Patenting a Matlab script

Hi, does anyone have experience about patenting a Matlab script ?

My code is not a function itself ( like ceil, stifness or magic() tu make some examples ) but i use some of the Matlab function to make something.

I appreciate if you write your knowledge about.

Thank you

[Walter Roberson]

It appears that you might be in Europe. The requirements to patent software in Europe are a bit complicated. If I read correctly, the technical innovation might be patentable but not the steps themselves. That looks to me to be nearly the opposite of the United States approach. In the United States, "algorithms" as such cannot be patented but concrete steps can be patented.

I would suggest that you should be examining resources specific to the jurisdiction you hope to patent your software in.

https://en.m.wikipedia.org/wiki/Software_patents_under_the_European_Patent_Convention

[Jan]

@Davide Di Luzio: What is the purpose of the patent? Should it limit the usage of your code by others? Would a written Non-Disclosure-Agreemen solve your needs? Do you need a kind of copy-protection? Should the patent be active in Europe or worldwide?

I'm surprised that your question concerns a script. Scripts are short hacks, which can have severe side-effects, so they are a bad programming pattern and avoided strictly for productive work.

[Davide Di Luzio]

Hi. Thanks for your reply.

The situation is very simple.

i'll make an (absurd) example : suppose that you write some code ( as a script but developable even as an application ) ,that calculates the design of a beam that can resist some various load. You are the first that thinks to use some algorithm in Matlab to front this problem. Now you want the title of merit about the fact that you are the first to make this and that your program is very helpful about every-day work about the argument ( beam design ) .

The patent be active worldwide

[Jan]

@Davide Di Luzio: So the algorithm is existing already and you are the first one, who has implemented it in Matlab?

I assume, that you will not get a patent on this. If there is no MP3 encoder in Matlab and I'm the first person, who implement this in a new programming language, this is not enough "intellectual creation" for a patent.

Anyway, this question has a very weak connection to Matlab only. Please search for a forum about laws or ask a lawyer who is speziallized in this topic.

[Davide Di Luzio]

nope.

I implemented it using Matlab, but on my search, no one makes it in any other programming language.

For sure my question is not on MATLAB only but i'm searching on experience of other people that have already front this problem.

Answer 1

[Steven Lord]

You're unlikely to get a definitive answer to a legal question, especially a very broad one like this one, here on MATLAB Answers. I know I wouldn't want to even appear to give such an answer, since I wouldn't want to open up myself or MathWorks to a lawsuit if I gave you incorrect advice.

You really should contact a lawyer in the country in which you're hoping to patent your invention. If you did this work as part of academic research or for your job, your educational institution or company may have in-house counsel or a contract with an external law firm that you can contact.

Answer 2

[John D'Errico]

Adding to what @Steven Lord has said, it is not that easy to do. And it is may even be the wrong path to take too. (Even though I have a few US patents in my name that are not so different from what you are asking to do. In my case, the patents came for free, as part of my job. And the company supplied the lawyer.) Anyway, a patent lawyer can help you a great deal. START THERE! Talk to a good patent lawyer. But even then, consider if a patent will really help you. (Of course, it is in the best interests of the patent lawyer that they do work with you on a patent, so they are may not be giving you completely unbiased advice. They earn their fees from you.)

Now, suppose someone steals your patent. Remember that when you write a patent, you are giving away a complete description of how to do what you are doing. Essentially, you are publishing your idea. Hey world, here is my great idea! Suppose some nefarious person decides to use your idea without your permission, without paying you royalties. Remember, this is software. It is pretty easy to hide code behind compilation, P-code, etc. Can you know they have stolen your idea? Can you prove it? And you will need to prove that in a court of law. And that means you will need to hire lawyers again, just to help you prosecute the person who you claim stole your idea.

This means you need to be able to clearly identify the results of what your code does. Else, how will you know someone stole your idea? You will also need to defend your patent there too, that your idea was not just something obvious that you took from other work that was already in the public domain. Prosecuting a patent can get EXPENSIVE.

What does all of this mean? It means that people who do have code they believe is innovative, etc., often do NOT patent it. Instead, they may sell it in a variety of ways, hidden behind layers of encryption.

At best, my suggestion is if you have an idea, that you THINK may be patentable, then I would establish that you did have that idea, on a certain specific date. Have your statement notarized, to establish the date. Now, if it turns out that someone else in the future decides you have infringed upon THEIR work, you now have an ironclad proof of when you came up with the idea.

[Walter Roberson]

Patent gives you the right to sue for purported infringement. It does not give you the right to win the lawsuit. If you end up being up against a company such as GE or Bayers then you can expect a vigorous and very expensive defense.

[John D'Errico]

That is a point I was hoping to make. Patents were fine when I was working for the little yellow box. They had lawyers, lots of them, and a big multi-national company is sort of the gorilla in the room - pretty hard to argue with. But a patent for an individual, on something that may be difficult to even know/prove you were infringed upon, will be a diffcult case to win.

[Walter Roberson]

I knew a fellow who had a patent on a particular form of MRI machine. Truly brilliant fellow; I have no doubt at all that he invented that particular form. Patent was stolen by a major company. Fellow sued... and lost, absolutely buried in legal expenses. He was very bitter about the experience.