Monday, January 10, 2005

Free Games, Part One

I have some things to say about this article, Hackers, Slackers, and Shackles. But it's a long bunch of things to say, so I'm splitting it in two. I'll post the rest in a day or so.

Before I say anything else, let me say that I mostly agree with the central argument of this article, that computer gaming as a whole would benefit from more open code. I've believed for some time in what Stallman is quoted as saying, that "A game scenario can be considered art/fiction rather than software. So it is okay to split the game into engine and scenario, then treat the engine as software and the scenario as art/fiction." In other words, open the engine, but protect the "content".

In fact, I've thought for a long time that the easiest way to "fix" the entire software-intellectual-property morass is as follows:

1) Repeal all copyright legislation passed since at least 1976, if not earlier.
2) Transfer functional computer software to patent law, rather than copyright.
3) Eliminate patents altogether.

Reasoning: In its original form, copyright law was pretty inoffensive, and in fact served largely as a set of limitations on what the copyright holder could do (e.g., you can't sue to prevent "fair use"). Returning to that would eliminate most of the current problems with IP.

From a libertarian perspective, I justify having copyright law of any sort as merely a sort of standardized license. What I mean is that from a free-trade perspective, I can justify "shrink-wrap licenses" - I, as a seller, have the right to refuse to sell to you unless you agree to the terms of my license. There is some benefit to having the terms of such a license be standardized from one product to another (avoiding confusion in the marketplace, for example), and as I said, early copyright law limited how restrictive that license was allowed to be.

Next, I know many people panic at the very thought of "software patents", but since step 3 in my plan is to do away with patents entirely, hopefully they will listen to my reasons for doing it this way.

Originally (simplifying a bit), copyrights were intended to apply to expressions of ideas (i.e., art), and patents to functional applications (i.e., machinery/processes). To me, a piece of software has more in common with functional machinery than with artistic expression. In fact, one of the basic tenets of computer science is that any software is theoretically reproducible in hardware - you could build a mechanical Microsoft Word device, if you really wanted to. It'd be big and ugly, but it could be done (and not to imply that the real MS Word isn't big and ugly).

The reason software is covered by copyright rather than patent law is that when the first cases reached court, the software companies wanted copyright protection, because it offered greater protections than patents. So they argued that since software code is, more or less, readable by humans, it is analogous to literary expression, which would be copyrightable, not patentable. The first judge to rule on this, not knowing any better, bought the argument. Well, software was an entirely new endeavor at that time, and there was really no guidance whatsoever in existing law. But we do know better now, and this should be fixed.

(There was, and is, also a body of case law saying that algorithms cannot be patented, and since all software is algorithmic, none can be patented. But then this would also imply that any mechanical device that can be reproduced algorithmically could not be patented, which means any machine that can be simulated in software, which probably means nearly any machine at all.)

Aside: As this Slashdot post suggests, the open source model probably doesn't work as well for "art" as for functional software. With code, there are objective criteria for deciding the merit of a given code change: Does it work? Does it avoid breaking anything else that used to work? Artistic expression is so subjective that every artist working on a project may have a different vision, and what usually happens (movies are a good example) is that either one person takes charge and wrests their own vision into being, or the result is a designed-by-committee monster, tending toward either blandness or overcomplexity, sometimes both. Of course, people are still free to use a Creative Commons license if they so choose.

And as for eliminating patents, this is simply because I find myself unable to come up with a libertarian justification for them. At most, if I'm a seller of something like a car, I might be allowed to have something like a "shrinkwrap license" prohibiting the buyer from reverse-engineering the car. But if you simply take some idea from the car and engineer your own implementation, I can't find any way to justify government intervention to prevent that.

No comments: