3 ms·

> Sometimes the government allows companies to get a certain amount of rent -- for example, the royalties from patents, which we protect in an attempt to encour

by a3n 9y ago

> Sometimes the government allows companies to get a certain amount of rent -- for example, the royalties from patents, which we protect in an attempt to encourage innovation. This is the classic, maybe the reason given to justify patents. The implication being that no one would bother innovating if patents didn't grant them rent for a time. But I wonder if the reality is merely that different people, and different kinds of people, would innovate in a patent-free world.

digi_owl 9y ago
The problem is not the concept, but the current application. And that is not even a construct of the law as written, but the courts application of said law. Patents started out as a 20 year protection on mechanical mechanisms or chemical processes as long as what was protected was publicly documented in full. Thus once the patent had run out, anyone could replicate it from the patent document. But then the chemical process side used to argue for a quasi-software patent in court, because the patent described a computer monitored mixing process. And thus the ball got rolling. Thing is though that 20 years is a very long time when we are talking software. By the time the RSA patent ran out, the algorithm described was largely obsolete. On top of that we have gotten a mass of patents that are so generic in terminology, that even if they describe something mechanical they can potentially be applied to something done in software. Not that patents have not been a problem even before computers. Serious refinements of the steam engine for example didn't happen until after the initial patent ran out. Similarly Smith and Weston sat on their refinements for the Colt revolver until the patent ran out.
- jb613 9y ago
  > Patents started out as a 20 year protection on mechanical mechanisms or chemical processes as long as what was protected was publicly documented in full. 1) Patents are not an American invention. England had them before, and the Romans before that. 2) I'll assume you implied U.S. patents. The Founders regarded patents so highly that they wrote them into the U.S. Constitution. It was written generically and not limited or fixed to only "mechanical or chemical" but rather "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". > Thing is though that 20 years is a very long time when we are talking software. By the time the RSA patent ran out, the algorithm described was largely obsolete. RSA patent expired nearly 20 years ago yet RSA is still used today. For example, the public key used to secure https://news.ycombinator.com https://news.ycombinator.com is RSA. > Not that patents have not been a problem even before computers. Serious refinements of the steam engine for example didn't happen until after the initial patent ran out. > > Similarly Smith and Weston sat on their refinements for the Colt revolver until the patent ran out. The myth is that patents block innovation. The reality is that blocking someone from doing something incentives them to find another way to do it. Afterall, if all you want to do is copy then how is that "blocking innovation"?

by a3n 9y ago

digi_owl 9y ago

The problem is not the concept, but the current application. And that is not even a construct of the law as written, but the courts application of said law. Patents started out as a 20 year protection on mechanical mechanisms or chemical processes as long as what was protected was publicly documented in full. Thus once the patent had run out, anyone could replicate it from the patent document. But then the chemical process side used to argue for a quasi-software patent in court, because the patent described a computer monitored mixing process. And thus the ball got rolling. Thing is though that 20 years is a very long time when we are talking software. By the time the RSA patent ran out, the algorithm described was largely obsolete. On top of that we have gotten a mass of patents that are so generic in terminology, that even if they describe something mechanical they can potentially be applied to something done in software. Not that patents have not been a problem even before computers. Serious refinements of the steam engine for example didn't happen until after the initial patent ran out. Similarly Smith and Weston sat on their refinements for the Colt revolver until the patent ran out.

jb613 9y ago
> Patents started out as a 20 year protection on mechanical mechanisms or chemical processes as long as what was protected was publicly documented in full. 1) Patents are not an American invention. England had them before, and the Romans before that. 2) I'll assume you implied U.S. patents. The Founders regarded patents so highly that they wrote them into the U.S. Constitution. It was written generically and not limited or fixed to only "mechanical or chemical" but rather "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". > Thing is though that 20 years is a very long time when we are talking software. By the time the RSA patent ran out, the algorithm described was largely obsolete. RSA patent expired nearly 20 years ago yet RSA is still used today. For example, the public key used to secure https://news.ycombinator.com https://news.ycombinator.com is RSA. > Not that patents have not been a problem even before computers. Serious refinements of the steam engine for example didn't happen until after the initial patent ran out. > > Similarly Smith and Weston sat on their refinements for the Colt revolver until the patent ran out. The myth is that patents block innovation. The reality is that blocking someone from doing something incentives them to find another way to do it. Afterall, if all you want to do is copy then how is that "blocking innovation"?