joegratz.net

Aaron Perzanowski has started a blog called Substantial Similarity, and it’s a really fun read. My favorite part of his post on Perfect 10:

Since Campbell, parody has been at the top of the fair use hierarchy. Any putative fair use that failed to take the piss out of the original, even if fair, has been seen as somehow less fair. And as for a use as utterly un-ironic as caching or indexing? Well, that’s hardly funny at all. How can it be fair?

The Op-Ed pages of today’s New York Times contain an essay inveighing against the current copyright term. The writer, Mark Helprin, finds the current term unjust, taking away vital resources from our children and grandchildren because of a rule better suited to an earlier time. The essay quotes Jefferson and opposes what the writer considers unwarranted regulatory interference.

More or less what you’d expect from an essay arguing for a shorter copyright term, right?

The funny thing is, though, that this essay is arguing for a longer copyright term. A term that lasts for seventy years after the author’s death is too short, Helprin argues, because property should not be “stripped from [his] children and grandchildren.” Instead, like other property, it should exist in perpetuity. In Helprin’s opinion, the Framers would have allowed a perpetual copyright term if they had “imagined that services and intellectual property would become primary fields of endeavor and the chief engines of the economy.”

The reasons Helprin is wrong as a matter of policy are numerous. Just a few:

• One additional year of copyright protection — the 71st year after the author’s death, let’s say — has an infinitesimal economic value today, literally on the order of a few cents even for the rare work that will continue to earn millions in that 71st year after the author’s death. That added value is simply not enough to lead an author to create a work he otherwise wouldn’t. Because the purpose of copyright is to act as an incentive to spur the creation of new works, a perpetual copyright term would have no effect other than giving an irrational author the warm fuzzy feeling that his heirs will reap the fruits of the sweat of his brow. The author’s grandchildren would be better off with a dollar invested today and passed on to them than with a perpetual copyright term.
• While the author’s heirs would reap the same amount as a few pennies invested today, the public will suffer. While Helprin may not be familiar with any uses for public domain works other than “price-reduced non-copyrighted” books at Barnes & Noble, the benefits of a robust public domain flow not only to “various business who will continue to profit from” works in the public domain, but also to the public itself. Archive.org, Project Gutenberg, and LibriVox are just a few of the nonprofit projects helping the public benefit from the public domain. In a hundred years, if Helprin’s works are mostly forgotten and no publisher finds it profitable to keep them in print, would Helprin want the works to be out of print forever, or would he want them to be available for free on the Internet — maybe even as an audio book?
• Helprin’s argument requires as a premise that no benefit to the public is great enough to justify depriving his grandchildren of a few pennies of net present value. This is not irrational, but it is unreasonable and mean-spirited.
• Finally, Helprin’s argument fundamentally misunderstands the American copyright tradition. Copyright are granted not because an author is naturally entitled to the fruit of his labor, but because granting copyrights gives incentives for the creation of new works. Helprin argues that this tradition is outdated, having arisen from an agrarian society where “[w]riters and inventors were largely those who obtained their sustenance from their patrimony or their mills.” But the importance of copyrights to the economy only strengthens the argument that copyright should be an economic right, not a moral one. If copyright is economically important, we should judge its effectiveness in economic terms.

Let us reflect, too, upon the irony that an argument in favor of perpetual copyright is being made by an author whose most successful work bears a title and a conceit appropriated from a work in the public domain.

Larry Lessig has set up a wiki on which the community is writing a response.

Yesterday, the Ninth Circuit Court of Appeals decided Perfect 10 v. Google. The case encompasses numerous issues, from section 512 to fair use to the bounds of secondary copyright liability. Here’s a summary of the most interesting parts to me (not to be confused with the most interesting parts to the litigants or to the court):

