joegratz.net

Bram Cohen, developer of the BitTorrent file distribution software, has opened a search engine for finding content available via the BitTorrent protocol. Other search engines for BitTorrent content already exist; some have been shut down due to legal problems. The new site appears to be intended to take advantage of the safe harbor for information location tools in section 512(d) of the Copyright Act.

I have two observations. One is that, as a legal matter, Cohen may have a hard time relying entirely on a requirement that copyright holders notify him of infringing torrents using the 512(c)(3) take-down procedure. Everyone at BitTorrent (which, I realize, may be only a handful of people) will have to remain entirely blind to all of the infringing torrents available through their engine. For example, if a BitTorrent employee ever clicked on this link, they’d be “aware of facts or circumstances from which infringing activity is apparent”, and they’d have to take down all of those links. I suspect that they are simply very careful never to type certain queries into their own search engine, preferring to wait for 512(c)(3) notice from copyright holders. But that’s unlikely to work as a long-term strategy; wouldn’t it be amusing if someone started sliding printouts under the door of the BitTorrent office over in SoMa, or holding up placards for employees to see on their way in and out of work? If I were a lawyer for a large copyright holder whose works were readily available through the search engine, I’d be drafting a complaint right now.

The second reason I think this is a bad idea is not really a legal reason at all. Until this point, Cohen has been able to say — and has said, repeatedly — that BitTorrent is just a tool for distributing large files, and that he doesn’t really know or care about the uses to which it’s being put. Cohen as the maker of a general-purpose data distribution tool is much easier to defend, as a matter of politics and rhetoric, than Cohen as a direct facilitator of infringing downloads. This is entirely apart from any legal arguments about the Betamax doctrine; I believe that it is simply a dumb rhetorical move to have this search engine come from the same organization that developed the BitTorrent software and protocol.

In yet another insightful and interesting post, Bill Patry writes about the history of the legislation that was upheld in yesterday’s Luck’s Music Library decision. Most intriguingly, he notes:

I did not believe that the legislation provided a direct incentive for the creation of new works, nor that, divorced from the Berne obligation, restoration represented good policy. But honesty in treaty adherence is good policy, and retroactive protection for U.S. works overseas is of benefit to both U.S. copyright owners and to the system as a whole. I thought those two objectives sufficient, from a policy and constitutional standpoint. And I say this as someone who believes, more fervently than even Larry Lessig, that the “Promotion of the Progress of Science” language in Article I, section 8, clause 8 is a real substantive limitation on Congress’s power. But I also believe that one must look to the system as a whole to see whether there has been a benefit provided by legislation, and in the case of GATT restoration, I answer that question affirmatively.

He’s right to divorce this issue from term extension; they are conceptually distinct. And I think that his view of the “promote the progress” language as a question of net effects is quite right. The Luck’s Music, Golan, and Kahle cases, taken together, simply argue that Berne implementation had a negative net effect on the progress of science — a matter of policy on which reasonable minds, defining progress differently, disagree.

The D.C. Circuit today released its opinion in Luck’s Music Library v. Gonzales, holding that the Uruguay Round Agreements Act, which granted new copyrights to foreign works which were in the public domain due (primarily) to failure to adhere to notice, registration, and deposit requirements, is constitutional.

The plaintiffs’ argument was twofold. First, that the Constitution, by directing Congress to provide incentives for creation of works, forbade the enactment of copyright laws that fail to provide such incentive, and that the law at issue here, by granting rights only to works that have already been created, violates that limitation. Second, they argued that both historical practice and Supreme Court precedents indicated that once a work enters the public domain, that status cannot be changed.

The D.C. Circuit rejected both of these arguments. The court found just a sliver of incentive for the creation of new works, in observing that since copyright protections tend to grow and grow, an author, for example, writing in Peru in 1932 might have decided to create a work because he knew that some day the United States might retroactively grant it copyright protection. This, said the court, is enough; Eldred requires only the barest hint of an incentive, and while the incentive here is minimal, it nonetheless, argues the court, exists.

The court rejected the historical and precedential arguments almost without considering their merits. The plaintiffs argued that never before had retroactive copyright protection been extended; the government argued that it had. Without careful analysis of the difference between previous laws and the URAA, the court concluded at least that the historical record was indecisive. And plaintiffs’ most direct argument — that Graham v. John Deere clearly states that “Congress may not authorize the issuance of patents whose effects are to remove existent knowledge from the public domain, or to restrict free access to materials already available” — was rejected as an irrelevant dictum. The court simply stated, rather unsatisfyingly, that copyright and patent doctrines were not always the same, and that this was one case where they were different.

