The trouble with patents?
An article in yesterday's New York Times (free via CNet) about Microsoft's manic goal to file 3,000 patents a year (and they're on track, incredibly) is up front about certain of its prejudices (the author is for open source development and a ban on software patents), but some of the prejudices that bother me are more subliminal. Take this paragraph:
"Abolishing software patents would be a very good thing," says Daniel Ravicher, executive director of the Public Patent Foundation, a nonprofit group in New York that challenges what it calls "wrongly issued" patents. Ravicher, a patent lawyer himself, says he believes that the current system actually impedes the advance of software technology, at the same time that it works quite nicely to enrich patent holders. That's not what the framers of the Constitution wanted, he said.
So, am I hearing Ravicher right: Is enriching patent holders a problem? Is the underlying message that the pursuit of wealth is somehow improper? Was Jefferson really a commie? Oh dear.
This kind of statement frightens and angers me. As an artist, I have first-hand experience of trying to toil in an area of human endeavor that our culture has declared to be "above" issues of commerce. Our cultural bylaws state that you can't make "true" art to make money (though if you're lucky you can make a lot of it by accident). Artists themselves wrote these bylaws--they ghettoize themselves by frowning upon commerce, and ultimately marginalize themselves by being "above" the market. I see it as a form of self-loathing. It's part of why I grew disillusioned with the arts. So when I see software folks gleefully building themselves an ivory tower and preparing to climb up and throw away the key, I feel dismay in the pit of my stomach.
In short, I'm tired of always seeing the debate over open source v. patents framed in terms of gentle communalists vs. evil capitalists. To me, the real underlying issues are far more mundane, and far more fascinating.
First, here's what I have always taken as the true role of the patent in our culture: Before patents, businesses were only able to protect their turf through secrecy. Trade secrets create lots of problems from industrial espionage to product safety ("Whaddya mean peanuts are part of your secret recipe? You coulda killed me."). They also stifle innovation.
What? Most who argue against patents say that patents stifle innovation--but in fact, patents offer the very protections that allow all of our innovations to be published and shared openly. You may not be free to use someone's patented idea--but because it's patented, you are free to read it, and learn. The patent library is a public record of our collective innovations. And if you want to use the technology, you can--but you must pay a licensing fee or strike a deal with the patent holder.
This is a supremely civil system. Why shouldn't the inventor receive the financial benefits of his or her idea for a period of time? And why shouldn't we all be gentlemen and -women about it and pay each other, while maintaining the open flow of our greatest capital--our ideas. If you ask me, the right to patent is right up there with freedom of the press.
There's a logical absurdity to saying that we should revert to secrecy in order to be free. Secrecy is necessary only if you have no legal recourse, i.e., if it's legal to steal someone's work. And just because it worked in the early days of software development isn't to say it is the system that should be institutionalized. No more than saying we should still settle arguments by duelling with pistols they way they did in the Old West.
Now, is Microsoft out of control? Absolutely. Is the hyperproliferation of software patents creating obstacles for developers? To be sure. And a system that worked for widgets might now work well for something like software. But there is a baby in that bathwater!
It seems to me that what we need is a new category of intellecutal property. What about a five year term of protection and a streamlined filing process--enough to let software developers get to market? What about a semi-open system based on a retail model, where anyone who wanted to use your software could, but had to bid for the license fee on an open market (rather than liciencing through laboriously negotiated deals)? Affordable, automated licensing could help enrich patent holders (expecially via the Long Tail) while maintaining the free flow of ideas.
I may have to think some more about what kind of system would work best. But one thing's for certain: if I do come up with a new way to issue patent software, the first thing I'm going to do is file a method patent.
ADDENDUM: Just read this from Mike Masnick on the Globalization Institute blog:
Isn't it time we started looking at real solutions for patent reform that recognize the fact that the current system obviously can't scale? This certainly suggests that a system of peer reviewed patentsthat allowed various experts in the field to weigh in on patents could work better. Other ideas include a more "open source" approach - where patent applications are published much earlier along with an easy way for anyone in the public to object to the patent and provide prior art. These are solutions that distribute the work outwards and aren't relying on the same patent examiner system that obviously cannot handle the load created by our existing approach to patents.


I'm not sure how people got the idea that when you get an idea, you have some sort of right to have the government guarantee that you profit from it exclusively, for any length of time. Perhaps they should patent that idea. There's no real evidence that patent protection, any more than any other government project, advances innovation or progress. The process introduces many distortions and much waste.
And (grammatical note): "man" means "man or woman", when used non-specifically. Why waste words?
Posted by: Robert Speirs | March 23, 2006 at 06:10 AM