It has taken almost eight years, but I am happy to share that United States Court of Appeals for the Federal Circuit has sided with Retail Decisions in a very important ruling, which, hopefully, will limit the US Patent and Trademark Office’s ability to grant weak, business method software patents. For full the full ruling, see Cybersource Corporation vs. Retail Decisions, Inc., United States Court of Appeals for the Federal Circuit, 2009-1358.
Some of you may recall my detailed write-up on one of the most influential intellectual property cases in recent history, referred to as In re Bilski. Back in October 2008, I wrote about the US Federal Court’s then-recent decision on Bilski and the potential impact it would have to business method patents, particularly in the technology and software arena (See my blog entry: Bilski from a Non-Lawyer: The Federal Court’s Decision on Bilski and Business Method Patents for full details).
At the heart of Bilsky was the Federal Court’s decision to impose more stringent criteria when evaluating a business method for patentability (See the Wall Street Journal article on October 31, 2008: Court Narrows Definition of Patent Process – A Tougher Test for Protecting Business Methods; Risk-Hedging Strategies Need Not Apply [paywall] or the Businessweek article: Federal Court Rules in Bilski Business Method Patent Case). In the October 2008 Bilski ruling, the court says:
The Supreme Court…has enunciated a definitive test to determine whether a process claim is tailored narrowly enough to encompass only a particular application of a fundamental principle rather than to pre-empt the principle itself. A claimed process is surely patent-eligible under § 101 if: (1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing.
The stressed, latter part of the excerpt above is really what sets the stage for a stricter business method criteria. I won’t get into the deep analysis here – you can read my previous blog entry or the thousands of pages of commentary that have been written about this – but what this essentially says is that you can’t invent a simple process in your head, put it on paper and patent it. In 2008, I quoted the Patent Law Blog, who explained this well:
To be clear, the machine-or-transformation test is not a physicality test – i.e., a claim can still be patentable even if it does not recite sufficient “physical steps.” Here, the court spelled out the specific issue in mind: a claimed process where every step may be performed entirely in the human mind. In that situation, the machine-or-transformation test would lead to unpatentability. “Of course, a claimed process wherein all of the process steps may be performed entirely in the human mind is obviously not tied to any machine and does not transform any article into a different state or thing. As a result, it would not be patent-eligible under § 101.” On the flip-side, “a claim that recites ‘physical steps’ but neither recites a particular machine or apparatus, nor transforms any article into a different state or thing, is not drawn to patent-eligible subject matter.”
Along this line, the court also dispelled two rising concerns, noting that (1) neither novelty nor obviousness have any relevance to the section 101 inquiry, and (2) the fact that an individual claim element is – standing alone – patent eligible does not render the claim unpatentable because patent eligibility is considered while examining the claim as a whole.
So why did I care about this?
Well, if you’ve read me previous writings on this topic or know a little bit about me, you’d know that I served as the Chief Technology Officer of Retail Decisions for over ten years. At Retail Decisions (ReD), I led a team that developed a number of industry-leading credit card and payments fraud detection technologies that became the cornerstone for fraud detection for many of the largest retailers, banks and telecommunications companies in the world, including WalMart, Sears, Kmart, Macy’s, Nordstrom, AT&T, Kohls and a large number of their international counterparts. The service is called RedShield and, in aggregate, ReD’s fraud prevention systems saw over 19 billion of the world’s 190 billion payment card transactions in 2010. The RedShield platform could essentially be considered my “life’s work”, if I ever had one, and I was referred to, by some, as “The Father of RedShield”, even though there were a large team of great people who actually created and evolved the system over the years. (RedShield and Retail Decisions are both still going strong, however, I left the company to join American Express in 2011).
In 2004, Cybersource Corporation (now owned by Visa, Inc.), filed claim against Retail Decisions, stating that ReDShield infringed on their patent no. 6,029,154, which, essentially, sent much of my professional life into a whirlwind over the course of many years. Defending your own invention, hard work and livelihood via the route of the US patent system and federal courts can simply suck the life out of you and I do not wish this on anyone. This was particularly frustrating given that the ‘154 patent is a poster child for the thousands of weak, “soft patents” that have been granted over the years (I’m looking at you, Amazon 1-Click Checkout). There was nothing novel about it, it was peppered with prior art and the techniques described in the patent went into absolutely zero detail about how the invention was actually used to detect or prevent payments fraud. But the one to blame for even allowing this suit to be born, however, is the US Patent and Trademark Office.
Bottom line: This joke of a patent should never have been granted, period.
But little did I suspect that the path to defending this patent would take over seven years…
Fast forward to March 2009: After years of preparation, the case finally made it to the courts and, thankfully for us, it was Bilski that drove the initial judgement in our favor. It’s easiest to just quote the March 24, 2009 edition of Intellectual Property Today:
Chief Judge Marilyn Patel ruled from the bench that all asserted claims from CyberSource’s U.S. Patent No. 6,029,154 are invalid based on the patentability standard articulated in In re Bilski. The patent relates to processes for detecting fraud in credit card transactions. CyberSource had alleged that the innovative ReD Shield and ebitGuard fraud detection products offered by Retail Decisions infringed several claims from the ‘154 patent. The ruling represents a victory for Retail Decisions in a case originally filed in August 2004.