1. Preliminary Injunctions and Fair Use. The court held, for the first time, that in order to get a preliminary injunction in a copyright case, the plaintiff must not only demonstrate a substantial likelihood of proving his prima facie case, but must also demonstrate a substantial likelihood of overcoming any defenses raised by the defendant. First, “the defendant is responsible for introducing evidence of fair use in responding to a montion for a preliminary relief,” then the plaintiff must “show it is likely to succeed in its challenge to the alleged infringer’s evidence.” Slip op. at 5767. The availability of preliminary injunctions in cases where the plaintiff’s prima facie case is easy to prove but the defendant has a strong fair use defense chilled speech, when it didn’t ban it outright. This holding is likely to mean that plaintiffs will rarely get preliminary injunctions in many of the cases copyfighters care about — cases where copying is obvious and the real question is fair use.
2. Preliminary Injunctions and Section 512. The court also held that a plaintiff seeking a preliminary injunction must demonstrate a substantial likelihood of overcoming a defense under 17 USC 512. Section 512 contains its own rules about injunctions, but the court held that because “the DMCA does not change copyright law,” general principles apply. Accordingly, if the defendant raises a section 512 defense, the court first decides whether the defendant is likely to qualify for protection under section 512, then applies the appropriate injunction rules depending on the answer.
3. The “Server Test.” In analyzing whether the copyright holder’s display right has been violated, we are told that the “displaying” is performed by the web server, not by the user’s computer. Displaying is “show[ing] a copy;” here, the copy being shown resides on the server, and is shown by transmitting it over the Internet to the user’s computer for display. In a slightly weird turn, the court says that a computer owner displays a work when she “uses the computer to fill the computer screen with the photographic image stored on the computer, or by communicating the stored image electronicallly to another person’s computer” (so the server is doing the displaying in the web context because it stores the image). What’s weird is that right in the middle of this analysis, the court cites MAI v. Peak, which held that copies held in RAM are sufficiently “fixed” to be copies for copyright law purposes. If RAM copies are “fixed” (which they aren’t, but we’ll argue about that another day), why isn’t the user the one displaying the work by “us[ing] the computer to fill the computer screen with the photographic image stores on the computer”? I have to imagine that the fact that web browsers store transmitted images in RAM before displaying them wasn’t before the court.
4. In-Lining Images Is Not a Display. Well, it’s not direct infringement, at least. IMG tags and links are just HTML instructions, the court says, and “[p]roviding these HTML instructions is not equivalent to showing a copy.” This is because the HTML doesn’t represent the photographic image; you can’t display the image solely by knowing the URL, because the server the URL is pointing to has to actually serve the image.
5. Framing Is Not a Display. Having another site’s content display inside a frame is not infringement, because the content itself is being served by the original website. It might confuse consumers, but the court rightly notes that confusion is dealt with by trademark law, not copyright law.
6. Linking is not Distribution. Because the data itself is coming from some other server, linking to infringing content is not a “distribution” of that content. Connectedly, the “deemed distribution” rule from Hotaling (the bad Hotaling, not the good Hotaling) doesn’t apply because Google doesn’t have a copy of the allegedly distributed full-size images, and you can’t distribute something you don’t even possess. (I expect that some will see this as an adoption of the Hotaling rule by the Ninth Circuit, but I think it’s dicta. Napster is bad enough.)
7. Art. I, sec. 8, cl.8 Gets Teeth. The court says that courts must be “flexible in applying a fair use analysis,” and that the four fair use factors must be “‘weighed together in light of the purposes of copyright’” (quoting Campbell). And then comes the best part. “The purpose of copyright law is ‘[t]o promote the Progress of Science and the useful Arts,” and to serve the welfare of the public” (internal quotations and citations omitted). So courts must weigh the factors together, balancing the benefit to the generation of knowledge that comes from greater protection against the public interest in making fair use of copyrighted works. Excellent. (I’m hoping for summary dispositions in really egregious fair use cases that just say, “The district court failed adequately to consider the Progress of Science and the Useful Arts. VACATED and REMANDED.”)
8. Superseding Use. Use that displaces sales of the copyrighted work has rarely been held to be fair use. But here, the court held that even if the defendant’s use supersedes some of the copyright holder’s revenue-generating uses, it can still be fair use if the value to the public from the defendant’s use outweighs the “significance” of the superseding use. This is perhaps the most important holding, in part because it helps to get rid of the circular argument that the copyright holder wanted to license uses like the defendant’s. Even if the copyright holder could have garnered some meager licensing revenue by prohibiting uses like the defendant’s, the “market harm” factor won’t weigh against the defendant if the benefit to the public outweighs the superseded benefit to the copyright holder.
9. Evidence of Harm. In order for the “market harm” factor of the fair use analysis to tip in the plaintiff’s favor, the plaintiff must come forward with evidence of actual market harm. Market harm cannot be presumed, and “hypothetical” harm isn’t good enough. This will help to erode further the circular “I could have licensed this use, therefore it is not fair use” argument; on the other hand, it makes the sham “licensing” of use that’s actually fair use all the more pernicious, because it makes the use less likely to be found to be fair use in the future.
10. The User’s Local Cache Is Fair Use. The court held that the user’s local web cache is a fair use copy because it is a transformative use — presumably because it is using the image data not as a photograph, but as a means toward faster web-browsing performance. The effect on the copyright holder’s rights is minimal — who’s going to license the image for use in their own local cache? — and the public benefit is significant, so it’s fair use. The court also expressed some doubt that “such automatic copying could constitute direct infringement,” presumably under a CoStar v. Loopnet volitional copying theory.
11. Substantial Noninfringing Use. The court held that substantial noninfringing uses for a product defeats a claim for secondary copyright liability based on an infringement-facilitating product design, but does not immunize a defendant from other sources of secondary liability. Thus, for example, if Sony had sent letters to each VCR owner exhorting them to duplicate and sell movies they rented from the video store, Sony would be subject to contributory copyright liability for sending the letters, even though it is not subject to liability for the infringement-facilitating design of its product.
12. Imputed Intent. The court held that intent is an element of a claim for contributory copyright infringement. Then, the court effectively read that element back out of the doctrine by holding that intent may be imputed if the defendant knew that infringement was “substantially certain” to result from her action. (Just like in Torts class.)
13. Contributory Liability. The court announced a new formulation of the contributory liability doctrine for internet sites: in order to be liable, the defendant must have actual knowledge that specific infringing material is available, and must then fail to take simple measures which would have prevented further damage. The court also called those “simple measures” “reasonable and feasible means.” Recall, however, that compliance with section 512 makes this analysis go away; it removes secondary liability for service providers if its procedures are followed. The district court hadn’t reached the section 512 analysis, so the court remanded this issue.
14. Vicarious Liability. As with all forms of secondary liability, a claim for vicarious liability must be based on some underlying act of direct infringement performed by someone. Here, that direct infringement was the unauthorized display of Perfect 10’s images on websites unaffiliated with Google. For a plaintiff to prevail on a claim of vicarious liability, he must show that the defendant has the right and ability to control the direct infringement, but fails to do so. Here, the court found that Google had no right nor ability to control the display of Perfect 10’s images on the unauthorized websites, since Google didn’t run those websites and couldn’t shut them down even if it wanted to. Perfect 10 argued that Google could have shut down the sites’ AdSense accounts, depriving them of revenue, but the court pointed out that doing so would not remove the images from the third-party websites.

