Slashdot Mirror
Cato Institute Critique of Software Patents
binarybits writes "I've written an article for the free-market Cato Institute about how patents impede innovation in the software industry. It points out that people tend not to realize how vast the software industry is. It's not just Google and Microsoft; virtually every organization has an IT department producing potentially-infringing software. Organizations as diverse as J. Crew and the Green Bay Packers have been sued for patent infringement. It's crazy to expect all these organizations to worry about potential patent infringement. Hopefully the Supreme Court's Bilski decision will lead to new limits on software patents."
do you really expect rational arguments in favor of the public good to be of any help against entrenched interests in this matter?
Patent attorney here. Yes, software patents shouldn't be there. Patents are there to stop people from sitting on their ideas. Stopping people from sitting on their ideas helps society, because it give society more knowledge. However, for software there is no need for this mechanism. There is no shortage of innovation because of lack of progress. If one person doesn't think of it, another one will. Also, the first sale doctrine doesn't work. If I have a patent on a resistor, and I sell it to you and you put it in a computer, you're free to do that. You don't have that with software. I can't buy a piece of, say, Word, and use it in my own programs. For the same reason, it is very hard to figure out whether your program is off the hook. Any aspect of a program could be patented. Finally, software patents are bad (with so called "wish" claims). I have programs developed for my company. It takes me 5 seconds to come up with an idea, but it may take the programmer 5 days, or 5 weeks, to implement. If I go with a software patent to a programmer, no time is saved.
The patent system is open source avant la lettre. An inventor has to provide all his knowledge (provide the best mode), in a way that can be replicated by an ordinary person skilled in the art, and it is available on line from patent offices. The "license" it comes with is a peculiar one (territorial limited/time limited), but it expires sooner than any copyright. But it is a rough tool. Fine for many types of inventions, including medical drugs, but not for software (or business methods).
Bert
It is not either/OR. So, they get the copyright too. Double whammy.
Bert
Anyone who's written a relatively small amount of software has very likely infringed on someones software patent. I happen to know for a fact I'm an infringer since I wrote software that did a simple zip-code distance lookup web program and years later found out someone actually managed to patent this. The application was taken down years ago because the organization I created it for ceased to exist and had extremely shallow pockets so there's no real danger of being sued over it. I don't recall how I found out about the patent, but it certainly wasn't from looking through granted patents.
The point being it's not that difficult to infringe on someones software patent and have absolutely no idea you've done so. It wouldn't surprise me in the least if I learned I've personally written code that infringed on dozens of other software patents. I'd be extremely surprised if the libraries I use every day didn't infringe on at least one software patent.
What is the greater tragedy?
Blizzard and Microsoft re-invent their gaming tech.
or
Blizzard and MS are at the mercy of Electronic Arts because EA managed to patent something that each could re-create in isolation?
Patents are meant to prevent wasteful re-invention or avoid the extreme case when re-invention is not likely.
The problem with patents today is that patents are being granted for trivial and obvious things that could be easily re-invented by a few undergraduate students.
A Pirate and a Puritan look the same on a balance sheet.
Well, I don't agree with patenting algorithms either.
I think when you think of an algorithm, you have in mind something like an image recognizing algorithm that took years to perfect.
But when I as a developer think of an algorithm I think of very basic building blocks, like binary search, quicksort, hashes, RLE compression, Hamming code. If any of those was patented progress would get slowed down for years.
It can get even simpler than that. Something trivial like "if( there_is_data_to_print && there_is_paper && there_is_ink ) print_document()" is an algorithm.
Allowing patents on this means giving somebody the ownership of a piece of math. That something could be illegal to calculate without paying somebody is completely insane IMO.
As a developer I say: I don't want software patents in any shape or form. Not of the "business method" sort. Not of the "algorithm" sort. There should never be such a thing as a line of code that can't be written without paying somebody for a license. Period.
Copyright is both a positive right (it gives the author the right to distribute and sell his product under his own conditions) and a negative right (you can forbid others to do certain stuff with your copyrighted work).
No, copyrights are purely negative rights. They permit the author to forbid others from doing certain things with their work under certain circumstances. But they provide the author no affirmative rights to do anything with his work. An author's right to create or publish a work is an exercise of his right of free speech and press, no different than if he were to use the work of another in a manner not prohibited by copyright. In the US, this right is guaranteed by the federal First Amendment and by similar provisions at the state level.
This is why, for example, an author could create obscene works, or child pornography, or libel, and have a perfectly valid copyright on them (there's no morals clause to what is eligible for copyright), but not be allowed to publish or perhaps even possess copies.
-- This and all my posts are in the public domain. I am a lawyer. I am not your lawyer, and this is not legal advice.
Your statement insinuates the reverse engineering person somehow has legal rights to the work of others just because reverse engineering the invention is hard.
No.
I am saying that they have that right by default (because knowledge fundamentally cannot be owned), and that reverse engineering being hard means that the fundamental justification for the patent system ("to promote the progress of science and the useful arts") cannot justify taking that right away in such a case.
All patents do is ensure the inventor, the person giving the benefits, is adequately compensated for his work, that's all.
No, they do not. They attempt to do so, but don't do a very good job and have a huge number of bad side effects (like blocking other inventions, turning the common case of simultaneous invention into a lottery, adding overhead to basically all research, etc).