joegratz.net

As my co-worker Elizabeth Rader says, “frivolous copyright infringement suits aren’t just lottery tickets.” The plaintiff sometimes gets burned. It happened last week in the case of Mattel v. Walking Mountain Productions. An artist took some obviously parodic photos of a Barbie doll. Mattel sued. The artist won, and the judge awarded the artist $1.5 million in attorney’s fees and almost $300,000 in other costs.

Mattel should have heeded the wise words of Judge Kozinski of the Ninth Circuit, who had earlier advised the company “to chill.”

Two op-ed pieces in this morning’s New York Times respond to the RIAA’s most recent round of lawsuits, both suggesting the implementation of alternative compensation systems for P2P file sharing. The first is by Terry Fisher of Harvard Law School; the second is by Kembrew McLeod of the University of Iowa.

In a really brilliant move, some attorneys (and interns?) over at the EFF have prepared this draft complaint for a suit over the iPod. The complaint demonstrates, in direct, concrete terms the ways that Orrin Hatch’s recently-introduced INDUCE Act will harm consumers and the IT industry. This complaint would definitely survive the summary judgment stage because the analysis required by the INDUCE Act is so fact-intensive — so even if the record companies eventually lost, which is far from certain, they would be able to chill technological innovation with the threat of lengthy court battles.

Bravo.

Ernie Miller dissects Orrin Hatch’s floor statement introducing the INDUCE Act. Very well done.

The Second Circuit has issued this opinion in Capitol v. Naxos which explores some very interesting, very tricky issues regarding sound recordings published outside the United States before 1972. In a somewhat anticlimactic move, they end up asking the New York state courts to tell them what to do.

Ernie Miller responds to my post about the cost of RIAA risk. I think he’s mostly right, arguing that some legal enforcement of copyrights against P2P users is necessary to prevent a drastic shift in copynorms and destroy P2P networks from the inside out by causing users to download and never upload.

I agree that my numbers may be off. I don’t know how many uploaders there are, and I haven’t assumed a decline in the uploader population as the pool shrinks and the risk increases. But the 2.95 million figure I use for the total P2P population may be conservative enough to correct for these, since there are a lot more than 2.95 million copies of the client software out there.

My argument was not that any sue-the-consumer, scorched-earth litigation strategy was doomed to fail in a world of rational P2P users, only that this one may be doing more harm than good to the RIAA. The real potential weakness in my argument, I think, is that it leads to the conclusion that the RIAA should probably be even more obnoxious. For example, if they sued 10,000 people a month, the monthly cost would increase to $10.17 — more than a Rhapsody subscription. Or, in the unlikely event they were able to extract even one full $150,000 statutory damage amount from each of their 532 defendants each month, the monthly cost of risk would increase to $27.05 — that’s plenty of new tracks from iTunes each month.

Of course, “assume a rational consumer” is a statement sure to get chuckles from all but the most hard-core Chicago School adherents. People aren’t rational, and the RIAA’s strategy maybe working for that reason. People (me included) always overestimate risks, and rarely do much math before they decide whether to undertake one course of action or another. But, at least with the current litigation strategy, few consumers have any reason to be worried.

So the RIAA has sued a few hundred more people. Ho hum.

What do we know about these suits? Well, they seem to be suing a maximum of 532 people a month. The settlements have been about $3,000 each.

Armed with these data, we can calculate how much each sharer’s monthly risk is worth. There are at least 2.95 million people on the FastTrack network (Kazaa). I’ll be conservative and assume that this number represents the entire P2P population, since people may use more than one network. Assuming the RIAA picks its opponents at random, this means that, each month, each sharer has at most a 0.018% chance of getting sued. Getting sued costs $3,000. So, each month, the risk of getting sued costs fifty-four cents.

This does not bode well for paid services that do not provide much added consumer value.

Of course, there are many other costs to P2P file sharing. Spoofing, interdiction, bad tagging, slow downloads, and poor search interfaces can also add to the cost.

My point, though, is that the RIAA is just making everybody hate them, needlessly. If people actually ran the numbers, they’d see that while there are many rational reasons not to use P2P networks to infringe copyrights, fear of these RIAA suits shouldn’t be a significant one.

I attended this morning’s Ninth Circuit en banc oral agument in Gator.com v. L.L. Bean along with Elizabeth Rader and incoming CIS Residential Fellow Colette Vogele. It was my first time attending a Federal appellate oral argument. Here are my impressions.

The Gator.com appeal deals with personal jurisdiction — whether it’s fair for one party to haul (or “hale”) another into court in a given location. Gator.com, as you may know, is a… uh… purveyor of… valuable advertising services to… uh… consumers who choose to install their software, which comes… bundled… with various no-cost software packages. (Whatever they do, they don’t make spyware, I’ll tell you that.) Anyway, the Gator software made Eddie Bauer pop-ups appear over the L.L. Bean website whenever one of Gator’s… uh… customers visited the site. L.L. Bean sent them a cease-and-desist letter telling them to stop, since, they said, the software infringed L.L. Bean’s trademarks. Gator is in California; L.L. Bean is in Maine.

