Digital Goods and the ITC: The Most Important Case That Nobody is Talking About

The following post first appeared on the Center for the Protection of Intellectual Property (CPIP) blog, and it is reposted with permission here.

By Devlin Hartline & Matthew Barblan

In its ClearCorrect opinion from early 2014, the International Trade Commission (ITC) issued cease and desist orders preventing the importation of infringing digital goods into the United States. The ITC’s 5-1 opinion has since been appealed to the Federal Circuit, with oral argument scheduled for the morning of August 11th, and the case has drawn a number of amicus briefs on both sides. Despite receiving little attention in media or policy circles, the positive consequences of the ITC’s decision are significant.

This case is important because the problem of the importation of infringing digital goods continues to grow. The ITC’s authority over digital goods can be a powerful tool for creators and innovators against a threat that has only gotten worse, and it would permit the ITC to go about doing what it’s always done in the intellectual property space—protecting our borders from the threat of foreign infringing goods. Interestingly, a look at the proceedings in the ITC and the briefs now before the Federal Circuit reveals how some parties now opposing the ITC’s authority over digital goods had argued for the opposite just a few years back.

The ITC Proceedings

This case began in March of 2012, when Align Technology Inc. filed a complaint with the ITC alleging that its only competitor, ClearCorrect Operating LLC, violated Section 337 of the Tariff Act of 1930 by importing digital goods that infringed several of its orthodontic patents. Section 337, codified at 19 U.S.C. § 1337, makes unlawful the “importation . . . of articles” that infringe “valid and enforceable” patents, copyrights, or trademarks, and it declares that the ITC “shall investigate any alleged violation of this section on complaint under oath or upon its initiative.”

There are two statutory remedies available to a complainant in an ITC proceeding. The first is an exclusion order, which dictates that “the articles concerned . . . be excluded from entry into the United States.” Exclusion orders are issued by the ITC and enforced by the U.S. Customs and Border Protection. The second remedy is a cease and desist order, which directs any person violating Section 337 “to cease and desist from engaging in the unfair methods or acts involved.” The ITC enforces its own cease and desist orders through the imposition of civil penalties, recoverable in the federal district courts.

Align’s complaint with the ITC involved its patented Invisalign System, a “proprietary method for treating crooked and misaligned teeth” using modern plastic aligners instead of old-fashioned metal braces. Align alleged that ClearCorrect violated Section 337 by importing “digital models, digital data and treatment plans that . . . infringe or induce infringement of” its patents, and it asked the ITC to “issue permanent cease and desist orders” prohibiting ClearCorrect from importing the digital files. In response, ClearCorrect argued that “no articles” had been imported since the digital data associated with the teeth aligners were not themselves “articles.”

This was the primary bone of contention: The ITC only has statutory authority over the “importation . . . of articles,” and if digital goods are not “articles,” then the ITC has no jurisdiction. After an administrative law judge (ALJ) determined that the digital files at issue were indeed “articles” within the meaning of Section 337, ClearCorrect petitioned the ITC to review that determination. The ITC took the case and solicited comments from the public as to whether electronic transmissions are “articles” under Section 337.

The ITC ultimately sided 5-to-1 with Align. On the threshold issue of whether electronic transmissions constitute “articles” under Section 337, the ITC affirmed the ALJ’s conclusion that they do: “[T]he statutory construction of ‘articles’ that hews most closely to the language of the statute and implements the avowed Congressional purpose for Section 337 encompasses within its scope the electronic transmission of the digital data sets at issue in this investigation.” This was consistent, said the ITC, with the “legislative purpose . . . to prevent every type of unfair act in connection with imported articles . . . and to strengthen protection of intellectual property rights.”

Appeal to the Federal Circuit

Having lost at the ITC, ClearCorrect appealed to the Federal Circuit. There, it focused its arguments on the statutory question of whether digital goods constitute “articles” under Section 337.

Public Knowledge and the Electronic Frontier Foundation (EFF) filed an amicus brief calling the ITC’s decision “sweeping and unprecedented,” and they urged the Federal Circuit to reject the ITC’s “overzealous construction” of the term “articles.” Aside from the statutory issue, the digital rights groups suggested that there were “important reasons” why Section 337 “ought not cover telecommunications.” They stressed the “real and unanswered questions about the enforcement role” ISPs would play, and they noted how ISPs “could be required to actively block transmission of certain content.”

It’s worth noting that no ISPs were involved in the ClearCorrect litigation—only ClearCorrect itself was subject to a cease and desist order. But this ISP question seems to be the reason why the case drew their attention: The real concern wasn’t whether ClearCorrect had infringed Align’s patents; it was whether the ITC had the authority to issue cease and desist orders to ISPs. This sentiment was echoed in an amicus brief by the Internet Association, which includes Google, arguing that the internet “should not be restricted to national borders” because of “the unforeseeable but far-reaching results that would follow.”