It’s a big opinion, with a lot of holdings, but most of them make sense and apply the law correctly. The case now goes back down to the district court for further factfinding and analysis. There’s a dispute over the adequacy of Perfect 10’s section 512 takedown notices, and after Perfect 10 v. CCBill, Google will have a much easier time prevailing on that issue.

The Ninth Circuit Court of Appeals issued its opinion today in Fair Housing Council of San Fernando Valley v. Roommates.com, LLC.

Roommates.com operates just like a dating service, except that it’s for finding roommates: people enter facts about themselves and preferences about the people they want to be matched with, then the system steers people to the profiles of people who match.

However, the housing world is unlike the dating world in one crucial respect: it’s legal to post a personal ad saying you don’t want to date people who have kids, or people who aren’t Hispanic, or people who are male. But the Fair Housing Act says it’s not legal to post an ad for housing saying you won’t rent to people who have kids, or people who aren’t Hispanic, or people who are male.

(Whether this provision of the Fair Housing Act applies to people seeking roommates, rather than offering rental units in which they don’t live, wasn’t an issue at this stage in the case, and I don’t have the expertise to opine on it.)

The Fair Housing Council sued Roommates.com under the Fair Housing Act, alleging that by asking users for their roommate preferences in categories like gender, familial status, and national origin, Roommates.com violates the Fair Housing Act. Roommates.com claimed that it was immune from this suit under section 230 of the Communications Decency Act. Section 230 says that online publishers of user-generated content aren’t liable for the content their users post (subject to certain exceptions not applicable here), so long as they aren’t “responsible, in whole or in part, for the creation or development of” the content.

This raises an extremely interesting question: If I have a website which encourages users to post unlawful content, am I “responsible, in whole or in part, for the creation or development of” that content, or am I immune under section 230? In his opinion, Judge Kozinski uses a colorful example:

Imagine, for example, www.harrassthem.com with the slogan “Don’t Get Mad, Get Even.” A visitor to this website would be encouraged to provide private, sensitive and/or defamatory information about others—all to be posted online for a fee. To post the information, the individual would be invited to answer questions about the target’s name, addresses, phone numbers, social security number, credit cards, bank accounts, mother’s maiden name, sexual orientation, drinking habits and the like. In addition, the website would encourage the poster to provide dirt on the victim, with instructions that the information need not be confirmed, but could be based on rumor, conjecture or fabrication.

(Yes, I registered the domain, and www.harrassthem.com now works, linking back to the Ninth Circuit opinion. Feel free to suggest amusing, non-liability-producing uses for it in the comments.)

Kozinski expresses doubt over whether such a site would be immune under Section 230. But after posing this rich, tantalizing question, and before deciding it, the court turns away from the question of whether Roommates.com’s questionnaire itself violates the FHA.

