joegratz.net

Margaret Mitchell’s Gone With The Wind is in the public domain in Australia but still protected by copyright in the United States. Project Gutenberg of Australia (based, unsurprisingly but importantly, in Australia) posted the book on their site. According to the file, the etext was prepared by Don Larson, whose email address indicates he’s from Canada (where the book is in the public domain).

So the Mitchell estate sent a letter to Project Gutenberg in the United States demanding that the work be taken down, advancing the theory that because the file is downloadable by users in the United States, both the Australian hosts and the Americans who the Mitchell estate claims are in charge of PG Australia are liable.

This quickly becomes a really good civil procedure hypo. Let’s assume that no American had anything to do with the posting of GWTW — an assumption I think is correct, given that it was prepared by a Canadian and posted on a server in Australia. So we have people in Australia doing things that are perfectly legal in Australia, but that are causing damage in the United States. This really brings to a head the question of exactly who’s doing what when I click on a link and download Gone With The Wind from Australia. If PG-Australia were in the United States, the easiest way to nab them would be to say they were making an infringing distribution, and to avoid all the messy stuff about exactly who’s making a copy or who’s causing a transmission. But here, the party making the distribution is doing so in Australia, where distribution is legal. It’s a bit of a thorny question, and I’m not sure what the answer is, or what the answer ought to be.

Thoughts?

The Sixth Circuit today reversed a district court’s holding in Lexmark v. Static Control Components that copying a 55-byte program that enabled aftermarket toner cartridges to work in Lexmark printers was a copyright infringement. This could have been a small win, but it turned out to be a big win. Jason Schultz has one of the choicest quotes; my observations are below.

• It’s quite clear that Lexmark saw this coming when it was designing the system, and that there was a copyright lawyer at the elbow of the programmer designing the authentication system. They very carefully designed a system in which it was impossible to make replacement cartridges without copying a small computer program, and they deliberately included a little bit of non-functional, original expression (the letters “LXK”, the company’s stock symbol) in that computer program in hopes that courts would find the program copyrightable. While the court saw right through it, I’ve got to hand it to Lexmark: were I attempting to pervert copyright law into a tool to enforce tying (which seems to be a popular pastime these days), this is just how I’d do it.
• The court noted that the copied code in this case has two uses. First, it is functional, incorporating an algorithm for figuring out how much ink is left in the tank based on sensor readings. Second, it is a key. The printer will only work if the program on the chip has a certain checksum. This means that the only practical way to get the printer to work is to make a verbatim copy of the program, so the checksum turns out the same. So, even though the program may or may not be copyrightable as an ink-level-sensing program, it is purely functional as the only key that allows the printer to work. The court rightly inquired whether Static Control had any other practical options and found that, while in theory some other program with the right checksum might exist, finding that program was so impractical that the program itself merged with its function (as in cases like Brandir).
• In the court’s fair use analysis, they get the relevant-market analysis really right. The district court ruled that the “effect on the market” fair use factor weighed against Static Control because their copying had a negative effect on Lexmark’s ability to sell its printer cartridges, since the copying destroyed Lexmark’s monopoly. The Sixth Circuit rightly reversed, noting that the relevant market is the market for the copyrighted work, not any market affected by the copying. The court found that to the extent there was a market for 55-byte toner-fill-level-sensing programs (the existence of which the court doubted), Lexmark had made no showing that the market had been harmed.
• In the court’s DMCA analysis, they interpret the statute’s “effectively controls access” language perfectly. Lexmark had claimed that Static Control was selling a method of gaining unauthorized access to a program stored in the printer’s ROM, since that program was accessed when the printer operated and non-Lexmark ink cartridges were not authorized for use in the printer. But the court saw that “access” to the program stored in the printer’s ROM occurs when the consumer buys the printer as well as every time the printer prints something. The program was not encrypted or otherwise protected, and could be read off the chip by any consumer with the appropriate equipment. To use the court’s analogy, it’s as if Lexmark claims that locking the front door while leaving the back door open prevents access to the house, since people can’t come in the front door. The idea that access occurs upon purchase when no encryption is involved is likely to prove a useful precedent.