When they got the letter, Gator sued L.L. Bean in Federal court in California, asking the court to tell Gator that their pop-up software didn’t infringe L.L. Bean’s trademark rights, as L.L. Bean’s letter claimed. (For you non-lawyers — this sort of “backwards lawsuit” made under the threat of an impending lawsuit is called a declaratory judgment suit. It happens all the time in situations like this.)

L.L. Bean moved to dismiss the suit for lack of personal jurisdiction. You see, L.L. Bean doesn’t have any stores in California. They don’t have any employees in California. They don’t pay taxes in California. They do, however, sell millions of dollars’ worth of goods to California consumers, and they did send that nasty letter to Gator. Without getting into too much civil procedure weenie-ness, the law isn’t clear on whether selling lots of goods to consumers in the jurisdiction with an interactive website and sending threatening letters to companies in the jurisdiction, without any physical presence in the jurisdiction, is enough to make it proper for the party to be able to be sued in the jurisdiction.

The district court ruled that there was no jurisdiction over L.L. Bean in California; a three-judge Ninth Circuit appellate panel reversed, holding that all of L.L. Bean’s Internet sales meant that they were doing business in California. L.L. Bean moved for rehearing en banc — by a larger panel of appellate judges. (In most circuits, en banc cases are decided by all active appellate judges in the circuit. But because the Ninth Circuit is so huge, they just select 10 judges at random and have them hear the case.)

The en banc panel focused more than I expected on the jurisdictional effect of the cease-and-desist letter. On one hand, L.L. Bean lobbed this potentially damaging document into California, and causing damage within the jurisdiction makes jurisdiction proper. On the other hand, all L.L. Bean was doing was enforcing their trademark rights, which they had a legal duty to do; the C&D wasn’t intended to cause any damage, only to ask Gator to stop infringing. Toward the end of the oral argument, L.L. Bean’s attorney (who was doing a great job with a hard case) made the point that they had no other way to tell Gator to stop, so if the court held that sending C&Ds caused personal jurisdiction to attach, every IP owner would be instantly amenable to suit anywhere someone infringed their IP rights, if they told the infringer to stop.

The panel decision had me convinced that L.L. Bean should be subject to California jurisdiction, but now I’m not so sure. Regardless, it was a great experience, and I really enjoyed watching a dozen smart lawyers hash over this minor, though thorny, ambiguity in the law of personal jurisdiction.

(Incidentally, on the way out, we noticed that L.L. Bean’s attorney’s briefcase was made of blue canvas, with brown leather piping and embroidered initials. Of course.)

So I’ve been using Gmail lately. I like it. But that’s not the point.

Next to a recent email about some public-domain materials, I got an ad for some websites (1 2) that make me sort of ill. This is not what we’re fighting for.

I think the fastest way to make lots of money exploiting public domain material is to charge $97 a pop for directions on how to make lots of money exploiting public domain material.

Frank Field points to this press release from Irdial Records describing their settlement of a copyright dispute with WEA International.

Here’s the story. Irdial put out a CD full of recordings of shortwave “numbers stations” called The Conet Project. The numbers stations are broadcast anonymously and more or less everyone acknowledges they have something to do with international espionage. For this reason, the recordings themselves are probably either not covered by copyright at all (in the case of recordings made by the United States government) or are protected by rights that are extremely unlikely to be enforced, since doing so would blow the broadcaster’s cover.

Wilco sampled one of these recordings at the end of “Poor Places” on their album Yankee Hotel Foxtrot — a numbers station repeating the words “Yankee… Hotel… Foxtrot”. The sample was taken from the Conet Project CD. Irdial sued WEA, Wilco’s record company, for copyright infringement in the UK. They claim, first, that their recording is unique because of the radio interference that surrounds it, and that this interference gives them a copyright in the recording. Second, they edited the recording to make it more interesting. Third, they processed the recording to make it clearer . Each of these, they say, gives them exclusive rights in their recording.

I don’t know UK copyright law very well, so I don’t know whether this claim has more merit there. But under American law, Irdial probably would have lost had the case gone to trial. First, simply recording a radio broadcast does not give a person rights in the recording. A recording of a preexisting transmission does not have the requisite originality for copyrightability. Second, Irdial’s editing may have been sufficient “selection and arrangement” to give rise to a copyright in the whole track, preventing wholesale verbatim copying. But from the description they give, there were no edits within the “Yankee Hotel Foxtrot” sample; the sample Wilco used was an unedited slice of Irdial’s source material, and thus Irdial’s edits cannot have given rise to copyright in the sample. Finally, the equalization and processing. Irdial admits that the EQ was “to remove noise” — not for any creative purpose. This would be the closest issue, I think, but WEA could argue that any new matter added by Irdial in the noise-reduction process is purely functional, increasing the clarity of the recording, rather than constituting protectible expression.

Given that last issue, I understand why WEA settled; this would end up being a fact-intensive dispute that would be expensive to litigate. But I don’t think the Wilco sample, in fact, infringed Irdial’s copyrights.