Editor: Tell our readers about your background and your current practice, including your participation in the Bilski case, as well as your firm's patent practice.
Kiklis: My focus is on software patent law, with an emphasis on litigation and counseling. I have both an undergraduate and a master's degree in computer science and worked as a software developer for six years before going to law school. Since graduating from law school 17 years ago, I have focused in this area. My participation in Bilski was only at the Supreme Court level. My partner, Tom Goldstein, and I co-authored the ABA's Supreme Court amicus brief.
Our firm's patent practice is focused on litigation and counseling. We handle patent cases all over the country and in all technological areas, including many high-stakes patent cases.
Editor: Give us an overview of the Bilski case.
Kiklis: The case began in 1997 when Mr. Bilski and Mr. Warsaw filed a patent application for a method of hedging risk in commodities trading. It described basically a business method directed at human activity and was not tied to a specific apparatus or machine.
The Patent Office examiner rejected the application for failing to constitute patentable subject matter under Section 101 of the Patent Act. Bilski appealed to the Patent and Trademark Office's Board of Patent Appeals and Interferences, which affirmed the examiner's decision.
Bilski appealed to the Federal Circuit, which ordered an en banc review. In its opinion, the Federal Circuit announced a new test for determining whether processes are patentable, known as the machine-or-transformation test. This test states that a process is patentable only if it is tied to a particular machine or apparatus or if the process transforms an article into a different state. Because Bilski's application was directed to human activity, it failed the test.
Bilski then appealed to the U.S. Supreme Court. The Supreme Court relied upon its most recent patentable subject matter cases, Benson, Flook and Diehr, as well as the ordinary meaning of the statute to resolve the case. Diehr, decided in 1981, was the Court's most recent patent case that dealt with patentable subject matter in this area.
The Supreme Court held that business methods, in fact, may constitute patentable subject matter, and it did not adopt any categorical exclusions for any subject matter, such as business methods or software, even though some of the amicus briefs had argued for such categorical exclusions. The Court found that, although not the sole test, the machine-or-transformation test was useful and provided an important clue for determining whether a process is patentable.
The Court then held that Bilski's patent application was nothing more than an abstract idea and, therefore, unpatentable because it attempted to patent a basic idea that would preempt its use in other fields. An abstract idea is one of the Supreme Court's three long-held exceptions to patentable subject matter. The other two are laws of nature and physical phenomena. At the end of the majority opinion, the Court said that it did not endorse any of the prior tests of the Federal Circuit, but it also didn't foreclose the Federal Circuit from adopting other limiting criteria. So, the Supreme Court did not announce any new criteria for identifying patentable subject matter, but rather resolved the case based on its existing case law and the text of the statute. In other words, in many respects, we're back to 1981.
Editor: Would having a machine involved enable a business method to be patented?
Kiklis: Business methods may constitute patentable subject matter. If your business method claim falls within the machine-or-transformation test, you should be safe. If it doesn't, then it will receive scrutiny on a case-by-case basis to determine if it is an abstract idea. We don't have any Federal Circuit cases yet, so it's uncharted territory.
If you have a business method that you want to patent, I recommend that you claim your business method as running in a machine. If you thought it was important to claim your business method outside of a machine, you could do so. However, I'd recommend that you also have a dependent claim that limits the steps to a machine. That way, if you lose the independent claim, the dependent claim should survive.
Editor: Where does that put software programs?
Kiklis: Software remains patentable subject matter. So, for any new computer technology, if you have a method in which there is transformation or if it's occurring within a machine and has sufficient ties to the machine, you should be safe. Software programs in the abstract - a program that is not tied to a machine - might not be patentable, but you should be safe if it's tied to a computer.
Editor: But isn't software always involved with a computer?
Kiklis: Actually , you could attempt to claim it as a sequence of steps without limiting the claim to a machine. But if you did that, then you'd be risking stepping beyond the machine-or-transformation test. Typically in a software patent application, you'd have apparatus claims, computer-readable medium claims, data structure claims and method claims. You only need to be concerned about the method claims. To be safe, you should consider tying your method claims to a machine. However, if the software happens to involve transformation, then perhaps you could step beyond the machine.
Editor: How does this "machine" concept apply to developing areas of medicine, medical devices, biotech, nanotech, communication and finance? Does each of these situations present problems?
Kiklis: The good news is there are no categorical exclusions. So, just because your invention deals with business methods, for example, it doesn't mean that it can't qualify as patentable subject matter. Further, if you can claim your method as running in a machine or if there is a transformation occurring for those situations where a machine isn't involved - say, some sort of chemical process - then your claim should pass muster for subject matter patentability. It's just in those technologies where you have to stray beyond the machine-or-transformation test that you can run into problems.
Editor: Does this mean that patent claims will be drafted differently after Bilski ?
Kiklis: For software and business method inventions, it has always been a good idea to draft patent claims such that the steps of your method are running within a machine. If you recite your method steps as being outside of a machine, then you can run into problems. I don't think that Bilski has changed that.
However, there remains uncertainty for claims that are not operating in a machine or where no transformation occurs, and we really don't have much guidance. All we have is whether it would constitute an abstract idea. A concrete test for determining where that boundary lies seems to be lacking at this point, so that is why I say uncertainty exists for anything beyond a claim in a machine or where transformation occurs. Bilski's invention was found to be an abstract idea by attempting to claim a basic concept that would preempt all uses in other fields. Also, although some of Bilski's claims added limitations, these limitations amounted to insignificant post-solution activities. So, right now, that is about all the guidance we have.
Editor: What do you believe Bilski's effect will be on business?
Kiklis: Having uncertainty is not good. If people perceive that their invention may not receive patent protection, it can have a chilling effect, which is why the patent system is set up the way it is. A broad, expansive view of patentable subject matter promotes innovation.
Editor: Did Justice Stevens in his concurring opinion advocate for limiting patentability to the machine-or-transformation test?
Kiklis: Justice Stevens said that he would not have made the machine-or-transformation test the sole test. In this respect, he agreed with the majority. However, he said that claims that merely describe a method of doing business do not qualify as a process under Section 101 of the Patent Act, so he carves out business methods as a categorical exclusion. Justices Ginsburg, Breyer and Sotomayor joined Justice Stevens in his concurring opinion.
Editor: What kinds of problems might occur if an existing foreign patent doesn't meet the Bilski standard?
Kiklis: Once again, there isa level of uncertainty. Other countries tend to have a narrower view than the United States as to what is patentable. For example, Europe has a technical effect requirement for software inventions. So, for software to be patentable, the claims would have to recite some sort of technological effect. At least in the software area, the U.S. provides broader protection. I don't see Bilski as changing that. It is just adding some uncertainty for technologies or patent applications that wouldn't satisfy the machine-or-transformation test.
Editor: In light of Bilski , how would you like to see the courts interpret the Patent Act in the future?
Kiklis: I'd like to see the courts continue as they always have, which is to expansively interpret Section 101. The Patent Act of 1793 was drafted by Thomas Jefferson, and his expansive language has remained largely unchanged. The statute is a forward-looking statute, and it's intended to cover unforeseen inventions, which it does. I take the Supreme Court's unwillingness to add additional limitations over the literal language of the Patent Act and its prior precedent as a sign that the Patent Act will continue to receive an expansive construction.