Saturday, January 4, 2014

Paul Graham - Are Software Patents Evil?

In the article Are Software Patents Evil, Paul Graham does a nice job explaining the role of patents in startups. It is based on his presentation to Google in 2006, which may be long ago, but I think describes patents better than many things I have read in recent years. It also has weight because Mr. Graham is not a lawyer, but a software programmer, a writer, an artist, and a startup advisor. See details in Paul Graham wiki. He is not trying to make us like patents so we will file more, he is trying to grapple with a difficult problem software companies face today.

I suggest reading the article, but for now here's a passage you don't hear often from the software community:

"Patents are a hard problem. I've had to advise most of the startups we've funded about them, and despite years of experience I'm still not always sure I'm giving the right advice.

One thing I do feel pretty certain of is that if you're against software patents, you're against patents in general. Gradually our machines consist more and more of software. Things that used to be done with levers and cams and gears are now done with loops and trees and closures. There's nothing special about physical embodiments of control systems that should make them patentable, and the software equivalent not.

Unfortunately, patent law is inconsistent on this point. Patent law in most countries says that algorithms aren't patentable. This rule is left over from a time when "algorithm" meant something like the Sieve of Eratosthenes. In 1800, people could not see as readily as we can that a great many patents on mechanical objects were really patents on the algorithms they embodied.

Patent lawyers still have to pretend that's what they're doing when they patent algorithms. You must not use the word "algorithm" in the title of a patent application, just as you must not use the word "essays" in the title of a book. If you want to patent an algorithm, you have to frame it as a computer system executing that algorithm. Then it's mechanical; phew. The default euphemism for algorithm is "system and method." Try a patent search for that phrase and see how many results you get.

Since software patents are no different from hardware patents, people who say "software patents are evil" are saying simply "patents are evil." So why do so many people complain about software patents specifically?

I think the problem is more with the patent office than the concept of software patents. Whenever software meets government, bad things happen, because software changes fast and government changes slow. The patent office has been overwhelmed by both the volume and the novelty of applications for software patents, and as a result they've made a lot of mistakes."

Few can describe how patents relate to startups. It takes time to master a domain, but also I think people don't have access to enough experiences. Perhaps his role at the YCombinator investment firm interfacing with literally thousands of software companies generates the data to get it right. See Randall Stross' The Launch Pad Inside YCombinator for details. That's why recent criticisms of his comments about lack of woman in computer science is off-base. He is not discriminating against women, but giving his candid impression based on data on a problem many would like fixed.

Copyright © 2014 Robert Moll. All rights reserved.