I wish the plaintiffs good luck before the Supreme Court. One wonders how different this opinion might have looked if the Golan case, which was filed around the same time as Eldred, had gone up on appeal first.

(Full disclosure: I worked on Golan, a challenge to the same law, while I was at the Stanford CIS; because of that work, I have corresponded with counsel to Luck’s Music and assisted once or twice with their research.)

Bill Patry writes on a subject I’ve been meaning to discuss in this space — whether Congress has the power to enact copyright-like laws that violate the constitutional requirements of “writings” and “limited times” under its Commerce Clause power.

The courts in the three reported bootlegging cases to date — Moghadam, Martignon, and KISS Catalog — each recognized that Congress may not bypass a limitation on an enumerated power by acting in the field of the limited power while claiming to act under the broader commerce power. Each relies on the Railway Labor Executives case, which presents a precisely analogous situation. One of Congress’s enumerated powers is the establishment of “uniform Laws on the subject of Bankruptcies throughout the United States”. Congress attempted to make a non-uniform bankruptcy law under its commerce power, establishing special rules for the Rock Island Rail Road that applied to nobody else. Congress was exceeding its constitutional authority, and the Supreme Court struck down the law.

Similarly, when Congress attempted to enact the anti-bootlegging statute — a copyright law that granted exclusive rights for an unlimited time — under its commerce power rather than its copyright power, the courts found that Congress had likewise exceeded its constitutional authority.

This line of cases raises a number of unanswered questions. The most important of these regards the bounds of the field of copyright law. How far outside the traditional contours of copyright law does Congress have to go before it’s no longer making a copyright-like law at all, but is instead making an allowable regulation of interstate commerce? Perhaps the field of copyright that preempts Congressional action outside the limitations of the copyright clause is coextensive with the field that preempts state action, recently examined in the L’Oreal right-of-publicity case. If so, the coming constitutional challenge to the anti-circumvention provisions of the DMCA (some day — be patient) will likely be successful. And the plaintiffs in the Luck’s Music Library and Golan cases, which challenge a grant of copyright protection to materials in the public domain, are more likely to succeed.

Perhaps in one of these cases, the Supreme Court will hand down the Copyright Clause version of the Railway Labor Executives decision, finding that Congress cannot circumvent the “writings” and “limited times” requirements by recourse to the Commerce Clause.

In a slightly strange turn, BMI is now issuing licenses for podcasts. As I’ve argued before, podcasts are reproductions, not public performances, and it’s not clear to me that a podcaster buying a BMI license is getting anything for her money.

It wouldn’t surprise me to find out that ASCAP and BMI think that podcasting is both a reproduction and a public performance. They would be wrong.

There are two events that go into a podcast. The first occurs when the user (or automated software set up by the user) downloads an MP3 over the internet. The second, which typically happens at a later time, is when the user plays back the MP3, typically on a portable music player like an iPod. The first event — the download — is simply not a performance, in copyright terms. Nothing is rendered; no sounds are made audible. The second event — playback — is a performance, but it’s a private performance, typically occurring at the direction of the listener through the listener’s headphones. That’s not a public performance; thus, it doesn’t implicate the public performance right; thus, a BMI license is unnecessary.

As BMI helpfully points out, a podcaster would need all sorts of licenses to make this happen. Specifically, they’d need a license to make a phonorecord embodying the musical work (which you can get from HFA with no trouble at all) and a license to make a copy of the sound recording (which you’d have to negotiate with each individual record company).

(Keep in mind, podcasters: I’m not a lawyer. Check with yours before listening to what I say.)

The Copyright Office has posted reply comments submitted in its Orphan Works proceeding. Of particular interest to me were the following:

