“But it is a pipe.” “No, it’s not,” I said. It’s a drawing of a pipe. Get it? All representations of a thing are inherently abstract. It’s very clever.”
-John Green, The Fault in Our Stars”
The Alice Vs CLS Case:
The words “software” and “patentability”, have always had issues with staying together. When they ever did, they have wrinkled many a patent examiner’s eyebrows. It makes sense, doesn’t it? SOFTWARE PATENTS ARE ABSTRACT IDEAS!!! ABSTRACT IDEAS DO NOT EXIST IN THE REAL WORLD!! You cannot patent something that doesn’t exist. What if someone patents a thought and sues you for thinking the same thought! So abstract ideas should never be allowed to be patented! Right? NO!! In a world where billions of dollars, business deals, and information related to almost everything under the sun is transacted in a virtual world, made up of technically “abstract ideas”, providing some kind of protection for these so called “abstract ideas” becomes an issue. “Hey!, we just wiped off all the balance in your Online bank account because we thought it to be a little too “ABSTRACT”. Now that’s something that you don’t want to hear from your banker. So lets all first agree to one thing, yes, we need a minimum level of protection for the so called “abstract ideas” that make up software patents.
Now, the reason why we are here talking about the patentability of software, is because of a fairly recent patent litigation between Alice corporation and CLS bank international. The Supreme Court concluded that Alice Corporation’s claims were directed at the abstract idea of intermediated settlement that merely required generic computer implementation, and thus ineligible subject matter under USC § 101”.
To throw a little light on the entire affair, Alice Corporation is an Australian firm that dabbles with the computerized trading of financial derivatives. The corporation holds patents dealing with computer trading of financial derivatives. CLS bank International sued Alice Corporation and sought a declaratory judgment of non-infringement and invalidity of three such patents of Alice Corporation on May 24, 2007. Later, a fourth patent was added to the list. All the four patents were deemed to be invalid by the Supreme Court. Yes, bad news for Alice Corporation. But what is that got to do with the rest of the companies with software patents? Well probably a great deal. One of a controversial opinion from the U.S. Court of Appeals for the Federal Circuit was that this decision could result in the death of software patents (Spoiler alert: It doesn’t).
In light of the recent events, a memorandum dispatched by Commissioner of patents to patent examiners details some new instructions for the preliminary examination of patents. The memo, dated 25 June 2014, outlines a two part test to test the patent eligibility of software patent claims. The two part test, informally called the Mayo framework, was drafted by the US Supreme court in a 2012 suit Mayo Collaborative Servs. v. Prometheus Labs. The test goes on as follows:
In this part of the test, the patent examiners test whether the claimed invention is in effect, an abstract idea. And why do they have a problem with abstract ideas? Let’s hear the answer to that question from the USPTO:
“Abstract ideas are excluded from eligibility based on a concern that monopolization of the basic tools of scientific and technological work might impede innovation more than it would promote it.”
Yes, true story. You would’ve had to do a lot of addition if someone had a patent for multiplication. So abstract ideas shouldn’t be patentable. Simple as that. However, the difficult part is determining what constitutes an abstract idea. For a lay person anything ranging from a thought, to programs, to financial instruments are abstract ideas. In fact, anything that is not tangible go straight into that class. But as we discussed earlier, sometimes we take some abstract ideas like bank account balances for example, a little more seriously. And we do need methods to protect them. Similarly, some inventions have aspects of abstraction that are essential to the invention. In fact, allow me to paraphrase what the commissioner for patents, USPTO had to say about patent eligibility of abstract ideas:
“At the same time, the courts have tread carefully in construing this exclusion [read exclusion of abstract ideas] because, at some level, all inventions embody, use, reflect, rest upon or apply abstract ideas and the other exceptions.”