According to Judge Patel, CyberSource’s patent did not satisfy either prong of the “machine-or-transformation” test In Bilski, and the patent was completely lacking innovation. “The court’s ruling is very important.” said Carl Clump, Chief Executive Officer of ReD. “The court’s decision demonstrates that ReD has always acted fairly and responsibly. We are disappointed that this case had to go to court to demonstrate this.”
“This is a significant result for Retail Decisions, and it should have a positive impact on consumers and merchants across a variety of industries,” said Scott Bornstein, Co-Chair of the Patent Litigation Group at Greenberg Traurig and lead counsel for Retail Decisions. The Greenberg Traurig Litigation team also included Allan Kassenoff, a shareholder in the New York office and Jim Soong, a shareholder in the Silicon Valley office. James Myers from Ropes and Gray also assisted in the case.
This was of course, excellent news and, at the time, was somewhat of a landmark decisions as it was one of the first cases won using the “machine-or-transformation” criteria set forth in the Bilski decision. As time passed, a number of similar decisions were handed down from courts around country.
Celebration was relatively short-lived, however. The impact of Bilski was starting to become widespread and, in 2010, the Supreme Court decided to take on the Bilski issues by hearing Bilski vs. Kappos (130 S. Ct. 3218, 561 US __, 177 L. Ed. 2d 792 (2010)), which was ultimately decided on June 28, 2010. In this case, the Supreme Court affirmed the judgement of the lower court but revised several important aspects of their decision. The most important revision was related to the good ol’ machine-or-transformation criteria, where it was held that this criteria should not be the only test used for determining patent eligibility, but should only be used as a useful tool in that determination:
The machine-or-transformation test is not the sole test for determining the patent eligibility of a process, but rather a useful tool. Bilski’s application, seeking a patent on a method for hedging risk in the commodities market, did not draw to patent eligible subject matter. Affirmed.
This was, of course, important in that many cases had recently been decided by examining this key machine-or-transformation criteria, including “my” case, Cybersource Corporation vs. Retail Decisions, Inc. As a result of this, Cybersource filed appeal on the original federal court ruling, since the machine-or-transformation issue was at its heart. In other words, it was time to go back to court…
It took more than another year, but the ultimate decision from the federal court of appeals also ruled in Retail Decision’s favor. The courts finding will, undoubtedly, have major implications in the granting of new business method patents and I am sure there are a number of attorneys filing briefs at this very moment attempting to overturn patents using this case’s outcome as the basis to their arguments.
So what did the court find?
First, the court looked at Claim 3 of the ‘156 patent and ruled that it did not meet the machine-or-transformation test. However, it did not stop there. As I wrote earlier, the Supreme court ruling says that failing to meet the machine-or-transformation test should not, in and of itself, disqualify an invention’s eligibility for patent. However, in this case, the court added that the claim “is drawn to an unpatentable mental process – a subcategory of unpatentable abstract ideas…All of claim 3’s method steps could be performed in the human mind, or by a human using a pen and paper.” In other words, the claim was too simplistic and you could easily execute it without much computation beyond pen and paper. In my words, the claim was simple and obvious, which I had held all along.
Second, the court then examined Clam 2, which essentially described another simple “pen and paper” process and ruled that executing this simple mental process on a computer medium does not make it patentable. Per the opinion: “”[M]erely claiming a software implementation of a purely mental process that could otherwise be performed without the use of a computer does not satisfy the machine prong of the machine-or-transformation test….”. These types of claims tied to computer-readable medium are often known as Beauregard claims. This is essentially saying that if the underlying software method is not patentable, then the so-called Beaurgard claim is also not patentable.
So what does all of this mean? Well, from what I read, the consensus seems to be that this ruling doesn’t necessarily invalidate patents that make Beaurgard claims, but it certainly opens them up to greater scrutiny. And how widely used are such claims, anyway? Well, Sunlight Research did a cursory analysis, looking at the existing patent portfolio of a few large organizations:
I have checked the portfolios of Motorola, InterDigital, Nortel and Kodak for claims directed to “computer readable medium” and have the following results to report:
Portfolio, No. of Patents/Apps with Beauregard Claims
This data does not mean that all of these patents have invalid claims. However, the patents should be checked to identify the extent of the damage. If anyone in the group would like to have one or more lists of patents/apps, let me know…
The most surprising results are the large number of Beauregard claims in the Kodak portfolio. These are the image capture claims, regarded as the patents in that portfolio that have value.
To the common software developer, however, I think I shall stick to my assessment of the initial Bilski ruling, back in October 2008:
From an inventor’s standpoint, business method patents are good. They allow me to protect my intellectual property and secure an exclusive place in the market for my technology. The patentability of these business methods, however, can work against you, as well.
Take the Amazon case, for example. I am one of those in the camp that truly believes that Amazon’s one-click checkout was not novel, fairly obvious and should not have been patentable. As a result of this patent, competitors are now barred from using such a feature on their own website, despite the fact that there really isn’t anything too innovative about it. These so-called, low-quality “soft patents” can truly stifle innovation by limiting the tools, techniques and features that technologists can keep in their arsenal.
I guess that if you’re inventing a lot of things, Bilski isn’t a good thing for you, since it may limit the scope of what you can protect. For everyone else, Bilski just might stop you from infringing on something which might have been patentable prior to today’s ruling.
One thing I can guarantee is that this isn’t the final chapter. But for me, and Cybersource vs. Retail Decisions, it’s nice to have some closure.
Note: I’d usually link to a few article of interest, but there have been, literally, thousands of write-ups about this in the past week. Search “Cybersource vs. Retail Decisions” for more details.