• Professors Jane Ginsburg and Paul Goldstein on the interaction between orphan works proposals and the Berne and TRIPS treaties. They have some misgivings about the compatibility of even a non-formalities-based solution with treaty obligations, but find that a narrowly circumscribed solution based on the inability to identify the rights-holder is likely to fulfill those obligations.
• Creative Commons and Save The Music renew their proposal for a “categorical” solution to the orphan works problem, requiring copyright holders to signal whether or not they wish to exploit their rights. These comments are particularly notable for their skillful use of comments submitted in the initial round as evidence for their argument — great work Chris, Lauren, and Stanford CIS students!
• The National Music Publishers Association and the Harry Fox Agency, arguing that the orphan works problem with regard to making new recordings of old songs is nonexistent (they’re right) and arguing that setting up an agent to set and collect “market rate” fees for all other uses for which the musical work copyright owner cannot be found is the best solution for all players.
• The Internet Archive submits comments proposing a notice-and-takedown procedure for library and archival use of all works, including orphan works. (Comments written by students and fellows at the Samuelson Clinic at Boalt Hall — great work!)
• The Recording Industry Association of America, believe it or not, is in favor of a solution to the orphan works problem. They oppose “categorical” solutions and see some problems with the CCI proposal that was widely praised in the initial round, but they remain positive about the general project. Of course, their ox is exceedingly unlikely to be gored; RIAA members deal in sound recordings, which have only been protected by federal copyright law since 1972. Because of their youth (and because RIAA members are generally easy to find), RIAA members’ recordings are unlikely to take on orphaned status.
• ASCAP still doesn’t think there really is an orphan works problem, especially with regard to musical works. They’re partially right — nobody who’s informed is wandering in the wilderness wondering whom to pay because they want to sing “At Last” on karaoke night at the pub. Their point about the “tyranny of the anecdote” is a good one, though in this case I think that the hundreds and hundreds of anecdotes received in the initial round of comments adds up to a real statement about a real problem.
• The MPAA (represented by the high-powered [and occasionally rather awesome] Washington copyright boutique Smith & Metalitz) is in favor of a due diligence based solution, though one with the same limits proposed by other copyright holders — for example, that each prospective user must do his own due diligence and may not rely on the search of a previous user. They call the Creative Commons proposal and those with similar registration-based solutions “outliers” and ask the Copyright Office to put them to one side and concentrate on a due diligence based solution. And on the last page, they helpfully include the phone numbers of the clearance departments for all of the major studios.
• The Authors Guild has general support for a solution, and reports from a survey of their membership that only 15% have run into orphan work problems. While this may seem small, this means that roughtly 1,200 of the Authors Guild’s 8,200 members — the most sophisticated of published writers — have been hampered in their literary goals by the lack of a solution to this problem.
• The Illustrators’ Partnership reply comments, like their initial comments, stake out the most copyright-maximalist end of the spectrum of submissions. They spend substantially their entire reply comment attacking the Creative Commons proposal and the CCI proposal. None of this is terribly surprising when one glances at the bottom of their letterhead and notices that their lawyer is arch-maximalist Bruce Lehman, he of the NII White Paper.
• Finally, the Electronic Frontier Foundation, in comments put together by Jason Schultz, has some very smart general observations about some of the limitations commenters seek to impose on any solution to the orphan works problem. He notes that escrow accounts are awfully inefficient and make no sense in the broad category of cases in which no fee would have been charged; that HFA, ASCAP, and BMI’s assertions that some types of works are categorically immune from orphan works problems is simply false; that the allowable uses of truly orphaned works should be unlimited in scope; and that the Directors Guild’s concerns about the vindication of the director’s moral rights simply have nothing to do with copyright.

There is startlingly broad consensus for some sort of action toward a solution to the orphan works problem, and while the next step is likely to be a series of rather uncomfortable roundtable meetings, it seems that there is a light at the end of the tunnel. It’s said that copyright legislation only happens when all of the players at the table have something to gain and can hash out a proposal among themselves; that certainly seems likely to happen eventually for orphan works, given the positions of the major players as set forth in their comments.

I just happened upon this page, showcasing the photos that Flickr users have uploaded and licensed under various Creative Commons licenses. What’s striking to me is the variance of the choices. People aren’t just choosing the first one on the list, or choosing the most or least restrictive option by default. They seem to actually be thinking about what they do and don’t want people to be able to do with their photos. Here’s a graph of the data.

Note that although it is not the default setting on Flickr, well over a million photos have been released under one of the Creative Commons licenses. Note, too, that a plurality of users choose one of the most restrictive licensing options, forbidding commercial use and either limiting or forbidding the creation of derivative works.

I’ve been meaning to write a piece on who chooses open-content licenses, which licenses they choose, and what factors affect their choices. I’d like to see, for example, how much of the time people choose the GPL because it’s the most famous F/OSS license, how much of the time they choose it because they want to incorporate GPL’d code into their project, and how much of the time they choose it because they actually know and like what it says. These data from Flickr suggest a greater variance among licensing options than I expected; perhaps the same would be true of an examination of the universe of F/OSS projects.

Blogging is likely to be light over the next week, as I’ll be finishing my last set of law school exams, attending commencement, and moving to San Francisco.