If courts keep on restoring sanity to the DMCA, we might just be able to live with it.

The BoingBoing bloggers have endorsed John Kerry for president.

In recent days, a growing number of news organizations have posted eloquent endorsements for Kerry. Some of them are particularly suprising, because they come from such unlikely sources. We encourage you to read them, and consider their content.

For us, the choice for Kerry involves simple things. Justice, liberty, privacy, transparency. Freedom of speech, thought, and technological expression. A woman’s right to choose. Equal access to health care, education, and economic opportunity for all. The rule of law, at home and abroad. Peace. The enduring value of the American Constitution.

These are wonderful things. The Bush administration has proven both inability and unwillingness to protect them. In 2004, Kerry is the one.

To paraphrase Siva — Oh. Uh. Joegratz.net endorses Kerry, too.

Richard Epstein has published an Op-Ed in the Financial Times questioning the long-term viability of the open-source software movement. He sees two principal weaknesses.

First, he says, section 2(b) of the GPL has not been tested in court. Epstein believes that courts are likely to enforce section 2(b) with copyright remedies rather than with contract remedies, meaning that a violator would be enjoined from further distribution until the offending code is removed and would have to pay damages to the copyright holder. The GPL would not have what Epstein claims is its promulgators’ “apparent intention” — to “infect” proprietary software and force it to be licensed under the GPL.

But the Free Software Foundation, promulgators of the GPL, have no such intention. Indeed, they see the GPL as a pure copyright license, not a contract. This means that if GPL’d code is incorporated into a proprietary product, a license is violated but no contract is breached. Under the terms of the GPL, all rights to use the GPL’d code terminate when the license is violated, and any further distribution is copyright infringement. At that point, the violator certainly may release the product under the GPL, curing the license violation, if that’s cheaper than rewriting the misappropriated code. But the violator can also just pull the product from the market, or rewrite the offending sections. The GPL Police don’t show up, seize CD-Rs full of source code, then dance in a circle singing victory songs. These are perfectly ordinary remedies, and the Open Source community is operating on the assumption that Professor Epstein is right — that the GPL is a license, not a contract, and that copyright remedies will suffice.

Second, Professor Epstein argues that the structure of Open Source projects is unsustainable. As he says:

The open source movement shares many features with a workers’ commune, and is likely to fail for the same reason: it cannot scale up to meet its own successes. . . . The bottom line is that idealistic communes cannot last for the long haul. The open source movement may avoid these difficulties for outside contributors who work for credit and glory. But how do the insiders, such as Linus Torvalds, cash out of the business that they built? And in the interim, how do they attract capital and personnel needed to expand the business? Traditional companies have evolved their capital structures for good reason.

Though he minimizes it, Professor Epstein recognizes the central feature that separates the open source movement from a workers’ commune. The point of a workers’ commune is that everyone is an insider, sharing equally in the monetary profits reaped through their collective labor. And when an insider wants to leave an organization, there has to be some way to separate out her share. But open-source projects are fundamentally different from workers’ communes in a number of ways that allow them to avoid the problems Epstein lists. Most importantly, where in a commune everyone is an insider, in an open-source project, everyone is an outsider. Everyone has the same rights to the code, and everyone has the same right to “fork” the project, using the same GPL’d code to create a new, improved, or different version of the program. The only thing that separates Linus Torvalds from any other kernel contributor is his reputation. People like using his version of the kernel more than they like using someone else’s, because they like the choices he’s made about what code is and isn’t incorporated. Epstein notes that organizational difficulties may be avoided with regard to contributors who work for “credit and glory”. But as far as the project maintainers know or care, everyone is working for credit and glory. Patches from people who IBM pays to hack the kernel for a living aren’t treated differently from patches submitted by any other talented kernel hacker. There is no central body handing out rewards.