Yes, be it your average bottle opener, or the Large Hadron Collider, all inventions are implementations of an abstract idea. So usually you have:
NOVEL, NON OBVIOUS ABSTRACT IDEA + its PHYSICAL IMPLEMENTATION = INVENTION
So now we have pretty much of an idea on when an abstract idea becomes patent eligible. It becomes eligible when it is given a physical implementation. But again there is a catch there. The physical implementation of that idea has to be novel. So in reality, it is:
NOVEL, NON OBVIOUS ABSTRACT IDEA + NOVEL PHYSICAL IMPLEMENTATION = INVENTION
An idea of using a lever to open a bottle cap became patentable when someone moulded a piece of iron into a bottle opener. An idea for taking a snake out for a walk became patentable when someone tied a leash around it. (I am not making this up, check out US 6490999).
Now you may ask, then why is there this problem with software patents? In a software patent, you have an abstract idea, which you make into a program, which you run in a physical processor.
Well, it is possible to look into it in this way: The programming code that implements the abstract idea, by itself, is an abstract idea. We load the code into the processor, and run it. We are merely “APPLYING” the particular code onto an existing processor. There isn’t anything innovative happening in the tangible world. We are not physically changing the processor in a way to implement the code. For example, in the case of a bottle opener, we are moulding a piece of iron in an innovative way in order to open the bottle cap with it. Though the topic is still debatable, in Alice Vs CLS bank, the Supreme Court has come to the conclusion that such a software that is basically an abstract idea is not patentable. Now what constitutes an abstract idea is a different question. The Supreme Court used its decisions in previous cases to tackle this problem. Some examples of decisions that the Supreme Court referred to are:
- In a decision for Benson v. Gottschalk., 1972, the court ruled that claims for an algorithm to convert binary coded decimal numerals into pure binary form constituted a patent-ineligible abstract idea.
- In a decision for Parker v. Flook 1978, the court ruled that mathematical formula for computing “alarm limits” in a catalytic conversion process constituted a patent-ineligible abstract idea.
- In a decision for Bilski v. Kappos 2012, the court ruled that method for hedging against the financial risk of price fluctuations is a fundamental concept in economic practice and therefore constitutes a patent-ineligible abstract idea.
Though Benson (1972) and Parker (1978) did not cause much of a problem for Alice Corporation, what really drove the last nail into the coffin of the case was Bilski (2012). The Supreme Court decided that similar to Bilski’s patent claims, the concepts mentioned in claims by Alice were fundamental concepts in economic practice and therefore constitutes a patent-ineligible abstract idea. However, The Supreme Court failed to provide a precise definition of what constitutes an abstract idea. The Supreme Court merely cited a few examples of abstract ideas.
Examples of abstract ideas quoted in the Alice Corp VS CLS include:
- Fundamental Economic Practises
- Methods of organising human activities
- An idea in itself
- Mathematical formulas
The above statements prove that analysing patent claims for abstraction shall continue to be a difficult terrain for both patent examiners and inventors. Especially because there is no definition for an abstract idea and neither does the court give one.
This brings us to the second part of the test. The second part of the test is performed only if there are abstract elements in the patent claim.
In the second step the examiner checks whether the patent claims contain elements to make the abstract ideas more than mere abstract ideas. The claims should not be a mere instruction to apply the software code on a hardware device. The abstract idea must use something more than a generic computer to implement the code. Examples of such cases include:
- Improvements to another technology or technical field
- Improvements to functioning of the computer
- Meaningful limitations beyond general linking of abstract ideas with processors.
In effect, this is where the examiner checks whether the implementation of the abstract idea is a novel physical implementation.
If the patent claims are found to fail the test, the claims are rejected under USC. § 101. And thus, the patent in question would not be granted.
Hence, the number of USC 101 rejections can be expected to be on the rise. Such rejections can hopefully be prevented by keeping in mind the new framework for preliminary examination and drafting the claims to be significantly more than the abstract ideas.
The only question that remains now is, “Is software patentable?”. The answer is simple. It is, if it passes part 1 and part 2. The only problem: The test is notorious for being as slippery as it is simple.
Author, Leo Paul Johnson, is a Senior Patent Engineer at IP ASTRA