Speaking of which, if anybody knows of a nice, cheap studio or one-bedroom in or near Hayes Valley (preferably west of Van Ness, east of Fillmore, north of Haight, and south of Ellis), let me know. I harbor no illusions about the mutual exclusivity of “nice” and “cheap”, but I’m at least hoping for a balance between the two. I’ve been trolling craigslist somewhat obsessively, but I’d welcome additional input.

This is yet another reminder that your adversary, whoever or whatever he or she is, probably isn’t evil — just fulfilling his or her legal duty to be a zealous advocate for his or her client.

Hilary Rosen, former head lobbyist for the RIAA, has been saying some shockingly sensical things lately, now that she’s stepped down from that post. Here’s the latest. Rosen excoriates Steve Jobs for not supporting Microsoft DRM on the iPod while acknowledging that consumers’ ability to strip the DRM from files they buy can serve an important interoperability purpose.

She says she’s hoping Steve Jobs gets “Betamaxed”. Irony abounds.

A huge win today for consumer technology innovation: the D.C. Circuit struck down the FCC’s Broadcast Flag mandate.

This means that the FCC’s order banning the manufacture of fully-functional HDTV hardware will not go into effect as scheduled on July 1 of this year. The FCC may appeal to the Supreme Court, but the Court rarely reviews D.C. Circuit administrative law decisions, and in any case the appeal would not be heard until well after the mandate was scheduled to go into effect.

Congratulations to the ALA, Consumers Union, EFF, and Public Knowledge, and the other groups who brought the challenge.

Derek Slater argues that hard-line copyright minimalists might be doing more harm than good.

Ed Felten writes about a new document issued by the Digital Media Project, a group attempting to design an interoperable DRM platform. As Ed notes, they seem to be designing copyright concepts from the ground up, with some rather strange results.

This seems to be indicative of a larger problem in discussions about copyright and related rights regimes. Both sides often seem to want to knock down the whole existing law of copyright and move forward into a new paradigm that takes account of all of the vast technological changes of the past twenty years. Some content companies argue that the internet changes everything, so we need to replace copyright protection with robust DRM and force its inclusion in all devices that handle media. Some copyright minimalists argue that the internet changes everything, so we need to move copyright toward a regulated industry model where anyone can do anything with a copyrighted work so long as they pay a government-determined fee.

But the internet doesn’t change everything — or, to be more precise, the internet changes everything to the same extent that the mimeograph machine or the photocopier or the phonograph or the compact cassette or the television or the transistor radio changed everything. The ability of a teenager to create a compelling work of art in her bedroom and show it to the whole world and still have some of her allowance left over is only as revolutionary as the ability of a big band playing in a dance hall New York City to be heard on a radio in a park in San Francisco. The ability of that same teenager to illegally download music she loves for free is only as revolutionary as her father’s ability, when he was a teenager, to illegally copy rented LPs to cassettes. They are equally revolutionary and equally disruptive, even if we can disagree about whether they are equally right or wrong or beneficial or harmful.

Technology has changed the way creative works are disseminated and experienced at a fairly steady rate for the past 500 years. The internet is an incremental improvement, not a reason to throw out copyright law and build it again from the ground up. The basic model of copyright protection — granting time-limited, circumscribed exclusive rights to creators and giving them a right of action for damages against those who violate them — has endured since the Statute of Anne, and it should endure forever. Illegal P2P file sharing is a problem caused by technology, but it cannot be solved by limiting the capabilities of technology. And overzealous copyright enforcement is a problem caused by fear and risk-aversion, but it cannot be solved by sweeping changes that will be fiercely opposed, circumvented, and poorly utilized by those who are fearful and risk averse.

Copyright is a mature, fully-theorized area of law; it works well now, and it’s worked well for hundreds of years. An incremental change like the internet can be effectively met only by incremental changes to copyright law. Those who seek to change its traditional contours — proponents, for example, of anti-circumvention paracopyright or of alternative compensation systems — bear a heavy burden, a burden that neither group has met.

Daniel Wallace is a solo software developer and critic of the open source movement; see, for example, his comment to this post of mine. He argues in that comment that the GPL is a contract, not merely a license. Wallace’s ability to prove that claim will be central to the pro se lawsuit he recently filed in federal court against the Free Software Foundation, alleging illegal price-fixing.

I’ve downloaded his four-page complaint from PACER; it’s available here.

It will be illuminating to see how Mr. Wallace will argue that a copyright licensor’s restriction that downstream licenses to its copyrights must be granted at no charge rises to the level of an antitrust violation.

The Supreme Court has declined to hear an appeal of the Ninth Circuit’s decision in the internetmovies.com case, which I discussed here.

The site, incidentally, now says in rather large type, “There are no Hollywood movies on this site.” A wise move.