The policy arguments made by Public Knowledge, the EFF, Google, and others were essentially circular: The internet should be “open” so we shouldn’t let the ITC “close” it. But that begs the question of what the ideal “open” internet looks like, and what illegal activities should or should not be tolerated in the digital space. We shut our borders to infringing physical goods. What makes infringing digital goods so special? A right is only as good as the remedies available to enforce it, so why should we give short shrift to the property rights of artists, creators, and innovators?

Align’s intervenor brief took the groups to task: “The amici briefs supporting ClearCorrect brim with hyperbole.” Align noted that the ITC “only asserts jurisdiction over the ‘articles” that are electronically transmitted, not over all acts of transmission.” It pointed out that it is the “owner, importer, or consignee” of the “articles” that violates Section 337, not the carrier, and it said that the claim that the ITC could issue cease and desist orders against ISPs for “data transmission activities” is “baseless.”

Supporting the ITC’s understanding of “articles,” an amicus brief filed by the Association of American Publishers explained that the ITC’s “authority over electronically transmitted copyrighted works is critical because . . . there has been rapid growth in digital publications.” It pointed to the rise in digital piracy “at the expense of U.S. creators and innovators.” It urged that affirming the ITC’s decision was “crucial” since it “will help ensure that unfair trade practices abroad do not harm the livelihoods” of those that “rely on copyright protection.”

An amicus brief filed by Nokia supporting the ITC also noted the importance of protecting intellectual property: “Stripping the Commission of its long-exercised authority over electronic transmissions could gravely damage the protection of valid patent rights through Section 337 investigations.” It pointed out that holding otherwise would lead to “absurd results” since the ITC would have jurisdiction over software “imported on a USB stick or CD-ROM” but not software disseminated by “electronic transmission.” Such a result would be “wholly contrary to the remedial purpose of Section 337.” Nokia concluded that the ITC’s “authority should not wax and wane as technology develops new methods of dissemination.”

The MPAA and the RIAA likewise submitted an amicus brief supporting the ITC. The industry groups pointed out that “illegal downloads and illegal streaming” account for most of the infringement losses they suffer, and they argued that “copyright protection is essential to the health” of their industries. They urged the Federal Circuit to affirm the ITC because “Section 337 is a powerful mechanism for stopping illegal electronic imports,” and doing so “would give effect to the intent of Congress that Section 337 protect U.S. industries from all manner of unfair acts in international trade.”

Who has the better argument here? Obviously, both sides argued that the text of Section 337 favored their positions. ClearCorrect and its supporters claimed that “articles” should be interpreted narrowly to include only tangible goods, while the ITC and its supporters wanted a read of the statute that allows the ITC to continue to fulfill its mission even as new technology and methods of trade become more common. What may come as a surprise, however, is that many of the groups now seeking to limit the ITC’s jurisdiction were arguing just the opposite a few years ago.

Remember the OPEN Act?

It may seem like ages ago, but it’s been less than four years since Congress debated the Stop Online Piracy Act (SOPA) and the PROTECT IP Act. Those two bills would have explicitly afforded artists and creators robust tools to use in the federal district courts against foreign rogue sites that aim their infringements at the United States. Many vocal opponents of the bills supported an alternative approach: the OPEN Act. Under the OPEN Act, the ITC would have been given explicit authority to investigate complaints against foreign rogue sites that import infringing digital goods into the United States.

The OPEN Act’s sponsors set up a website at where members of the public could see the text of the bill and suggest changes to it. The website included an FAQ to familiarize supporters with the thinking behind the OPEN Act. As to why online infringement was an issue of international trade, the FAQ pointed out that “there is little difference between downloading a movie from a foreign website and importing a product from a foreign company.”

When advocating for the OPEN Act as a good alternative to SOPA and the PROTECT IP Act, the bill’s sponsors touted the ITC as being a great venue for tackling the problems of foreign rogue sites. Among the claimed virtues were its vast experience, transparency, due process protection, consistency, and independence:

For well over 80 years, the independent International Trade Commission (ITC) has been the venue by which U.S. rightsholders have obtained relief from unfair imports, such as those that violate intellectual property rights. Under Section 337 of the Tariff Act of 1930 – which governs how the ITC investigates rightsholders’ request for relief – the agency already employs a transparent process that gives parties to the investigation, and third party interests, a chance to be heard. The ITC’s process and work is highly regarded as independent and free from political influence and the department already has a well recognized expertise in intellectual property and trade law that could be expanded to the import of digital goods.

The Commission already employs important safeguards to ensure that rightsholders do not abuse their right to request a Commission investigation and the Commission may self-initiate investigations. Keeping them in charge of determining whether unfair imports – like those that violate intellectual property rights – [sic] would ensure consistent enforcement of Intellectual Property rights and trade law.

