If you thought software patents were bad...

Okay, this is going to get controversial. Just bear with me.

The Siggraph conference is basically defined by the intersection of art and technology. The ideal Siggraph attendee is both an artist and a technologist, though lots of us are only one or the other (personally, I can't even draw flies).

At Siggraph this year, the well-known patent expert Greg Aharonian gave a presentation arguing that movies and movie scripts ought to be patentable.

You're probably aware that movies, like books and music, can be protected by copyright.

Copyright protection is automatic, but narrow. The owner of the copyright has the exclusive right to copy the work and distribute the copies, to display or perform the work, and to create derivative works (like a screenplay from a novel).

A copyright, however, doesn't give the owner any rights to the basic ideas in the work. Even if you were the first person to write a story about a boy meeting a Martian princess (it's been done), you couldn't use your copyright in that story to stop other people from writing their own stories using that simple plot.

But there is a way to protect ideas under the law. First, you reduce the idea to practice: you design a system or process that embodies the idea. Then you file a patent application. If you're the first person to describe such a system or process, you get a patent. The patent gives you the right to stop others from using (making, selling, importing, etc.) your invention. (It may not give you the right to use the invention yourself, however; if you patent a better mousetrap that relies on a patented bait hook, you can't just start making the bait hooks. But the fellow who makes the bait hooks can't just start making your mousetrap, either.)

That's all clear, I hope, but here's where it gets tricky.

In his presentation, Aharonian points out that Title 35, Section 101 of the U.S. code says that "any new and useful process, machine, manufacture or composition of matter" is entitled to patent protection.

Aharonian's central argument is that just as a piece of clothing is manufactured by a well-defined process, movies are also the result of a manufacturing process. The processes for manufacturing clothing are certainly patentable; almost every new method of weaving and stitching has always been patented. Why shouldn't the processes for making movies be just as patentable?

He isn't speaking here of the mechanical process of aiming a camera at actors, editing together the film, and projecting it in front of an audience--that's old news. I don't know if anyone ever got a patent on live-action movie production, but Walt Disney got a patent on animation.

Lights, camera--patent?
Instead, he's talking about what amounts to the instructions for making a specific movie: a script.

A script defines the appearance of sets, lines for actors to read, camera angles and lighting to be used during the production, and a specific sequence of scenes that express a story. By any reasonable standard, a script is a description of a process. It seems to meet the requirements of 35 USC 101: it's new (in the sense of being original or novel), and it's useful because it tells us how to make a movie.

The resulting movie is itself covered by 35 USC 101 for the same reasons; it's new and it's useful for entertaining a theater full of people.

Studios pay money for scripts, and movie-goers pay money to see movies, which is pretty good evidence that something useful is going on, right?

Aharonian also says that movie scripts meet the requirements of business-method patents, which have become moderately popular over the last decade. In the so-called State Street decision, a federal court held that in order to be patentable, business methods must be concrete, tangible, and useful. A movie script is all of these things.

And Aharonian drew a comparison between a movie script and a computer algorithm. You can't patent a bare algorithm, but if you show how to use an algorithm to achieve a useful purpose, you can get a patent. A script is sort of like an algorithm executed by actors and a crew.

Three more sections of federal law-- 35 USC 102, 103, and 112-- define specific tests for patentability. To qualify for a patent, an invention must be novel (new), non-obvious, and described in sufficient detail that it could be implemented by a person of ordinary skill in that particular area.

Once again, Aharonian explained that movie scripts meet these tests. But here, I think, he started to do a little bit of hand-waving to support his case.

It seems to me that it would take a really unusual movie script to be "novel" in the sense required by 35 USC 102. (Here's where I ought to say: I Am Not A Lawyer. This is not legal advice, just my opinion.)

It's sometimes said that there are only 20 basic plots in all of literature. Or 7, or 27, depending on who you ask. Is "boy meets Martian princess" really new in light of "boy meets girl?" Aharonian points out that patents have been issued in technical subjects on similar grounds. I don't think that's good enough, and I'll point out that Aharonian makes a living by busting just this kind of bad patent.

The recent Supreme Court decision in the case of KSR v. Teleflex was an attempt to make it more difficult to prove non-obviousness under 35 USC 103. This makes me wonder if "boy meets Martian princess" isn't obvious under the combination of "boy meets girl" and "Mars has life." (Edgar Rice Burroughs wrote "A Princess of Mars" after hearing of Percival Lowell's arguments that the appearance of canals on the Martian surface meant there must be an advanced civilization there.)

Finally, Aharonian himself raised the question: is a movie script really sufficient to describe how to make a "useful" (successful) movie? A movie script in the hands of a director "of ordinary skill" is not going to achieve nearly the success that it would if made into a movie by Steven Spielberg. But ultimately I think this is actually the strongest of these three arguments, since I suspect even an ordinary director could make something reasonably entertaining from a good script, and that ought to be good enough.

Anyway, I can't begin to guess whether the patentability of movie scripts will ultimately be upheld by the U.S. Patent and Trademark Office and the Federal courts. I could argue both sides if I had to, but my preference would be to impose extremely high standards.

For example, there must be some disk array over at Pixar that contains all the data and programs necessary to re-render its recent movie, Ratatouille. It must be possible to construct a valid patent around all that information.

But just another unoriginal buddy-cop movie script? I don't think that would make the grade even if one of the cops is a sapient squid. (With a tip of the hat to Bruce Schneier.)

Aharonian gave me permission to distribute his slide presentation (here for now; once he puts it up on his own site I'll change this URL).

He also has a whole website devoted to the intersection of art with patents: http://www.patenting-art.com/

It's worth a visit if you work in this field. It's something you might need to start thinking (or worrying) about...