And this is the reason the Free Software Movement won’t collapse under its own weight. When IBM or Red Hat or TiVo hire programmers to work on open-source projects, neither the companies nor the programmers expect to be able to appropriate any portion of their contribution. This is not to say they are acting altruistically, but only to say that the corporations believe — rightly — that under certain circumstances they can maximize their profits by waiving proprietary rights to certain code written by their employees. If corporations continue to provide institutional support for high-profile open source programmers for these reasons, the present model is sustainable. If they cease, the worst that could happen is a return to the old, pre-boom open source development structure, where things moved a little more slowly, but everyone played for the love of the game.

Via Larry Lessig.

I’m a big fan of Wikipedia, a collaboratively-written online encyclopedia that is quite comprehensive and perfect for casual research — satisfying curiosity, linking to unfamiliar terms from blog posts, that sort of thing.

In two good posts today at the Volokh Conspiracy, Eugene Volokh and Orin Kerr question the wisdom of citing to Wikipedia in legal documents.

I’ve written before about the dangers of citing to Wikipedia in briefs or opinions. Professor Volokh expresses concern that if Wikipedia turns out to be mistaken, or if a clerk or judge knows about Wikipedia’s collaborative nature, credibility could be lost. Even more potentially troublesome, though, is the possibility of out-and-out meddling. Were Wikipedia regarded as an authoritative source, an unscrupulous lawyer could insert his preferred facts into the encyclopedia, or frame the facts in a light favorable to his client. Likewise, a net-savvy lawyer opposing a party citing Wikipedia in a brief could edit the cited article, reframing the facts such that the opposing lawyer appeared to be misrepresenting the source’s content — or even change the facts completely. Such meddling could even be carried out anonymously.

This might prove far too tempting, and could turn Wikipedia into even more of a battleground than it already is. So, please, don’t cite to Wikipedia. (The difficulty in bluebooking a cite to a source that changes many times daily seems nightmare enough to recommend against the practice.)

At a rummage sale yesterday, I found a copy of the 1964 yearbook from St. Lawrence University in Canton, New York. I figured it might be a source of silly, anachronistic candids for some future project. But then I got to page 67. Read on for more.

(more…)

Retropod, a site selling ancient Sony Walkmans converted into ironic iPod cases (and supplies for doing the conversion yourself) has been shut down following a cease-and-desist request from Sony.

The letter read, in part:

Sony recently learned that you are selling a case for carrying an iPod personal stereo that is made from a WALKMAN tape player. The product is being offered at your website at www.retropod.com.

Your use of casings for such a purpose is a clear infringement of the SONY and WALKMAN marks because it is deceptive. Consumers likely will be misled and deceived into believing that Sony is somehow connected with the iPod personal stereo when in fact it is not. Moreover, they will be misled into thinking that Sony is backward in its design of products and is going away from miniaturization, as the size of the tape player housing is quite large by today’s standards.

Extremely lame. Sony would almost certainly lose if this went to trial. Here’s why.

There’s some case law on whether it’s a trademark infringement to sell converted or modified goods without removing the original trademarks. For example, while it’s OK to leave the trademarks on a spark plug (Champion Spark Plug Co. v. Sanders 331 U.S. 125 (1947)) or golf ball (Nitro Leisure Products, L.L.C. v. Acushnet Co. 341 F.3d 1356 (Fed. Cir. 2003)) when you recondition it, it’s not OK to replace the original guts of a Rolex with non-Rolex guts and sell it as a Rolex (Rolex Watch, U.S.A., Inc. v. Michel Co., 179 F.3d 704 (9th Cir. 1999)). Cases on whether selling cars converted into stretch limousines are divided, but there seems to be a consensus that selling a conversion service is always legal, while selling converted cars themselves may or may not be.

