How to Patent Software in a Post Bilski Era
|Written by Gene Quinn
Patent Attorney & Founder of IPWatchdog
Zies, Widerman & Malek
Follow Gene on Twitter @IPWatchdog
Posted: Jun 29, 2009 @ 12:55 pm
Those familiar with the ongoing debate regarding the patentability of software and business methods know, the United States Supreme Court has decided to take up In re Bilski during its next term, which will start in October 2009. In the meantime, what exactly do you do if you want to patent software or computer related inventions? My firm has developed a specialty in handling these types of inventions, and we had success pre-Bilski, and we are having success post-Bilski. As it turns out it might not be as difficult to obtain a patent on software or a computer related invention as was previously feared by some, including myself, in the immediate aftermath of the Bilski decision rendered by the United States Court of Appeals for the Federal Circuit. Nevertheless, there are many attorneys who are advising clients that software is no longer patentable at all. Recently I obtained a client who had been told that he did not have anything that could be patented thanks to the Bilski decision. After we (Mark Malek and myself) looked at the invention we came to the same conclusion. There is no single invention here that can be patented, rather there are about 8 to 12 different inventions that can be patented, depending upon how the patent examiner wants to issue restrictions. The omnibus patent application we will soon be filing covers a really exciting innovation, will be the foundation of what could be a rather large portfolio and should be extremely lucrative for the client.
While it is true that the Federal Circuit has largely made “software” unpatentable, they did not prevent the patenting of a computer that accomplishes a certain defined task. Given that a computer is for all intents and purposes completely useless without software, you can still protect software in an indirect manner by protecting the computer itself, and by protecting a computer implemented process. You see, processes have been patentable since 1790, and despite the misinformed protestations by computer scientists and mathematicians, you can protect software because the software provides the instructions for the computer to operate, can be defined as a process or method like any other patentable process or method. Rather than describing the process as it would be done by a human actor, you simply define the process as it is done by a computer. If you know what you are doing it really is not that hard, at least conceptually. The inability to actually call the invention “software” does, however, mean that the disclosure needs to be much more verbose and perhaps even a bit circuitous.
It is puzzling to me that patent attorneys have not been able to figure out how to get around Bilski. While pure business methods are likely in trouble, processes have always been patentable and with the Patent Office merely requiring reference in the claims to some sort of computer implementation, if anything getting an examiner to recognize allowable matter seems to have become easier since the Federal Circuit’s Bilski decision. Perhaps that is also due to PTO initiatives to start issuing patent applications and working with inventors and their representatives to find patentable subject matter if it exists, I am not sure. Whatever the reason we are having success, and I know others are having success as well. It is not rocket science, although it does admittedly require a certain mindset that many patent attorneys do not have. The key is to think and write the specification as would a software engineer. What is critical is understanding that getting software to work requires far more than math or binary code, despite what computer scientists and mathematicians like to believe. There is a human element to writing software, software is not predictable and creating a working software application requires overcoming traditional engineering hurdles. That is why software is inventive and why it does get protected and rightly deserves the protection it receives.
Many patent attorneys are just not capable of writing patent applications for software or computer related inventions, and I suspect that is why some are telling clients that Bilski presents roadblocks, when in fact Bilski and the PTO interpretation thereof provides a road-map. This is not to cut down or criticize patent attorneys who do not operate in the software/computer space, but rather to point out that this is a peculiar specialty much in the same way chemical and biotech practice is a peculiar specialty. I don’t know anyone who dabbles in biotech, and when it gets outside process chemistry (i.e., A + B, mix and apply) I don’t know many who dabble in chemical applications either. Similarly, there is not much room for dabbling in software or computer related applications. You either do it and understand it or you don’t and should stay away. Of course, if you are a computer sophisticated patent attorney with some level of familiarity with electrical engineering or writing computer code, you could well take to this area like a duck to water. Regardless, if you operate in this space or want to operate in this space and see Bilski as a roadblock you really should sign up for the PLI hour long presentation – Patent Subject Matter: Data Transformation Claim Strategies After Bilski, which will be live via teleconference tomorrow, June 30, 2009 at 1pm Eastern.
So how do you protect software not knowing what the Supreme Court will do? I think the way you need to approach any software related patent application is by realizing first that there is a strong likelihood that the law will change at least once, perhaps two or three times, before the Patent Office ever gets around to reviewing a software related patent application on the merits. With that in mind, it is prudent to have in the specification a generally defined process, a process that is tied to a computer, discussion of what the computer does and how it will accomplish the task and as much more detail as possible. Essentially, what you want to do is have disclosure in the specification that would support claims appropriate under State Steet, under Bilski and under Diamond v. Diehr. I personally cannot see the Supreme Court dialing back so far that software and computer related inventions are not patentable, but I could see them going back to a rule much closer to Diamond v. Diehr, which might mean that we have to start protecting software as a machine, which is intellectually dishonest and naive, but who cares as long as your client’s innovations are protected!
About the Author
Gene Quinn is a US Patent Attorney, law professor and the founder of IPWatchdog.com. He is also a principal lecturer in the top patent bar review course in the nation, which helps aspiring patent attorneys and patent agents prepare themselves to pass the patent bar exam. Gene started the widely popular intellectual property website IPWatchdog.com in 1999, and since that time the site has had many millions of unique visitors. Gene has been quoted in the Wall Street Journal, the New York Times, the LA Times, USA Today, CNN Money, NPR and various other newspapers and magazines worldwide. He represents individuals, small businesses and start-up corporations. As an electrical engineer with a computer engineering focus his specialty is electronic and computer devices, Internet applications, software and business methods.