Deven Desai wonders if copyrights might be just as harmful to innovation in the software industry as patents:
It seems to me that copyright with its rather long term and (as Mr. Lee acknowledges) the ease with which one can obtain copyright protection could have as much if not more detrimental effect than the software patents to which he is opposed. I would offer Dan Burk and Mark Lemley’s Designing Optimal Software Patents which argues “Optimal software patent doctrine would constrain scope to deal with patent thicket while lowering the non-obviousness standard to validate more issued software patents” as a place to begin investigating this question but also welcome input about this debate.
I’ve actually read the Burk/Lemley paper and have been meaning to write about it for some time. But I don’t have time to dig into that now, so let me instead just quote an excellent response from Michael Chermside:
First of all, keep in mind that today we *HAVE* software copyrights. Under US law (and most other countries are similar) all code is copyrighted except the miniscule amounts that written for the US Govt or are intentionally granted to the public domain. (I don’t know of any code old enough to have passed out of copyright due to age, but it may exist.) Occasionally there are even lawsuits over copyright violations — SCO, for instance, accused IBM of copying some of its code and incorporating it into Linux. To me, it feels somehow “fair” that when one person (or company or whatever) spends hours writing a program, that another cannot simply start using it (without permission). The important difference from patent law is that if person A has copyrighted something and person B wants to do the same thing, they MAY… they just have to do the work themselves. (Copyright law *is* robust enough to handle situations like an attempt to copyright a loop or some other basic pattern — there are allowances for similarity if it really does arise by coincidence not copying.) This contrasts with patent protection where if person A has patented something then person B is out of luck. No matter how badly B wants it, no matter how hard B works, B isn’t allowed to do that thing without a license from A. And *that* is what obstructs progress.
Exactly. Copyright only allows the copyright holder to control software they created themselves. They don’t allow them to restrict the creation of competing software from scratch. In contrast, if somebody gets a patent on “checking email wirelessly” or “translating phone numbers to IP addresses,” he gains the power to shut down any software performing that function regardless of whether or not it’s derived from the patent holder’s original product. As we’ve seen in recent years, this opens the door to all sorts of harmful rent-seeking.
I agree with Desai that copyright terms should be shorter, and there might be other ways that copyright could be improved. But I think it’s hard to argue that the copyright system as it exists today is a net negative. It’s both far more effective than patent law at rewarding innovation and far less prone to abuse.