Some of the groups now arguing that the ITC shouldn’t have jurisdiction over digital goods openly supported the OPEN Act. Back in late 2011, the EFF stated that it was “glad to learn that a bipartisan group of congressional representatives has come together to formulate a real alternative, called the OPEN Act.” The EFF liked the bill because the “ITC’s process . . . is transparent, quick, and effective” and “both parties would have the opportunity to participate and the record would be public.” It emphasized how the “process would include many important due process protections, such as effective notice to the site of the complaint and ensuing investigation.”

Google likewise thought that giving the ITC jurisdiction over digital goods was a great idea. In a letter posted to its blog in early 2012, Google claimed that “there are better ways to address piracy than to ask U.S. companies to censor the Internet,” and it explicitly stated that it “supports alternative approaches like the OPEN Act.” Google also signed onto a letter promoting the virtues of the ITC: “This approach targets foreign rogue sites without inflicting collateral damage on legitimate, law-abiding U.S. Internet companies by bringing well-established International trade remedies to bear on this problem.”


The ITC has been protecting our borders against the importation of infringing goods for nearly a century now. As technology and trade evolves, it makes perfect sense to let the ITC continue to do its job by protecting our borders against the importation of infringing digital goods. This is an important tool for our innovators and creators in combating the ever-growing flood of foreign infringing goods.

The fact that many of those who supported the OPEN Act are now supporting ClearCorrect suggests that for them this appeal isn’t really about whether digital goods are “articles” under Section 337. The ITC is an appropriate venue for all of the reasons the supporters of the OPEN Act publicized just over three years ago: The process is transparent, there’s ample due process protections, the commissioners are experienced and independent, and their decisions are consistent.

As the 5-1 opinion suggests, affirming the ITC’s decision should be an easy choice for the Federal Circuit. Let’s hope the Federal Circuit does the right thing for our artists and innovators.

How Rhetorical Epithets Have Led the FTC Astray in its Study of Patent Licensing Firms

The following post first appeared on the Center for the Protection of Intellectual Property (CPIP) blog, and it is reposted with permission here.

We’ve all heard the narrative about patent licensing firms, often referred to pejoratively as “patent trolls.” These patent owners, who choose to license their innovations rather than build them, are the supposed poster-children of a “broken” patent system. It’s as if commercializing one’s property, just like a landlord leases his land for another to use, is suddenly a bad thing. Nevertheless, the power of this “troll” rhetoric cannot be denied. Many provisions in 2011’s Leahy-Smith America Invents Act were aimed at starving out these “trolls,” and no less than five bills currently under consideration in the House and Senate seek to further deflate their sails.

Another example of the powerful appeal of the “patent troll” rhetoric is that the agencies charged with enforcing antitrust law have also been convinced that there is something amiss with the commercial licensing of patented innovation in the marketplace. This has been a key feature of the deployment of patented inventions in America’s innovation economy since the early nineteenth century, as scholars have shown. Last year, the Federal Trade Commission (FTC) instigated its own investigative study of what it calls “patent assertion entities” (PAEs), which is merely a more formal and neutral-sounding synonym for the popularized “patent troll” epithet.

In a new paper published in the George Mason Law Review, Sticks and Stones: How the FTC’s Name-Calling Misses the Complexity of Licensing-Based Business Models, CPIP Senior Scholar Kristen Osenga takes a closer look at the FTC’s ongoing study of PAEs and finds that it is destined to fail for two simple, yet inescapably obvious, reasons.

The first is the basic definitional problem of the FTC’s characterization of PAEs, which puts all patent licensing firms in the same boat. Failing to take a more nuanced approach, Osenga warns, “fires up the rhetoric but obscures thoughtful discussion and debate about the issue.” Building upon her previous work, she explains:

[T]he real problem is that patent licensing firms are treated as a homogenous category, with no attention paid to the wide range of business models that exist under the patent licensing firm umbrella. The categorical determination of patent licensing firms as “problems” imputes to a large, diverse group of firms the negative actions and qualities of a small number of bad actors.

Since not all “trolls” are alike, Osenga cautions, it’s “naïve and inaccurate” to lump them all together. And when the FTC makes this mistake, it leads to a situation “where words actually can hurt, much more so than sticks and stones.” The FTC’s study is explicitly “premised on a one-size-fits-all conception of patent licensing firms.” Rather than shedding much-needed light on the complex innovation ecosystem, the study promises to squander the opportunity by failing to recognize that not all “trolls” are the same.

