Software is patentable subject matter in the United States. There are some restrictions, such as that the software must be embodied in a machine, but software inventions are regularly given patent protection. However, there is strong opposition to protecting software inventions, and many countries, particularly European Union nations, do not allow the patenting of software inventions.
The opposition to software patents is based on a number of good arguments including that patents stifle innovation and competition, that many over broad and obvious patents were granted, and that the United States Patent and Trademark Office (USPTO) doesn’t have the necessary expertise to evaluate software inventions. There is some truth to these arguments, but protecting software innovation is more important than ever because so much economic growth and innovation is centered in software products.
It is true that software patents stifle innovation and competition. That is the objective of any patent, to give the inventor the right to exclude others from practicing her invention. All modern societies give these innovation-discouraging non-competitive rights to inventors for one reason – it encourages more innovation than it discourages.
Invention is a difficult, time-consuming, and expensive process. It is simply hard to get something new to work in a useful way. In contrast, copying an invention is easy, quick, and cheap. Without patents, there would be little incentive to innovate because copying is so much easier. We see this today in countries with weak patent protection – copying is far more popular than innovating.
However, bad patents can stifle innovation. A bad patent is generally over broad, protecting far more than the inventor invented. An over broad patent can stop all innovation in a large technology segment by giving one inventor all rights to a very general solution, even though the inventor never developed or even considered all of the solutions that are covered in her patent claims.
Unfortunately, there are a disproportionate number of overboard patents for software inventions. One reason for this is that so much of our software technology is of recent invention. We are close to the Big Bang of software innovation, so many core software concepts are still covered by patents. For example, the explosive growth of the internet is still within the twenty year term of a patent, so many basic software innovations like one-click ordering are protected by patents. This is similar to the early days of the automotive industry when patents covered most basic automotive technologies like steering wheels and drive trains.
As a result, programmers can infringe patents using relatively basic design principles. In contrast, the core innovations of older technologies such as electronic logic or engine design are not longer protected by patents, so basic design principles can be employed without violating patents.
Bad patents are often granted for very obvious inventions. Patents should be awarded for doing something new and non-obvious, and not for being the first to file an application for a predictable, obvious solution. Unfortunately, the USPTO had to meet a very high standard to show obviousness when many applications for software inventions were examined. As a result, patents were issued for some relatively obvious inventions. Since then the standards for showing obviousness have been relaxed, making it much harder to patent an obvious solution. Older patents can also be reexamined using the new obviousness standard, reducing the chances that patent holders will try to assert questionable patents.
Some bad patents were also issued because the USPTO simply wasn’t staffed with examiners with the right expertise to evaluate the explosion of software applications. However, the USPTO has done a good job of adding and training examiners with software expertise. As a result, the examination of software applications has steadily improved, with fewer bad software patents being issued.
While much of the criticism of software patents is valid, protection for software inventions is increasingly important for innovation and economic growth. The factors that created the large number of innovation stifling bad software patents are diminishing as more basic software technology enters the public domain, changes in patent law reduce obvious patents, and the USPTO becomes adept at examining software inventions. The problems and abuses are being corrected.
At the same time the importance of software inventions to innovation and competitiveness is growing. Programmers are solving real problems with inventive, non-obvious software solutions. They should be encouraged by being allowed to patent their inventions. It would be wrong to deny protection for software inventions because they are inventions. And it would be foolish, because the creation of software is a vital economic activity where innovation should be encouraged and protected. Software patents are more important than ever because software innovation is more important than ever.