Is Your New App Patentable? Uncertainty Surrounding Computer-Related Inventions

February 1, 2013

Edmund J. Walsh

(as published in IP Today)

Corporations like Microsoft, Google, and Facebook—and the billions they have generated— have provided proof that software can be inventive, but the Supreme Court has weighed in again and drawn us back to the dark ages of uncertainty surrounding computer-related patents, raising questions whether such inventions are too “abstract” to merit patent protection.

The US Court of Appeals for the Federal Circuit (Federal Circuit), charged with interpreting patent law, has issued recent decisions that have magnified this uncertainty. Although this court and others have attempted to develop a succinct test for computer-related inventions, no test has survived the Supreme Court’s review as being a definitive test. Despite the uncertainty, recent cases provide some clues about how companies with computer-implemented inventions can maximize their chances of protecting their business with patents.

Computer-Related Claims under Federal Statute

The relevant statute (§ 101) states:

[w]hoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.

Patents directed to computer-related technology possibly qualify as patent-eligible subject matter under the categories of “process” (formulated as method claims), “machine” (apparatus or system claims) or “manufacture” (often formulated as computer-readable medium claims).

Despite the many ways computer-related technology might fit within one of these categories, patentability has never been guaranteed. Over the years, judicially-created exceptions have arisen, excluding claims directed to abstract intellectual concepts. Problems arise for computer-related process claims because they are often drafted abstractly to encompass myriad computer systems on which an invention might be implemented.

Enter Gottschalk v. Benson

Our era of uncertainty began 40 years ago in the US Supreme Court case, Gottschalk v. Benson, in which the patentability of a technique for converting binary-coded decimal numbers to pure binary form was challenged. The Court found the claim at issue to be too abstract, viewing it as a fundamental idea or algorithm. The Court worried that allowing such a claim could exclude fundamental R&D tools from the public domain.

The Benson Court also noted that, administratively, the patent system was not prepared to handle the burgeoning numbers of computer-related applications. Because a Presidential Commission had just rejected a proposal that “programs” be patentable, the Court acknowledged an uncertainty as to whether programs were allowed by statute. Against this backdrop, the Court issued a strong caution that:

Indirect attempts [to patent programs] by drafting claims as a process, or a machine or components thereof programmed in a given manner, rather than as a program itself, have confused the issue further and should not be permitted.

Despite advances in technology in the 40 years since, this caution has resurfaced as a tool for refusing patent protection for computer-related inventions.

The Impact of Recent Cases

Challenges to the patentability of computer-related claims subsided after the landmark 1998 State Street decision, and patent practitioners settled into a comfortable groove for getting patents granted. A disruption occurred in 2010 when the Supreme Court, in In re Bilski, relegated the settled machine-or-transformation test for determining which computer-implemented inventions could be patented to being merely a “useful clue”.

Several perplexing Federal Circuit decisions followed. Some decisions and dissents combined other requirements for patentability (novelty, non-obviousness) with the identification of patent-eligible subject matter. Others viewed the inquiry as simply a threshold question. Some cases focused only on the language of the claims, while others focused on the gist of the invention. These recent Federal Circuit decisions have failed to provide clarity as to how patent-eligible, computer-related, inventions can be separated from unpatentable abstract ideas.

A notable criterion for invalidating all types of claims within a computer-related patent emerged from the Benson caution and became entrenched as a “form-over-function” sword. An example of its use appears in the Bancorp Services v. Sun Life case (decided July 2012). The patents were directed to computer-implemented methods and systems for tracking values of life insurance policies subject to market volatility. The court first evaluated the system and computer-readable-medium claims to be equivalents of the process claim. The court invalidated the process claim as directed to patent-ineligible matter, and swept the apparatus (machine) and computer-readable-medium (manufacture) claims into the same invalidity bucket as clever attempts at claim drafting, i.e., form over function. Such a readiness to ignore the words of the claims that recited things squarely within the categories deemed by statute to be patentable came as a shock to many patent practitioners.

Within the first 10 months of 2012, the Federal Circuit decided two other notable cases, CLS Bank International v. Alice Corp. and Dealertrack v. Huber. CLS involved patents directed to computer-implemented methods and systems for tracking volatile asset values of two parties obligated under a purchase and sale agreement. The patents in Dealertrack were directed to a computer-aided method for processing auto loan applications.

The primary analysis used in Bancorp and Dealertrack to invalidate the claims followed an “underlying invention” approach. In Dealertrack, the claims were notably distilled down to a few paraphrased steps. The analyses looked at the underlying inventions and declared them either abstract or nothing more than an automation of manual tasks. The CLS claims survived the § 101 microscope, because the court took an opposite approach, analyzing the claim as a whole in light of the specification, and spoke out against paraphrasing and over simplifying a claim. Whether CLS stands, however, awaits a full court rehearing.

These seemingly inconsistent results highlight the problem of determining whether computer-related subject matter is patent eligible, or whether it constitutes an unpatentable abstract idea. With enough work, any claim can be distilled to an abstract idea, and the ultimate conclusion on abstractness depends on a subjective decision of when to stop distilling.

The Federal Circuit attempted to counter this potential problem in Research Corp. v. Microsoft (decided December 2010) by deeming claims unpatentable only if they are “manifestly abstract.” However, as more recent cases show, this view has not been universally followed.

Measures to Bolster Patent-Eligibility

Despite the discord and uncertainties, there are measures patent drafters can take to increase the survivability of claims to computer-related inventions. A first and important step is to gain a detailed understanding of the invention. Patent attorneys strive to present every invention in broad terms, not restricted by any specific implementation. Describing the details of code or system architecture could render a patent too easy to design around. Yet, the tried and true practice of drafting a broad description greatly increases the risk that a claim will be deemed too abstract to be patent-eligible.

Balancing these competing concerns means that a patent drafter must really understand the invention, and discern new, commercially valuable functionality. To provide the best chances of surviving a challenge, a claim must express that functionality in a way that encompasses how competitors would use a machine to implement it.

The CLS, Dealertrack, and Bancorp courts, though arriving at opposite results, were united in their views that patent-eligible, computer-related claims should set forth a specific way in which a computer or processor is integrally involved in the method, and that the computer involvement must do something more than could otherwise be done manually. This is particularly important for areas of technology directed primarily to financial services and information exchange. Merely executing a process faster or with greater efficiency using a computer has been deemed insufficient to confer patent eligibility—which forces a patent drafter to really understand how a specific structure or operation in a computer system leads to the benefits of the invention.

Equally important is detailing these aspects in the specification. The specification should not merely summarize what the invention can do, but should leave no doubt that it is intended to be implemented with some computer hardware. Detailed flow charts have sometimes been key in saving a claim. Pertinent sections of computer code may also help sell a court that the invention is more than an abstract idea.

It is also important to assess whether the claims reach for too much scope. Every patent holder wants a broad claim, but for computer-implemented methods, a broad claim more easily runs the risk of being found abstract, particularly by courts who draw in requirements of novelty and non-obviousness into the abstractness analysis. Worse, the fatal contagion can spread from method claims to invalidate system and computer-readable medium claims. Analysis of the prior art before claim drafting may provide extra benefits in these scenarios. It may also be worth filing a separate application with only system claims, avoiding contagion from a process claim deemed too abstract.


Confusion over the test for computer-related subject matter has raised the bar for protecting valuable computer-implemented inventions. A deeper understanding of the invention, and its interaction with a computer system, will be necessary to draft claims that survive the test for excluding patenting of abstract ideas—if and when that test is ever articulated by the courts.