Osenga notes that the FTC is uniquely situated to obtain nonpublic information about how these patent licensing firms operate using its investigative power under Section 6(b) of the FTC Act. Unfortunately, however, the study is premised on the faulty notion that the only upside of patenting licensing firms is to “compensate inventors.” But this focus on patents-as-incentives misses the forest for the trees, Osenga urges, as it fails to account for the larger patent-commercialization network:

[T]here are many steps between invention and the introduction of an actual product to the market and consumers. These steps include transforming an idea in to a marketable embodiment, developing facilities to produce the marketable embodiment, creating distribution channels to bring the embodiment to the consumer, and making the consumer aware of the new product. Each of these steps requires its own additional resources in the form of both capital and labor.

The FTC study, like many patent skeptics, fails to consider the benefits of the division of labor that patent licensing firms represent. Not every inventor is willing or able to bring an invention to the marketplace. Osenga’s point is that patent licensing does more than simply compensate inventors for their troubles; it creates liquid markets and solves problems of asymmetrical actors and information. These exchanges increase innovation and competition by playing the role of match-maker and market-maker, and they place valuable patents into the hands of those who are better positioned to exploit their worth.

Osenga points out that there are indeed possible negative effects with patent licensing firms. For example, they sometimes engage in ex post licensing, waiting to offer licenses until after the would-be licensee has already adopted the technology. These firms can be better positioned litigation-wise since their potential exposure is typically less than that of the infringers they sue. Finally, patent aggregators tend to have greater market power, and it can be difficult to judge the quality of any given patent that’s asserted when they offer to license their entire portfolio.

As with all things, Osenga stresses, there’s both good and bad. The problem is figuring out which is greater. The FTC could conduct a study that reveals a “detailed understanding of the complex world of patent licensing firms,” she laments, but that’s not what the FTC is doing:

[T]he configuration of the study is slanted in such a way that only part of the story will be uncovered. Worse still, the study has been shaped in a way that will simply add fuel to the anti-“patent troll” fire without providing any data that would explain the best way to fix the real problems in the patent field today.

This leads to the second problem with the FTC study, which follows as a necessary, logical consequence from the first definitional problem: There are serious methodological problems with the study that will undermine any possible empirical conclusions that the FTC may wish to draw.

Osenga says that the FTC’s study is simply not asking the right questions. Painting a complete picture of complex licensing schemes requires more than just counting the number of patents a firm has and adding up the attempts to negotiate license deals. To really get to the bottom of things, she contends, the FTC should be asking why patentees sell their patents to licensing firms and why licensing firms buy them from patentees. Better still, ask them why they decided to become patent licensing firms in the first place.

This insight is powerful stuff. It’s not enough to simply ask these firms what they’re doing; to really understand them, the FTC must ask them why they’re doing it. And the results are likely to be varied:

Some, of course, begin with this business model in mind. Others invent new technology but are unable to successfully commercialize it themselves, despite making efforts to do so. Still others exist as practicing entities for years or decades before something changes—supply change issues, rampant infringement by competitors, and regulatory initiatives—and they are no longer able to exist as a viable practicing entity.

Similarly, the FTC could ask them what kind of firms they are, and these answers are also likely to be diverse. Osenga’s point is that the FTC’s questions aren’t designed to showcase the vast differences between the various types of patent licensing firms. If the FTC wants to get to the bottom of how these firms affect innovation and competition, the first step should be to realize that they’re not all the same. The FTC’s study is as clumsy as those who refer to all such firms as “patent trolls,” and the lack of nuance going in will unfortunately produce a study that lacks nuance coming out.

In the end, Osenga agrees that deterring abusive behavior is a good thing, and she worries about innovation and competition. However, unlike many in patent policy debates, she is also concerned that the rhetoric is having an undue influence on policymakers. Throwing all patent licensing firms into the “patent troll” bus will not get us the narrowly-tailored reforms that we need. Sadly, the FTC’s approach with its ongoing study appears to have swallowed this rhetoric wholesale, and it seems unlikely that the results will be anything but more fuel for the “patent troll” pyre.

Unintended Consequences of “Patent Reform”: The Customer Suit Exception

The following post first appeared on the Center for the Protection of Intellectual Property (CPIP) blog, and it is reposted with permission here.

In the last two weeks, the House and Senate Judiciary Committees marked up wide-ranging patent legislation ostensibly aimed at combating frivolous litigation by so-called “patent trolls.” But while the stated purpose of the House and Senate bills—H.R. 9 (the “Innovation Act”) and S. 1137 (the “PATENT Act”), respectively—is to combat abusive litigation, a closer look at the actual language of the bills reveals broad provisions that go far beyond deterring frivolous lawsuits. This far-reaching language has raised concerns in the innovation industries that, instead of curbing ambulance-chasing patentees, Congress is preparing to fundamentally weaken the property rights of all inventors, emboldening patent infringers in the process.

