Why Software Patents Are Bad, Period

We have two types of intellectual property protection. Patents and copyright. And they are oriented towards too different categories.

Copyright is designed to create a working market for artists, writers, musicians, etc. These people's work is characterized be low output and hit and miss success. A writer might complete a two dozen works and two of which make any money. So you need protection on the order of a human lifetime.

Patents. Patents are totally misunderstood. The core of a patent isn't an idea. Nor is it the answer to a question, how do we do X. Really the core of the patent is the question itself. Knowing which questions are important is what is hard. And patents traditionally apply to physical things, because the physical world is very hard to reason about. It takes a lot of work to figure out and that is what needs protection. The proper period of protection is the course of a product lifecycle. So 20 years, something like that.

Software doesn't fit into either of these. Primarily because for most software the output is determined very mechanically from the requirements. (Unlike, gee a flying car would be nice, or write me a beloved novel that people will still buy 50 years hence). And software ages like yesterdays fish. His comment is, Windows 3.1, 25 years old and who cares. But it's copyrighted for another 75 years.

So software needs it's own category of protection, yes but that is much shorter.

Patents are designed to "promote the progress of science and the useful arts".[1] In the case of software, patents are a net drag on innovation and thus not worth the cost. Software developers are innovating in spite of patents, not because of them.

[1] "The Congress shall have power ... To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries"

Also, this author is apparently unaware that "pure software" patents have never existed, and do not exist today. This is especially true in the post-Enfish world we now live in.

The author is spot-on that obviousness is broken, that's true. Also a good point that patent quality is ill-defined. But those of us who work in this industry absolutely know it when we see it.

Overall, this article doesn't do a great job of explaining what is actually broken and why software patents suck. But saying that software as a class should not be protected like other forms of labor is insulting to software developers. What a software developer does is just as important as what a mechanical engineer does.

- I got a software patent once, on a blockchain related idea. Why did I patent it? Because my partners thought that if we didn't, we'd be in a bad position if things took off. So then we had to spend time and money researching what lawyer to get, explaining to them what the invention was, and so on. Time that could be better spent building a product. And to be honest, the idea was trivial to anyone who knows how to code. But that leads me to another point...

- The fact that things can be patented leads to a culture of ideas being some sort of resource, like gold or oil. They're not. And we shouldn't deify people who have good ideas. Ideas are simply a natural consequence of exposure to an environment. Literally every idea ever has been "let's take X and Y, and use it for Z". Stick someone in the world, and they will have ideas. Sometimes two people will have very similar ideas.

- A friend of mine is an inventor. Of the old school, the kind of guy who has a 3D printer because his inventions are objects rather than software. Like any tech guy, he gets ideas. Every time he gets one, he has to spend time looking for an existing patent. He's canned several projects because something turned up. For instance my wife started doing some research in the medical field because he'd come up with some device, and it basically ended up getting canned because someone in China had done something similar. He's spent significant time training himself in how to do a patent search, which is not a trivial thing to do.

The market should just be free. You invent a better mousetrap? Why shouldn't I make it and sell it? Maybe I can sell it more cheaply. Or I can make one that fits the needs of the consumer of some segment better. Why should I be stopped?

And what about the argument that it incentivises research? I don't know what the state of the evidence for this claim is, but I would think that just about any interesting line of research will be known to more than one group of researchers.

For instance, DNA was being looked at by several groups before Watson and Crick published. In science space, it's somewhat ok, maybe it's bad that one group gets a lot more credit than those who went down a different line, but those other guys can still publish papers with valuable information.

But if you're researching a cure for cancer (imagine it's just a certain unobtainium complex), and someone else finds it first, you are screwed. In fact, if you know that many people are making progress on a problem, you have a powerful disincentive to work on it. In fact, you could even have an equilibrium where a line of research is not pursued because someone else might get ahead.

The econ 101 idea that people will invent stuff on the basis that their investment will be protected is naive. They may never get to the point where they're protected.

Aside from that, look at fields where intellectual property is not protected from copycats. Imagine if you could only buy suits from one brand. Or if baseball caps only had one logo. And so on. There's plenty of fields where me-too marketing is beneficial to society.