Should software be patent-able? As we all know, a lot of companies in United States are based on software. I can think of Microsoft, Google, Amazon, Apple, Oracle, Cisco(although they do hardcore hardware too), Twitter, Facebook. You name it. I am sure that they have some software in place.
We all know that software takes a long time to write. It's also very valuable since we can see that many software(or hardware companies doing software) companies are doing so well in the United States. Some say that software shouldn't be patented because mathematics cannot be patented in the US. Since software is involving heavily with theory and algorithms, it's really up to debate whether it can be patentable or not. In this article, Microsoft, HP, and Adobe tries to prove that software should be patented. I thought this was interesting because I do agree to some extent. But on some areas, it went too far. Microsoft tries to explain their claim through example they describe the evolution of the mechanical typewriter.
This is a direct excerpt from the article regarding Microsoft's statement:
Basically, this is saying "This is the key mistake of the software patent problem: taking earlier mechanical inventions and re-implementing them in software is patentable."Functionality previously achieved through gears and levers—like proportional font spacing — was re-developed using software that provided instructions to computing hardware. Those digital improvements served the same function as, and were no less technologically innovative than, their mechanical forebears. Accordingly, the inventions were patentable.
And they go on to state:
Software is heavily dependent upon the use of algorithms. And it is “clear that a process is not unpatentable simply because it contains a * * * mathematical algorithm.” . . . Where a software patent describes a practical application of an algorithm as part of a process that produces a technological effect or useful result on a computer — e.g., encryption, data compression, higher- speed processing — that should be patent eligible under this Court’s precedent.
This is pretty much stating although it's the law that mathematical algorithm cannot be patented, fields relating to use of encryption, data compression, higher speed processing should be eligible for patents.
And, Microsoft also goes onto state that they will provide service with professional witnesses to USPTO if they cannot understand the patents. The article goes onto state that all software programs are list of symbols that embody ideas. I also thought it was interesting that the author mentioned that software implements abstractions on a computer. But United States court decided otherwise, and therefore, making software companies spend money on patents. So, should all software be patentable? Are Microsoft, Adobe, and HP doing the right thing? These are all valid questions that may come up in the later years, definitely.
Source: http://www.zdnet.com/microsoft-hp-adobe-try-to-justify-software-patents-7000027033/
I had previously thought that software are patentable. I guess that was really the ideas which had a software-backed support. For example, the slide-to-unlock feature is coded but it may not necessarily have to be coded, that is, if there is some other ways to achieve it (if any). Software, as parts, perhaps are not able to achieve anything significant yet until pieced together with something else. In this respect, it is still confusing exactly what about a software can or should be patented. We will have to see how the progress goes about software patents in the coming years.
ReplyDeleteI thought this as well, but not all software are patentable at the moment, especially the ones dealing with theory. It's really up to debate as there's no clear justification which software is patentable and which one isnt. There's a lot of controversy debate going on regarding this topic. You should check out some blogs on the web. There's also ruling on US supreme court, who will decide on this matter. Also I was able to get the sense that it's very subjective based on the ruling.
Deletehttp://www.washingtonpost.com/blogs/the-switch/wp/2013/12/06/the-supreme-court-could-abolish-software-patents-next-year-heres-why-it-should/
DeleteThis is interesting article that you might want to check out.
I really appreciate your focus on the patentability of software. The issue is quite controversial. After reading the Washington Post article, I found it interesting that the author wanted to promote the notion that patents cannot apply to software. There are two main camps to this issue based on this discussion:
Delete-recognize pure software patents for sake of innovation
-deny software patents to prevent patent trolling
It is difficult to determine if there is a need to satisfy both sides. Regardless of the decision, patent trolling will continue for patent litigation with ideas not related to software.
Furthermore, I believe the system is not streamlined to make assessing software patents a smooth process. The first thing that the court system would have to do is determine if the software patent is purely to cover the algorithm or describing a process that involves the software. Perhaps, the USPTO agency would need to develop a specific process for evaluating software as they do with regards to natural phenomena.
Thank you for sharing the article! It is certainly a very complex matter with many implications. It kind of made me think of this example. Let's say, there was a lever with some mechanical connection which interfaces with a pulley, making it easy to lift something heavy. Supposing that it is just recently invented, that would be patentable right? However, if it is a digital lever on a screen, remotely connected to a motor to lift something, assuming the previous invention did not even exist yet, is it patentable?
DeleteIn the mechanical case, a different method of lifting the object such as through a multi-pulley system can probably be patented. However, what happens if a different software is able to lift the object in a different way? I think the software patent appears too strong in this case, eliminating other digital techniques. However, if it is tagged to the mechanical parts as a support, other companies can invent a multi-pulley system using a similar software interface without infringing the patent.
Very interesting. Software patents are not something that I had touched upon heavily in the past. But I'll definitely try to write about them more in some future blogs! Anyhow, I think if we were to keep in mind the notion that if an invention does not build on its mechanical forbears in terms of innovation that it is unpatentable then the same should be said for software. Technically speaking the code for all programs are just recycled from one program to the next. Even though they are arranged differently and the source code differs, in my opinion I don't think that software should have full patent rights. There should be a different classification and even system through the USPTO specifically for software. And I think that in the future as more and more software patent issues come up, the framework will be refurbished and revamped.
ReplyDeleteI agree. From the basic knowledge I have of coding, a lot of it is indeed recycling. It's quite amazing, actually, how two similar codes/frameworks can give completely different results! Knowing this, it'd be unfair to grant a whole software patent rights, as this would hinder others with separate, innovative ideas from moving forward on their separate paths.
DeleteThere seems to be more debate going on regarding software patents. I believe Congress is trying to decide what to do with software patents soon. On some level, there are many who believe software patents should be justified, while others do not. As a computer scientist myself, I believe software to some extent should be patentable. But thinking back how anyone can come up with same algorithm kind of makes it difficult to decide.
DeleteThese questions are even valid today! I'd like to address your analysis of Microsoft's typewriter example: "This is the key mistake of the software patent problem: taking earlier mechanical inventions and re-implementing them in software is patentable."
ReplyDeleteI don't wholly agree with this because I don't see this to be a problem. In fact, once a patent expires, this idea becomes public knowledge again. The fact that someone then takes the time to build on this old yet innovative-in-its-period idea is perfectly sound. Otherwise, how do we expect to advance as a society if not to build on our history?