The “customer suit exception” or “customer stay” provisions that appear in both bills are particularly troubling. These provisions direct courts to stay patent infringement suits against “retailers” and “end users” in favor of suits involving manufacturers higher up the supply chain. While the basic idea makes sense—we’ve all heard stories of coffee shops being sued for patent infringement because of the Wi-Fi routers they used—the provisions are drafted so broadly and inflexibly that they invite abuse and gamesmanship by infringers at the expense of legitimate patent owners.

Both the Innovation Act and the PATENT Act provide that “the court shall grant a motion to stay at least the portion of the action against a covered customer” that relates “to infringement of a patent involving a covered product or covered process” if certain conditions are met. The first condition in both bills is that the “covered manufacturer” must be a party to the same action or to a separate action “involving the same patent or patents” related to “the same covered product or covered process.” In other words, so long as the manufacturer is challenging the patentholder, the customer is off the hook.

The two main problems here are that (1) the definition of “covered customer” in both bills is exceedingly broad, such that almost any party can claim to be a “customer,” and (2) the provisions leave the courts no discretion in deciding whether to grant a stay, forcing them to halt proceedings even when it’s not warranted.

Both bills define “covered customer” as “a retailer or end user that is accused of infringing a patent or patents in dispute.” “Retailer,” in turn, is defined as “an entity that generates” its “revenues predominantly through the sale to the public of consumer goods and services,” and it explicitly excludes “an entity that manufactures” a “covered product or covered process” or “a relevant part thereof.” Thus, a “retailer” is a “customer,” but a “manufacturer” is not.

This language is far broader than necessary to achieve the stated purpose of protecting downstream retailers and end users. The Senate’s section-by-section breakdown of the PATENT Act claims that the “customer stay is available only to those at the end of the supply chain.” But the actual definitions in both bills are so broad that almost any entity in the supply chain would be eligible for a mandatory stay. This is so because almost all manufacturers are also retailers of other manufacturers; that is, almost all manufacturers could claim to be a “customer.”

Take, for example, a smartphone company that sources its components from a third-party manufacturer. If the smartphone company were sued for patent infringement over a component, it could claim to be a “covered customer” under both bills. Many smartphone companies generate “revenues predominantly through the sale to the public of consumer goods and services,” and they would not be considered “an entity that manufactures” the component. As a “retailer,” the smartphone company would be entitled to a mandatory stay, even though it’s nothing like the mom-and-pop coffee shop the customer stay provisions are designed to help. A district court would be forced to grant the stay, even if doing so hampered a legitimate patentholder’s ability to enforce its property right.

Against this backdrop, it’s important to keep in mind that the decision to stay proceedings has historically been left to the discretion of judges. Sometimes there are indeed good reasons to grant a stay, but each case is unique, and courts frequently weigh many factors in deciding whether a stay is appropriate. Instead of recognizing this dynamic, the Innovation Act and the PATENT Act mandate a one-size-fits-all solution to an issue that is best determined on a case-by-case basis. In effect, the bills tie the hands of district court judges, forcing them to stay suits even when the equities dictate otherwise.

While in some cases a manufacturer may be the more appropriate party to litigate a patent suit, it is not always true that efficiency or justice dictates staying a suit against a customer in favor of litigation involving the manufacturer. Courts generally balance several factors, such as convenience, availability of witnesses, jurisdiction over other parties, and the possibility of consolidation, when deciding whether to grant a stay. Courts consider whether the stay will lead to undue prejudice or tactical disadvantage, and they examine whether it will simplify the issues and streamline the trial. The decision to stay involves an extensive cost-benefit analysis for both the court itself and the litigants.

The Supreme Court has often emphasized the importance of judicial discretion in deciding whether a stay is warranted. As Justice Cardozo wrote for the Court in 1936, the decision to stay “calls for the exercise of judgment, which must weigh competing interests and maintain an even balance.” Justice Cardozo warned that the judiciary “must be on our guard against depriving the processes of justice of their suppleness of adaptation to varying conditions.” In the patent law context, Justice Frankfurter, writing for the Court in 1952, declared: “Necessarily, an ample degree of discretion, appropriate for disciplined and experienced judges, must be left to the lower courts.”

The problem with the House and the Senate bills is that they take away this important “exercise of judgment” and threaten to remove much-needed flexibility and adaptation from the litigation process. The customer stay provisions take the “ample degree of discretion,” which is “appropriate for disciplined and experienced judges,” and place it into the hands of the alleged infringers. Infringers are not likely to be motivated by important notions of efficiency or justice; they’re likely to be motivated by self-interested gamesmanship of the system to their own advantage.

The proponents of the customer stay provisions claim that they’re necessary to help the little guy, but the provisions in both bills just aren’t drafted like that. Instead, they’re drafted to tie the hands of judges in countless cases that have nothing to do with small-time retailers and end users. The courts already have the power to stay proceedings when the equities tip in that direction, but these bills disrupt the judicial discretion on which the patent system has long depended. Customer stays certainly have their place, and that place is in the hands of judges who can take into account the totality of the circumstances. Judges should not be forced to make the important decision of whether to grant a stay based on overbroad and inflexible statutory language that goes far beyond its stated purpose.

