Software patents must survive: American practice

For about a hundred years, the Patent Office and the Supreme Court of the United States have been fighting and prohibiting the patenting of abstract ideas, natural phenomena, and mathematical algorithms as such by means of precedents. Nevertheless, potential patent owners are struggling to deceive the patent office and the courts of all instances, trying not to fall into the "forbidden" categories, especially in our time with software solutions that are not tangible and can be easily attributed to abstract ideas or programs .



What is especially interesting is that courts everywhere recognize that if an abstract idea has practical application, it is patentable, but dividing technical solutions into practically applicable and not applicable is quite a challenge. And if in the courts of first instance the jury of 12 housewives takes the decision on your invention, then the problem is exponentially more complicated. This complexity is particularly relevant to software and solutions that use medical / biological phenomena (for example, diagnostic methods — it’s also not easy with them now). Very often, in these areas, abstract ideas are the most important elements of technology and the courts must draw a clear line between inventions with the practical application of ideas and just ideas as such. Unfortunately, this distinction often eludes ships.



In determining what your invention belongs to and whether it can claim patent protection in the United States, you should pay attention to the most significant major precedents to which the United States Patent Office or the courts now often refer in their requests. After analyzing several cases, you can understand the concept of the definition of abstractness.



For example, the very famous case of Bilski v. Kappos (2010) , in which the Supreme Court confirmed the refusal to grant a patent for an invention, which described how to hedge losses in one segment of the energy industry by investing in other segments of this industry, based on the fact that this is an abstract investment strategy. Although the method described that it can be implemented through a computer, the court did not accept this argument. It is not enough to add a processing device, a sensor, a data storage to the formula of your invention in order to avoid abstraction.

')

Next was the Mayo Collaborative Svcs case. v. Prometheus Labs , where the court analyzed the patentability of an invention relating to adjusting the dose of a drug based on the response of the individual patient to the drug. The court acknowledged that the patient’s response to the drug was a natural law and decided that this method could not be patented. This decision was very controversial. Many of our colleagues - state patent attorneys who work in biomede, said that the invention formula (which forms the scope of protection) was simply poorly written. However, the precedent has arisen. In this case, in fact, for the first time, a two-step test was used to determine whether the idea was an abstract idea or not. This test is called the Alice test, because it was first described in a very large case by Alice Corp. v CLS Bank . According to this test, if the idea is abstract, the court should see if the invention contains “something more” that constitutes the “idea of ​​the invention”. It would seem to come up with a good test and now everything will be fine.



Unfortunately, the courts again have difficulty interpreting "something more." Maybe soon come up with a test by definition "something more"? And here it smacks of recursion. As a result, many patent attorneys and inventors in the IT field are very concerned that it is not clear what to predict how the court will conduct the Alice test.





Interestingly, in a fairly new case of November 1 ( Amdocs v Openet Telecom ), the appeals court in the federal district ( CAFC ) stressed and honestly admitted that there was still confusion among the courts on the application of the Alice test, and recommended just trying to analyze any similar enough cases applying this test.



What is this “something more”?



Basically, in practice, this is a software or hardware solution that can be technically improved in a special way. In other words, if the invention is directed to a technical improvement in the performance of a computer (or a specific component), then it can be patentable (at least it should be). It became noticeable that the American practice is moving toward another test for "technical improvement". A technical solution should solve a technical problem through some technical improvement. What is particularly interesting is that in Russia this practice has been working for a long time and there is a clear concept that the applicant, when filing a patent application, must indicate and is called this “technical result”. The technical result is a characteristic of the technical effect, phenomenon, properties, etc., objectively manifested in the implementation of the method or in the manufacture or use of the product, including when using the product obtained directly by the method embodying the invention. It turns out that the US patent practice is approaching the Russian one through precedents.



It can be concluded that software can be patentable if it offers a technical solution to a technical problem. This is a very important point to which special attention should be paid. Of course, one should not forget about the criteria for patentability: world novelty, inventive step (not obvious to an expert in this field) and industrial applicability.



The reality is that any law must come back to normal and provide some predictability. The applicant must understand what he may face. Unfortunately, for the time being we see a “transitional period” in the practice of the USA. Software can be very difficult (from an architectural point of view, not programming), innovative (a buzzword), and meaningful to society, therefore, to say that software cannot be patented or software patents will soon disappear, probably recklessly. Patenting software would have fewer problems and many developers would not have been negative about it if the inventors who were abusing themselves did not try to patent an abstract heresy, trying to capitalize on this and block the development of the industry.