Repurposing an ancient Walkman is more like reconditioning a spark plug than substituting cheap guts in a Rolex. Consumers don’t think this is a Sony-made iPod case; they know it’s (1) a joke and (2) a conversion. In trademark infringement actions, we can look to the sophistication of prospective buyers. Here, the intended market is iPod-owning hipsters, who, as a group, tend to recognize ironic craft projects when they see them.

The best fact for the defendant in this suit would be the survey in which 99% of respondents, when asked “Is this a Sony product or something else made out of a Sony product?”, answer correctly. Sony might ask consumers, “After looking at this product, does Sony make iPod cases?,” and confuse some chunk of them. But I think Sony would lose that factual inquiry.

Sony seems to be assuming that he’s selling finished retropods; in fact, he was mostly selling kits to convert Walkmans into iPod cases, and selling such conversion kits is definitely not trademark infringement. Selling something called “Kit to convert a Sony Walkman into an iPod case” is merely a descriptive use of the mark, and a contributory infringement claim would fail because, among other reasons, there’s no likelihood of confusion (as noted above).

Still, it looks like the creator of the Retropod didn’t feel like fighting, since the joke was sort of already done with when he got the C&D. All this has really accomplished is making a few hipsters hate Sony, which they probably did anyway.

Jacques Derrida died.

We have a new overbroad copyright bill pending, the Intellectual Property Protection Act of 2004. How overbroad? Well, it certainly seems to criminalize passing a VHS copy of a TV show to a friend.

Here’s how:

It changes the definition of criminal copyright infrongement to include the following in 17 USC 506(a)(3):

infringes a copyright by the knowing distribution, including by the offering for distribution to the public by electronic means, with reckless disregard of the risk of further infringement, during any 180-day period, of … (C) 1 or more copies or phonorecords of 1 or more copyrighted pre-release works

Seems to be aimed at people posting copies of movie screeners or unreleased CDs, right? Unfortuantely, the definition of “pre-release work” is awfully broad:

“‘(1) PRE-RELEASE WORK.—The term “pre-release work” refers to a work protected under this title which has a commercial and economic value and which, at the time of the act of infringement that is the basis for the offense under subsection (a)(3), the defendant knew or should have known had not yet been made available by the copyright owner to individual members of the general public in copies or phonorecords for sale, license, or rental.

Mere transmission to the public does not move the work outside the definition of “pre-release”. So everything broadcast on television or the radio that isn’t also available on some physical medium, like a VHS tape or a DVD. So if you tape a TV news broadcast and give the tape to a friend, you’re a criminal. That broadcast has not been made available to the public in physical copies, and you knew it. And you distributed the tape; elsewhere in the statute, “distribution to the public” is required, but in subsection (a)(3), any distribution at all will do.

If this dangerous new law is passed, you could go to jail for three years for taping a show for a friend.

Silliness.

UPDATE: As reported by Ernie, the anticipated frankenbill never made it to a vote. So we’re safe. For now.

On the heels of the news that the INDUCE Act is dead for now, a story of some very cool inducement right here in Minneapolis.

The Walker Art Center, a museum of contemporary art that really gets it, is closed this year for a major expansion. While they’re closed they’ve put together numerous “Walker Without Walls” programs. Perhaps the coolest is called Radio Re-Volt. In workshops all over town, they teach people how to set up their own milliwatt pirate radio stations, each with a range of about one city block. Driving around Minneapolis with radios set to 97.7 FM, one will be able to catch snatches of tiny, personal broadcasts.

Just, nobody tell ASCAP.

Michelle Delio of Wired News has this write-up.

Creative Commons moves out of the basement of Stanford Law School, and all of a sudden they’re getting coverage in Newsweek. Nice.

Chris Anderson, Editor-in-Chief of Wired Magazine, has a very smart article in the current issue titled “The Long Tail.” In it, Anderson argues that new dissemination technologies make relatively unpopular works a much more important source of revenue. For example, even the largest chain bookstores can stock only a fraction of the titles available on Amazon.com; even the largest music stores can stock only a fraction of the tracks available on Rhapsody. If you make a graph with the number of sales on the vertical axis and arrange items by their sales rank along the horizontal axis, you get a graph with a large peak at one end, representing the popular, mainstream items that sell many copies each. But from there, the graph drops steeply, then trails off very slowly — a “long tail” of items that each sell very few copies, but that in aggregate could represent a substantial proportion of overall sales.