CloudFlare Enjoined From Aiding Infringers: Internet Unbroken

The following post first appeared on the Center for the Protection of Intellectual Property (CPIP) blog, and it is reposted with permission here.

Just how far does a court’s power to enjoin reach into cyberspace? It’s clear enough that those directly posting or hosting infringing content are subject to an injunction. But what about a company such as CloudFlare that provides content delivery network and domain name server services? Does an injunction under Rule 65 against anyone acting in “active concert or participation” with an online infringer apply to an internet infrastructure company such as CloudFlare? CloudFlare recently argued that its service is “passive” and untouchable, but a district court vehemently—and rightly—disagreed.

The controversy started with the shutdown of the Grooveshark music streaming service pursuant to a settlement agreement with the major record label plaintiffs this past April. Back in September of 2014, Grooveshark and its two founders were found directly and indirectly liable for copyright infringement. After the district court held that their infringement was “willful,” thus subjecting them to potential statutory damages exceeding $736 million for the 4,907 works-in-suit, they consented to paying $50 million in damages and shutting down the site rather than risk it with a jury.

But the demise of Grooveshark was short-lived, and just days after publicly apologizing for failing “to secure licenses from right holders,” two copycat sites popped up at different top-level domains: and The record label plaintiffs filed a new complaint and obtained ex parte relief, including a temporary restraining order (TRO), against the new sites. Upon receipt of the TRO, Namecheap, the registrar for both sites, disabled the .io and .pw domain names. When another copycat site was established at, the domain name was quickly disabled by Dynadot, the registrar, after it received the TRO.

Undeterred, the defendants publicly taunted the plaintiffs and registered yet another copycat site at Rather than continuing this global game of domain name Whac-A-Mole, the plaintiffs served the TRO on CloudFlare, the service utilized by the defendants for each of the infringing domains. And this is where things got interesting. Rather than swiftly complying with the TRO, as the domain name registrars had done, CloudFlare lawyered up and contended that it was beyond the court’s reach.

In its briefing to the court, CloudFlare argued that it played merely a passive role for its customers—including the defendants and their copycat site—by resolving their domain names and making their websites faster and more secure. CloudFlare disavowed the ability to control any content on the copycat site, and it denied that it was in active concert or participation with the defendants:

Active concert requires action, and CloudFlare has taken none. Participation means assisting a defendant in evading an injunction. CloudFlare has not so assisted defendants and, in fact, has no ability to stop the alleged infringement. Even if CloudFlare—and every company in the world that provides similar services—took proactive steps to identify and block the Defendants, the website would remain up and running at its current domain name.

CloudFlare did not deny that the defendants utilized its services; it instead argued that the TRO would not remove the infringing site from the internet. Thus, CloudFlare’s position hinged on its own passivity and on the futility of enjoining it from providing services to the defendants.

A moment’s reflection reveals the superficiality of this position. The fact that CloudFlare had no control over the content of the copycat site was not dispositive. The question was whether CloudFlare aided the defendants, and there was no doubt that it did. It was not only the defendants’ authoritative domain name server, it also optimized and secured their copycat site. That the defendants could have used other services did not erase the fact that they were using CloudFlare’s services. And once CloudFlare was served with the TRO and made aware of the copycat site, its continued provision of services to the defendants constituted active concert or participation.

CloudFlare’s policy arguments were similarly unpersuasive. It suggested that the TRO “would transform a dispute between specific parties into a mandate to third parties to enforce” the plaintiffs’ rights “against the world in perpetuity.” Of course, that is not what happened here. The question was not who else in the world the TRO reached; the question was whether the TRO reached CloudFlare because it aided the defendants.

CloudFlare further argued that it could not be enjoined because “Congress explicitly considered and rejected granting such authority to the courts with respect to Internet infrastructure providers and other intermediaries for the purpose of making a website disappear from the Internet.” To enjoin it, CloudFlare proposed, would be to pretend that the Stop Online Piracy Act (SOPA) “had in fact become law.” This argument, however, completely ignored the fact that courts have long been empowered to enjoin those in active concert or participation with infringers.

In reply, the record label plaintiffs rebuffed CloudFlare’s claim that it was not aiding the defendants: “CloudFlare’s steadfast refusal to discontinue providing its services to Defendants – who even CloudFlare acknowledges are openly in contempt of this Court’s TRO – is nothing short of breathtaking.” They pointed to how CloudFlare continued to aid the defendants, even after being on notice of the TRO: “[T]he failure of an Internet service provider to stop connecting users to an enjoined website, once on notice of the injunction, readily can constitute aiding and abetting for purposes of Rule 65.”