Instead, the court held that Roommates.com’s use of the data to control access to users’ profiles and to generate emails listing only “matches,” rather than the site’s publication of an unlawful housing preference, that strips it of Section 230 immunity. By making its site obey a user’s discriminatory housing preferences, the court held, Roommates.com was no longer a “passive pass-through” or mere “facilitator of expression by individuals.” Instead, it was creating new content of its own — an “additional layer of information” it is responsible, at least in part, for developing.

I find this holding somewhat troubling. It seems that the court was concerned with Roommates.com’s implementation of its users’ discriminatory preferences. By the court’s reasoning, it seems, merely repeating something a user posted, even in edited form, is permissible, but taking actions or generating new information (such as “match” emails) based on the user’s post can give rise to liability. This raises difficult line-drawing questions, since it makes some automatic processes (like a script that deletes all the dirty words in a post) cause a site to retain its immunity, while making others cause a site to lose its immunity.

Thankfully, I think discrimination laws may be among a very small set of laws to which such a rule could apply. Taking action or generating new content based on user-generated content doesn’t give rise to liability in many areas outside intellectual property, and intellectual property laws are already categorically excluded from CDA immunity.

Still, legal rules that make the automatic parsing of data (User: “I want a female roommate”) into metadata (Server: “Hmm, perhaps I should match this user only with females”) a legally actionable event make me uncomfortable.

In copyright, there are two ways to prove a prima facie case of copying. First, the plaintiff could show that (1) the defendant had access to the plaintiff’s copyrighted work, and (2) the defendant’s work is substantially similar to the plaintiff’s copyrighted work. Or, second, the plaintiff could show that her work and the defendant’s work are so similar that copying is the only realistic basis for the similarities — in other words, the plaintiff could show “striking similarity.”

On Wednesday, the United States District Court for the Southern District of New York granted summary judgment in favor of defendant Brian Transeau (yes, that Brian Transeau, a/k/a BT, who did those great Tori Amos remixes back in 1996 — you know, the “Talula” remix that was the second B-side on the Professional Widow 12″? But I digress . . .) and against plaintiff Ralph Vargas.

Vargas alleged that BT copied a drum riff from a breakbeat LP Vargas released in 1994.  The allegedly copied riff appears in BT’s track “Aparthenonia,” off Breakz from the Nu Skool.  BT claimed that the riff wasn’t copied; he’d put it together himself in Reason while working on his tour bus. (You can hear the two recordings side by side at the Columbia Law School Music Plagiarism Project.)

Vargas didn’t sell many copies of his breakbeat LP, and he was unable to show that BT had access to it.  So, to prevail, Vargas would have to show that his riff and BT’s were “strikingly similar.” Both sides presented expert testimony on similarity, and BT re-created the drum riff from scratch in Reason for the court. Vargas’s experts waffled during their depositions on whether they could rule out independent creation, stating that the works are “significantly similar, if not identical,” but failing to explain why or to rebut BT’s expert, who did a FFT spectrogram of the two recordings that showed differences.

The court found that Vargas’s expert’s conclusory statements about striking similarity were insufficient to create a disputed issue of material fact that would prevent summary judgment and necessitate a trial. Because the burden was on the plaintiff to prove each element of his case, and the plaintiff had failed to present any expert who found striking similarity (rather than merely substantial or significant similarity), the defendant prevailed on summary judgment.

This seems a little odd, but the oddness, I think, comes from the doctrine itself. “Striking similarity” is not a standard that admits much analysis or evidence; almost by definition, the similarity strikes one or it doesn’t.  All the evidence is in the two recordings. If the court is to “apply logic and experience to determine” whether copying is the only explanation, as Mel Nimmer said it should, why should the plaintiff be required to produce an expert to avoid summary judgment at all? The court reached the right result here — the evidence of independent creation was substantial — but it’s hard to say why there wasn’t a genuine issue of material fact just because their expert couldn’t rule out independent creation (and it’s even harder to say why Vargas didn’t find himself some expert who would rule out independent creation).

I’m not crazy about “striking similarity” as a doctrine for precisely this reason: it collapses the whole analysis into the question of whether the factfinder thinks the two works could or couldn’t have been created independently. The doctrine withstands analysis so poorly that it’s hard to say why and whether any evidence or advocacy is needed at all: at its core, the “striking similarity” doctrine says that, when the plaintiff can’t prove access, the factfinder is to listen to the two works and decide who wins, with close cases going in favor of the defendant.

(For more beating up on substantial similarity (just smarter than mine), see this post at Bill Patry’s blog.)

Congratulations to the Stanford Fair Use Project, which was co-counsel for BT during this case’s long history.