An article I published yesterday – Praying the Supremes Get Bilski Right in 2010 – has already started quite a stir, bringing out the anti-software patent advocates. These folks always claim to be innovators and as innovators they know best and they state with an obviously flawed confidence that as innovators they can state with certainty that innovators despise software patents. Of course, this is utter and complete nonsense. Those who are anti-software patent are simply pro copying. They don’t want software patents because then they cannot copy the work of others freely and without fear of being sued. Those in the software industry who are not interested in software patents are not innovators, they are copiers. They steal the work of others. They also claim to have a sophisticated understanding of constitutional law and patent law, but reading what they say makes it apparent to anyone who is knowledgeable that they don’t know the first thing about law in general, let alone constitutional law or patent law in particular. It serves no purpose to retroactively kill patents and applications that could have satisfied the standard announced in In re Bilski, but were written to satisfy the now defunct State Street test. That is changing the rules in mid-stream and violates all ideals of fundamental fairness and due process.
One particular comment demands further exploration because it demonstrates a complete lack of understanding of basic patent law and patent applications in general. In the above referenced article I discussed how it is fundamentally unfair to change the law in the middle of the game and leave at least a decade of patent applications and issued patents worthless. This argument was characterized as “weak” by a anti-software patent commenter who said:
The idea that it will inconvenience a bunch of existing patents, which probably should exist under changed rules anyways, and forcing people to refile patent applications as too much of a hassle is a pretty weak argument indeed.
While I go back and forth and get heated with this regular contributor who goes by the name “POP,” my intention here is not to make fun of him, but rather to address this issue. In my experience as a teacher, when someone has a question it is likely that others have the same or similar question. The law surrounding this is fundamental in the patent world, and explaining in depth the issues will clarify exactly why the Federal Circuit decision is nothing more than utter nonsense and fundamentally unfair, perhaps to the point where the decision creates a constitutional unfairness. It should also be educational for inventors and others who are interested in learning more about the patent process.
First, at the time you file a patent application your invention is defined and locked in. You are not allowed to update your disclosure to add material or information that was not in the application. The only way to include new information, which the patent laws characterize as “new matter“, is to file a new patent application. This new patent application will get a new filing date and the information in the application that was not present at the time of the earlier application will receive a new priority date. An example will likely be useful, and since I am in the process of teaching the first live patent bar review course of the year in Washington, DC, I will use an example loosely based on something in our materials that I discussed earlier today with students.
Inventor Smith files a provisional patent application defining a widget on January 6, 2010. After several months of continuing to work on the invention Smith makes a breakthrough and significantly improves upon the widget. Smith has the option to file a subsequent patent application covering the improved widget. This secondary application could be another provisional patent application or a nonprovisional patent application. Lets say Smith files a second provisional patent application on March 5, 2010 covering the improved widget. Smith would then be allowed to file a nonprovisional patent application wrapping together both provisionals, but there will be two priority dates — January 6, 2010 for the original and March 5, 2010 for the improvement. Even if Smith files a nonprovisional patent application on March 5, 2010 that covers both the original and the improvement, the two versions of the invention will have different priority dates. The rule is that whenever an invention appears in a patent application that is the priority date. Priority dates are important for determining what is prior art. Essentially, something that happens after your priority date cannot be prior art, so having an early priority date can be exceptionally important.
When you file a patent application and leave something out, on purpose or accidentally, you need to add it by filing a new patent application. In the Bilski invention scenario almost all inventions subject to a patent application since the State Street decision in 1998 could be patented under the Bilski machine or transformation test, but the Bilski machine or transformation test is quite different than the useful, tangible or concrete result test from State Street. So those who followed the law announced and considered settled for 11 years in State Street would find that under Bilski they cannot obtain a patent or the claims obtained quite commonly will be invalid. This is not because the invention could not be defined to meet the Bilski test, but because the invention was defined to meet the then existing law.
Now, another illustrative example. Lets say Inventor Doe filed a patent application in 2003 on a Bilski related invention. The application was filed to satisfy the State Street requirements, and the patent issued in 2006. Upon filing the patent application in 2000 Doe started using the invention, which is allowable and would not create a problem under existing patent law. Now the law changes to the machine or transformation test and the application filed to satisfy the useful, tangible or concrete result test does not recite enough to satisfy the machine or transformation test. The claims would be invalid, and that is exactly what is happening over and over again in the district courts across the US. Unfortunately for Doe, since the invention has been in use since 2000 there is no ability to go back and file another US patent application to rectify the situation. Under 102(b) a new patent application that would satisfy the Bilski test would be rejected because the invention has been in public use, and potentially on sale, for more than 12 months, which means no US application can be filed. Doe is out of luck because the law changed and there is nothing Doe can do even though the application could have easily been written to satisfy the Bilski standard had that been the law in 2000. The rules have changed and Doe is out of luck, even though Doe did everything right and followed the law. That is unfair.
A similar result would be the same case even if the patent had not yet issued. Inventor Jones filed a patent application in 2005 on a Bilski related invention. The application satisfies State Street, but does not contain enough disclosure to satisfy Bilski. The patent application gets rejected and there is nothing Jones can do, because like Doe, the invention has been in use and a new patent application would violate 102(b). Jones did everything correctly, followed the well established law and then due to a fundamental shift winds up losing everything not because the invention has changed, but because the way the invention needs to be characterized is now different. I know life is not fair, but this is ridiculous, and historically when property rights are affected in this way the Supreme Court has said changes in the law need to be prospective, not retroactive. Those who order their affairs based on settled law should not have the rug pulled out from under them for having done so, which is exactly what is happening presently.
The patent laws prohibit applicants from making alterations to their applications. That being the case principles of fundamental fairness demand that when substantive changes occur to the legal standards applicants should be given the ability to amend their application to bring them into conformance with the new requirements, or the new requirements be applied only prospectively. Erasing value and forcing applicants to have satisfied requirements that could never be envisioned and are far different than existing legal requirements is not only unfair, but likely unconstitutional. This has to violate due process.
Whether anyone likes it or not, regardless of the outcome of Bilski at the Supreme Court software will remain patentable. Patent attorneys have always been at least one step ahead, and even if the Supreme Court tries to kill software patents we will figure out a way to characterize it so that it will be patentable. It is ridiculous to have to act like the machine is unique when it is the set of instructions that turn the machine into a device that provides a new, unique and cool functionality. Time will tell whether those on the Supreme Court understand patent law, technology and the US economy. I think they do, and I think the resulting Bilski standard will kill pure business methods, but allow for software, medical devices, diagnostic process and treatments to remain patentable, and more easily so.