Wednesday, July 10, 2002

how I learned to stop worrying and oppose IP Send this entry to: Del.icio.us Spurl Ma.gnolia Digg Newsvine Reddit

how I learned to stop worrying and oppose IP OK -- I've been periodically posting anti-IP links without ever giving any real background info. So here it is: My initial IP misgivings grew out of libertarian ideas. I was strongly influenced by Roderick Long's essay The Libertarian Case Against Intellectual Property Rights, which made me appreciate that intellectual property is not a necessary reaction to scarcity (the way that, say, private property in land is) but is rather a grant of monopoly privilege in order to create artificial scarcity. What's more, intellectual property rights conflict with tangible property rights: I own my computer and my printer, but copyright law means that there are certain (say) poems which would be illegal for me to type out (on my computer), print out (on my printer), and sell. That didn't sit well with my libertarian ideals either. At the same time, one of the projects I was working on at MSFT was "licensing compliance." Through monstrous data mining and spreadsheet calculation, I would figure out that Company X owned 100K licenses of Windows 2000 but had 130K computers. (This was harder than you'd think, given Microsoft's arcane licensing schemes and sketchy data.) Armed with my analysis, sales reps would then demand 30K * (a lot of $$) to ensure "compliance." This made me feel thuggish and intuitively queasy. If I buy a CD, I figured, why shouldn't I be able to install the included software on as many computers as I want? I also watched IP laws used to prosecute/threaten Dimitry Sklyarov, Ed Felten, and and 2600 Magazine all for doing things which seemed pretty reasonable to me. So I started asking around -- why should I support intellectual property laws? -- Some people (many libertarians and most Randroids) will argue that IP is like tangible property and that "stealing ideas" is no different than stealing, say, a car. Copying a MP3, they argue, is no different from stealing a CD (even though the former preserves the original while the latter doesn't). But if that's the case, then patents and copyrights should never expire. After all, property rights in land and cars and computers don't expire. To take this point of view, you'd have to believe that using a patented idea the day before the patent expires is "stealing" but using it the next day isn't (which seems ludicrous), or you'd have to believe that patents and copyrights really shouldn't expire. (I do know people who believe the latter.) this was such a long post that it needed a pictureI learned not to argue with the aforementioned true believers -- it was no more productive than a "does God exist" debate between a theist and an atheist. Most people, however are not such zealots and agree with me that "Intellectual Property" is not really property, but merely a grant of monopoly privilege. The disagreement, then, is whether such a grant of monopoly privilege is a good idea. Now if the government granted me the exclusive right to sell hot dogs or operate a laundromat or sell used books, I imagine that even supporters of IP would be outraged. (Most of the ones I know would be.) It's true that none of these is my idea. But should even the person who opened the first cinema / drafthouse be given a monopoly? Should the person who came up with idea of airline hubs? I think we'd agree that this seems a little outrageous. So there must be something about writing a novel, recording a CD, coming up with the idea for an operating system with DRM built in, inventing the hyperlink, or developing a new pharmaceutical which is different from the sort of invention involved in combining movies and beer, or in routing planes through a couple of big airports. But for the life of me I couldn't figure out why some merit monopoly protection and others don't. The main argument is that -- without IP protection -- much of what's copyrighted/patented would be unprofitable and would never get created in the first place. In other words, IP is needed to "incent" the right amount of innovation. But central planning doesn't sit well with me -- when Congress tries to set the right amount of milk production, steel imports, or mohair growth, free-market types rightly complain. In my view, IP protection is no different. It's likely that without IP laws we'd have less production of music, blockbuster movies, books, and new chemicals and drugs. (Though it's not certain -- see below.) Similarly, it's likely that without mohair subsidies we'd have less mohair. Neither of these strikes me as an intrinsically bad thing. First, because I don't trust central planning to provide the "right" amount of innovation. And second, because I believe there are alternative business models which will make up for the lack of monopoly. Jim at Objectionable Content wrote a great (really long) piece about this a few weeks ago:
If you and I want to see a film of Tom Stoppard's Rosencrantz and Guildenstern are Dead, we have an incentive to help Tom, Terry Gilliam, Gary Oldman, and the rest of the cast and crew in making the movie. We want to make sure they don't starve while they focus their energies on creating entertainment for us. If only there was some way to share the fruits of our labor with them in exchange for their work in creating the movie... Perhaps you have heard of such an arrangement. It's called commerce.
In other words, the IP paradigm is backwards. If we're interested in a new CD of Eminem music, we can decree that Em "owns" whatever info he produces and that anyone who uses the information in an unapproved way is a criminal. Or we can set up a market which allows us to pay Eminem to create a CD (perhaps through pre-orders). Both provide incentives for Eminem to create his new CD. But the second has three HUGE advantages: (1) The market decides. Maybe, after 3 CDs, we're all sick of Eminem. Under the IP system, he'll go out and waste a lot of time and money producing and (especially) promoting an album no one wants, which will then not sell any copies. But under the pay for creation system, Em will make a new album only when there's demand for it (or when he's trying to initially prove himself). (2) Enforcement costs. IP creates a class of criminals who must be caught, prosecuted, etc..., and requires pouring lots of resources into costly legal battles. As technology advances, it also requires ever more draconian restrictions on freedom to make it work. These are both huge costs of the IP system. In contrast, the pay for creation system (for example) simply requires a marketplace. (3) No deadweight loss. Under a properly enforced IP system, only those who value the Eminem CD at $17.99 or more get to enjoy it, even though it costs less than a dollar to produce. (A downloaded copy costs only pennies to produce.) Under the pay for creation system, anyone who values the music at more than the cost of downloading (or burning) will get to enjoy it. Now it's true that no one will want to pay the unknown Eminem to create his first album. (Just as you wouldn't want to pay an unproven electrician to rewire your house.) That them's the breaks. another pretty pictureI consider that to be a pretty reasonable case against copyright. The case against patents is similar -- but rather than paying a specific person to create a cure for AIDS, instead I'd help endow some sort of AIDS cure reward fund. (The Clay Math Institute, for instance, has endowed a prize fund for the solution of several hard math problems.) Perhaps we'd use some version of the Wall Street Performer Protocol. And researchers could offer serendipitous discoveries for sale. ("I've discovered a way to make gasoline out of water. I'll make it public in exchange for $10 million.") Would this result in the same level of (say) pharmaceutical R&D we have now? I have no idea. But who's to say what is the "right" amount of R&D? It's not "as much as possible" -- otherwise we'd conscript the world's best and brightest and force them to research drugs. The "post-IP business models" I offered above are only examples; there are others I haven't mentioned (tipjars, bundling, etc...) and others still I haven't thought of. (Item #1 on my research agenda is developing such business models and markets.) Hence, faced with the costs of IP and the existence of alternative business models, I came to the conclusion that IP was a bad idea. -- The argument against trademarks is different, but boils down to the following three points: (1) There are ways to uniquely identify my product (involving, e.g., cryptography or holograms) which are Very Hard to imitate. This makes government trademarks at best superfluous. (2) Trademarks can be (and are) used to intimidate and silence criticism. (3) Sometimes consumers WANT imitation products (fake Rolexes, imitation Gucci bags, etc...). -- (I'm going out for the evening, so I probably won't be able to respond to any comments until tomorrow.)