In the real world, CloudFlare markets the benefits of its services to its customers. It touts its content delivery network as delivering “the fastest page load times and best performance” through its “34 data centers around the world.” It boasts having “web content optimization features that take performance to the next level.” It offers robust “security protection” and “visitor analytics” to its customers. And its authoritative domain name server proudly serves “43 billion DNS queries per day.” But when it came to the defendants’ copycat site, it claimed to be a “passive conduit” that in no way helped them accomplish their illicit goals. This disingenuousness is, to borrow the plaintiffs’ term, “breathtaking.”

District Judge Alison J. Nathan (S.D.N.Y.) made short work in rejecting CloudFlare’s shallow denials. She noted that there was no factual question that CloudFlare operated the defendants’ authoritative domain name server and optimized the performance and security of their copycat site. The question was whether these acts were passive such that CloudFlare was not in “active concert or participation” with the defendants. Judge Nathan held that the services CloudFlare provided to the defendants were anything but passive:

CloudFlare’s authoritative domain name server translates as entered in a search browser into the correct IP address associated with that site, thus allowing the user to connect to the site. Connecting internet users to in this manner benefits Defendants and quite fundamentally assists them in violating the injunction because, without it, users would not be able to connect to Defendants’ site unless they knew the specific IP address for the site. Beyond the authoritative domain name server, CloudFlare also provides additional services that it describes as improving the performance of the site.

Furthermore, Judge Nathan dismissed CloudFlare’s argument that it was not helping the defendants since they could simply use other services: “[J]ust because another third party could aid and abet the Defendants in violating the injunction does not mean that CloudFlare is not doing so.” And to CloudFlare’s concern that the TRO was overly broad, Judge Nathan reasoned that the issue before her was CloudFlare’s own actions, not those of other, possibly more attenuated, third parties: “[T]he Court is addressing the facts before it, which involve a service that is directly engaged in facilitating access to Defendants’ sites with knowledge of the specific infringing names of those sites.”

This TRO wasn’t about the “world at large,” and it wasn’t about turning the companies that provide internet infrastructure into the “trademark and copyright police.” It was about CloudFlare knowingly helping the enjoined defendants to continue violating the plaintiffs’ intellectual property rights. Thankfully, Judge Nathan was able to see past CloudFlare’s empty and hyperbolic position. Protecting intellectual property in the digital age is difficult enough, but it’s even more challenging when services such as CloudFlare shirk their responsibilities. In the end, reason trumped rhetoric, and, best of all, the internet remains unbroken. In fact, it’s now even better than before.

Further reading: Leo Lichtman, Copyright Alliance, Bringing Accountability to the Internet: Web Services Aiding and Abetting Rogue Sites Must Comply With Injunctions

Ninth Circuit Gives “Short Shrift” to First Amendment in Garcia v. Google

The Ninth Circuit, sitting en banc, released its long-awaited opinion in Garcia v. Google this past Monday. I assume readers are familiar with the facts, but the gist is that Garcia claimed copyright ownership over her five-second performance within a fourteen-minute trailer for a film. The filmmaker uploaded the trailer to YouTube, and Garcia sent Google several DMCA takedown notices to have it removed. When those notices were ignored, Garcia sued Google and sought a preliminary injunction. The district court denied that relief, but a panel of the Ninth Circuit reversed and issued a temporary injunction. The en banc court has now dissolved the panel’s injunction and affirmed the district court below.

In an earlier post, I looked at Google’s claim that an injunction against it would constitute a prior restraint violative of the First Amendment. The Ninth Circuit panel, in an opinion by then-Chief Judge Alex Kozinski, brushed aside Google’s prior restraint argument with a single paragraph. That analysis was quite simple: Since the context was Google’s public performance of somebody else’s copyrighted work, and since Google had not argued that its performance was excused by the fair use doctrine or the idea/expression dichotomy, it could not be a prior restraint. The prior restraint doctrine is only concerned with suppressing speech that may turn out to be protected. Without either of the two First Amendment-based defenses in play, it was a simple matter for the panel to determine that its injunction would not suppress protected speech.

In Monday’s opinion, the en banc court takes issue with the panel’s injunction: “The takedown order was unwarranted and incorrect as a matter of law, as we have explained above. It also gave short shrift to the First Amendment values at stake.” I don’t think this criticism is fair. The prior restraint argument was a nonstarter since Google hadn’t raised any First Amendment defenses. What’s more, after complaining of the panel’s curt analysis, the en banc court then proceeds to give its own “short shrift” to the First Amendment:

Although the intersection between copyright and the First Amendment is much-debated, [n.14] the Supreme Court teaches that copyright is not “categorically immune from challenges under the First Amendment.” Eldred, 537 U.S. at 221 (internal citation omitted). To be sure, this is not a case of garden-variety copyright infringement, such as seeking to restrain the use of copyrighted computer code. The panel’s takedown order of a film of substantial interest to the public is a classic prior restraint of speech. Alexander v. United States, 509 U.S. 544, 550, 113 S.Ct. 2766, 125 L.Ed.2d 441 (1993) ( “Temporary restraining orders and permanent injunctions—i.e., court orders that actually forbid speech activities—are classic examples of prior restraints.”). Prior restraints pose the “most serious and the least tolerable infringement on First Amendment rights,” Hunt v. NBC, 872 F.2d 289, 293 (9th Cir.1989) (citation omitted), and Garcia cannot overcome the historical and heavy presumption against such restraints with a thin copyright claim in a five-second performance.