As I see it, this phenomenon has three important ramifications for free culture, two overwhelmingly positive, the other potentially negative.

On the plus side, the diminishing marginal cost of stocking works means that consumers have access to a much wider range of culture than they would otherwise. Much of it is unpopular because it isn’t very good, but much of it is unpopular because it caters to a very small audience. To take one example, there’s this band that plays traditional Jewish songs in the 50s surf guitar style. They’re called Meshugga Beach Party. There is, as I’m sure you can imagine, an extremely small audience for this band. But they’re really good, if you’re into ashkenaz irony. Even the biggest, coolest music store in town wouldn’t stock their album. But it’s available through two distributors that take advantage of the long tail; you can buy the CD from CD Baby, and you can stream the whole album via Rhapsody. The small, dispersed set of people who like the CD have their lives enriched just a little by having access to it.

More important, though, than consumers getting access to niche works, long-tail retailing allows niche artists to sell more product, making it easier to be a niche artist. A hundred streams a month on Rhapsody or a dozen album purchases a month on CD Baby or Amazon won’t pay the rent, but they might turn an expensive hobby into a modestly remunerative one, or help morale by demonstrating that there really is someone out there listening. And for artists with substantial followings who fly just underneath the radar of the mainstream music industry, long-tail retailing can make national tours both possible and profitable. This is especially true of artists who don’t fit into any well-established genre around which communities can grow — like GrooveLily, a personal favorite of mine.

The commercial viability of formerly unprofitable works is great for consumers and creators, as I’ve noted. But the same distribution channels and technological advances have the potential to hamper creativity in another way. Until 1976, works had to be renewed 28 years after creation in order to be protected for an additional 28-year term. If the owner failed to renew, the work enriched the public domain. The vast majority of owners — around 85% — failed to renew, since their works no longer had much commercial value. Even though the works had no commercial value, they had cultural value, and other creators were free to make new works from bits and pieces of the old works. As the plaintiffs in Kahle v. Ashcroft argue, the renewal requirement was such an integral part of the American copyright system that removing it should trigger heightened scrutiny, possibly leading to the reinstatement of some system of formalities.

But if the long tail of works that otherwise would have passed into the public domain is suddenly commercially viable, more copyrights would, in a new system of formalities, be renewed. A smaller percentage of works would enrich the public domain. This has the potential to weaken arguments for the reinstatement of the renewal formality and weaken the impact of renewal if it were reinstated. Of course, the long tail’s other benefits would do much to ameliorate these problems; if a million GrooveLilies bloom, it will be a great victory if even a small percentage reach the public domain more quickly.

The web offers cheapskates many delights, from FatWallet Forums to sites full of coupon codes to Froogle. Most of those center on legitimate ways to get the best price for goods purchased online (even if you have to send in for three rebates and apply two coupon codes to get it).

But if you aren’t above cheating a little, Movie Madness is for you. It’s a tool that lets you pick a day and a movie theater, and then designs a theater-hopping schedule for you, trying to pack in as many movies as possible for the cost of one admission based on show times and the length of each movie. I like this mostly because it’s a neat hack, not because I intend to spend a day at the Mall of America seeing Wimbledon followed by Mr. 3000 followed by The Forgotten followed by Ladder 49 followed by Shark Tale.

Incidentally, this neat hack runs on the web server software designed for another neat hack, OKCupid.

I just got an email from the University of Minnesota, where I’m a law student, letting me know that Rhapsody subscriptions are now available through the school for $2 per month (an 80% discount over the normal rate).

It will be interesting to find out whether this is a pure market share grab by RealNetworks or whether it’s subsidized by the University or by the record companies.