Hacker News Reader

I find it very hard to believe that the patent system, broken though it is, is anything other than a HUGE net positive for innovation. I'm talking about the patent system across all industries, but I suspect the same would be true for the tech industry specifically.

The Economist (hardly a bastion of the free software movement) quoted a 2008 study showing:

A study in 2008 found that American public companies' total profits from patents (excluding pharmaceuticals) in 1999 were about $4 billion—but that the associated litigation costs were $14 billion.[1]

Clearly the litigation costs have grown significantly since 1999. It is unclear to me if the profits have kept pace, but even if they have that still would mean patents cost over 3-times the financial benefit they bring.

[1] http://www.economist.com/node/21526370

That seems intuitively wrong. $4 billion across all public companies? Apple makes more than 10 times that on its patented inventions yearly...

But how much of that revenue would they miss out on without patents, taking crosslicensing out of the picture? I suspect very little. People buy iphones for the whole product's look and feel, not for any specific feature.

It's a myth that software must be protected through patents. Copyright and trade dress law is sufficient.

> You have no way of knowing whether this is true.

There is a whole field dedicated to answering this kinds of questions, it is called economics, and there is quite a bit of research in this area analyzing historical evidence from different legal systems, times and industries.

And most of the evidence indicates that patents hinder and stop innovation, and that most innovation and progress happens when there are no patents:

http://levine.sscnet.ucla.edu/general/intellectual/against.h...

You're correct that economics is a science, but you make an error when you say "most of the evidence", and then link to an ideological position.

Science doesn't work that way. Even if the preponderance of evidence were the standard, you haven't provided evidence that the preponderance of the evidence supports your position. And even if every economist in the world shared the same opinion, it would be a meaningless claim (even if you got a petition and they all signed it) from a scientific standpoint.

I believe you are wrong, but I would fall into the same error if I simply made the opposite claims you do, and I cannot prove my position scientifically, so I won't argue it.

You're correct that economics is a science, but you make an error when you say "most of the evidence", and then link to an ideological position.

It is a mistake to simply link to a book and said it is evidence, because that's just being a lazy person. People should be able to come up with their own arguments in a discussion.

That being said, it's an economic book that should be evaluated on its own merit. Calling a book "ideological" is quite unfair unless you're prepared to make an argument against the book's arguments.

>> And secrecy is a patent that potentially never expires.

Do you seriously believe, no one else can come up same/similar functionality/product? I would like a system where they keep the patent secret(nothing is published) for 5 years, and if in 5 years no one comes up with same/similar idea, patent is granted for next 10 years (rather than current 20 years from filing), otherwise patent is not granted. My guess is 99% of software patents will go away.

That would slow innovation to a standstill.. After you invent something, you couldn't use it for 5 years. Everyone would be even more secretive and it would favour big companies who could afford to 1) sit on ideas for 5 years 2) register as many patents as possible.

There is a control group for the impact of patents. It's called software.