[n.14: See, e.g., Joseph P. Bauer, Copyright and the First Amendment: Comrades, Combatants, or Uneasy Allies?, 67 Wash. & Lee L.Rev. 831 (2010); Mark A. Lemley & Eugene Volokh, Freedom of Speech and Injunctions in Intellectual Property Cases, 48 Duke L.J. 147 (1998).]

The cite to Eldred is curious. There, the Supreme Court did indeed say that copyright laws are not “categorically immune” to First Amendment challenges. However, the context was a substantive challenge to the CTEA, not a procedural challenge like here with Google’s prior restraint argument. The issue in Eldred was whether the CTEA should be analyzed under heightened First Amendment scrutiny. The Court held that, since the CTEA and the First Amendment were balanced at the definitional level through the fair use doctrine and the idea/expression dichotomy, there was no need to apply heightened scrutiny. Mere rational basis scrutiny would suffice. Importantly, the level of scrutiny applicable to such a substantive challenge tells us nothing about whether an injunction is a prior restraint.

The en banc court pays lip service to the First Amendment by declaring the trailer to be of “substantial interest to the public,” and it claims that the panel’s injunction is “a classic prior restraint of speech.” This makes little sense, however, given the context of Garcia’s copyright claim. Of course, the trailer itself is protected speech, but that First Amendment protection isn’t the same for the copyright owner as it is for Google. Google can’t violate the copyright owner’s public performance right simply because the public finds the trailer interesting. The First Amendment only protects Google’s use of the copyrighted trailer if that use is excused by the fair use doctrine or the idea/expression dichotomy, but Google never argued either defense.

As noted in the law review article by Professors Mark Lemley and Eugene Volokh, cited by the en banc court in the footnote quoted above, “preliminary injunctions are granted pretty much as a matter of course” in copyright cases. The reason such injunctions aren’t prior restraints is because, at least in run-of-the-mill cases of nontransformative copying, there is no risk that the speech will turn out to be protected by the First Amendment. Professors Lemley and Volokh explain why this is so:

The danger of preliminary injunctions is that they may temporarily suppress speech that ultimately proves to be protected. If a judge can, at the time of the preliminary injunction hearing, make a reliable finding that the speech is constitutionally unprotected, rather than just that it might be unprotected, then there should be no problem with issuing the preliminary injunction.

This could happen whenever the defendant has made identical or nearly identical copies of the plaintiff’s works, and there is no claim of fair use but only some other copyright defense (such as that defendant has a supposedly valid license, or that plaintiff’s copyright wasn’t properly renewed, or something along those lines). Most cases alleging outright piracy, as well as most claims that the defendant has exceeded the boundaries of its license, would fall within this category.

In this situation, the defendant’s conduct is not constitutionally protected speech, even if the defendant ultimately proves that it’s not a copyright infringement.

Google is able to publicly perform copyrighted works because copyright owners grant it licenses when the works are uploaded to the YouTube platform. While the works themselves are protected by the First Amendment, Google’s public performance of those works is protected only by the licenses it is granted. It needs those licenses because the First Amendment doesn’t otherwise protect its performance of the copyrighted works. It’s a simple matter for a court to “make a reliable finding that the speech is constitutionally unprotected,” especially where, as here, Google never argued that either of the First Amendment-based defenses applies to its performance.

The en banc court’s invocation of the prior restraint doctrine thus misses this point. Garcia’s claim was anything but run-of-the-mill, but there was never any doubt that Google’s public performance of the trailer was not protected by the First Amendment. It may have been protected by the license granted by the filmmaker when he uploaded the trailer, but then it’s the license that protects Google’s performance, not the First Amendment. By ignoring the context of Garcia’s copyright claim, and by ignoring the fact that Google’s use could only be protected by a license, the en banc court itself gives “short shrift” to the First Amendment. Assuming Garcia is not the copyright owner, the case should simply be thrown out for lack of standing. There’s no need to wax poetic on the prior restraint doctrine when Google’s performance of the trailer isn’t protected by the First Amendment in the first place.

Follow me on Twitter: @devlinhartline

© 2015 Devlin Hartline. Licensed under the Law Theories Public License 1.0.