Stop Online Piracy Act and the DMCA

By , November 10, 2011.

The online safe harbors for copyright infringement liability, enacted as part of the DMCA in 1998, have been credited for everything from saving the web to removing former Egyptian president Hosni Mubarak from power.

But while there is certainly disagreement over the scope of immunity for online service providers — appeals are currently pending in the Second and Ninth Circuits over many of the details of the provisions, for example — the law is generally considered to strike the appropriate balance where it applies.

Now with the introduction of the Stop Online Piracy Act in the US House, critics are concerned that the bill — specifically Section 103, which creates a notice mechanism for preventing sites “dedicated to theft of US property” from profiting off infringement similar to the notice-and-takedown procedure in the DMCA for removing infringing content uploaded by a service provider’s users — will upend this balance.

The EFF warns that the bill threatens to “effectively eliminate the DMCA safe harbors.” Public Knowledge writes that “SOPA would undermine the DMCA.” The Internet abounds with others raising the alarm over what would become of DMCA safe harbors should SOPA pass.

I had previously written about the need for copyright law to continue to adapt alongside technology. It makes as much sense to freeze the law relating to the online world as it was a decade ago as it does to freeze the technologies that shape the web as they were.

The Stop Online Piracy Act in many ways represents an update to the Digital Millennium Copyright Act — DMCA 2.0 — rather than an end-run around it. The existing safe harbors are good for certain things — user-generated sites and social networking sites to name just two — but not good for others.

Active and Passive Infringement

One major feature of the safe harbors, which illustrates both why they are ineffective against the type of commercial piracy targeted by SOPA and why SOPA enhances rather than eliminates these limitations on liability, is that they are primarily concerned with passive infringement: infringement that occurs through services that is inevitable, but which would be overly burdensome to fully prevent.

The online safe harbors were intended to encourage content owners and service providers to cooperate in order to “detect and deal” with online infringement.  They were also meant to give “greater certainty to service providers concerning their legal exposure for infringements that may occur in the course of their activities.” 1Final Conference Report on H.R. 2281, House 105-796, pg. 72 (Oct. 8, 1998). Note the passive “infringements that may occur” — the safe harbors weren’t made for service providers who set out to enable or facilitate infringement.

Perhaps the most popular safe harbor is the one laid out in 17 USC § 512(c), which immunizes service providers from liability for “Information residing on systems and networks at direction of users.” The text of 512(c) supports the idea that the safe harbor was designed for inadvertant and incidental infringement. It shields service providers for copyright infringement liability that arises only “by reason of the storage at the direction of a user of material.” The safe harbor does not apply if the service provider has actual knowledge that material or activity using the material infringes copyright, or if the provider receives a directly attributable financial benefit where it has the right and ability to control infringing activity.

It shouldn’t be controversial to say the DMCA safe harbors were crafted as a shield for service providers, not as a sword against creators and copyright holders. The notice and takedown procedure in Section 103 of the Stop Online Piracy Act thus builds upon the originally intended goals of the DMCA to target only those sites that actively set out to benefit from third-party infringement.

The provisions can only be used against sites that are “primarily designed or operated” to offer services in a way that “engages in, enables, or facilitates” copyright infringement; sites that take “deliberate actions to avoid confirming a high probability” that the site is used to carry out infringement; or sites that promote their use to engage in infringement “as shown by clear expression or other affirmative steps taken to foster infringement.”

Taken as a whole, the bill can only be used against sites that have purposefully taken actions to profit off piracy, actions that are well beyond the scope of the DMCA safe harbors.

DMCA safe harbors remain alive and well

This is not a departure from existing law. The third definition of a “site dedicated to theft of US property” repeats word-for-word the Supreme Court’s articulation of inducement — “one who distributes a device with the object of promoting its use to infringe copyright, as shown by clear expression or other affirmative steps taken to foster infringement, is liable for the resulting acts of infringement by third parties.” [Emphasis added.] 2MGM v. Grokster, 545 US 913 (2005). And as one court noted, “inducement liability and the Digital Millennium Copyright Act safe harbors are inherently contradictory. Inducement liability is based on active bad faith conduct aimed at promoting infringement; the statutory safe harbors are based on passive good faith conduct aimed at operating a legitimate internet business.” [Emphasis added.] 3Columbia Pictures v. Fung, CV 06-5579 SVW, Order granting plaintiff’s motion for summary judgment on liability (CD Cali. Dec. 21, 2009).

It’s also important to keep in mind that nothing in the bill negates the safe harbors. A copyright holder can only proceed in court against a site after it has sent a notice to an advertising or financial transaction provider and either those providers failed to comply with the notice or the site operator sends a counter-notification. Even then, the court is limited solely to granting an injunction against the site to cease and desist from doing further any of the actions described above that made it subject to a notification in the first place. 4Which make the claims that SOPA would lead to increased litigation, “shakedowns” of new services, or a boon for trial lawyers absurd: copyright holders already can sue a site for copyright infringement, and if successful, can collect monetary damages. A cause of action that is much narrower than copyright infringement, for which only limited injunctive relief is available, isn’t exactly attractive to someone looking to make a buck.

In practice, the DMCA notice and takedown provisions are ineffective against site operators who take an active role in facilitating, promoting, or turning a blind eye toward the use of their sites for infringing copyright. Many creators would find it a full time job to send notices against these types of sites, and most find that specific content removed quickly returns elsewhere. This results in sites playing a game of “catch me if you can” while continuing to profit off the work of others. And they are especially ineffective against sites that are directed at and easily accessible by US residents but located outside the US and dismissive of US law.

Rogue sites sidestep the rule of law by using the DMCA safe harbors to shield themselves from liability while they take deliberate steps to profit off of piracy. With the Stop Online Piracy Act, the US House is hoping to truly secure the exclusive rights of creators; doing so not only protects creators but also ensures that the development of innovative and sustainable services for consumers to access and enjoy media and content is not hamstrung.

References

References
↑1 Final Conference Report on H.R. 2281, House 105-796, pg. 72 (Oct. 8, 1998).
↑2 MGM v. Grokster, 545 US 913 (2005).
↑3 Columbia Pictures v. Fung, CV 06-5579 SVW, Order granting plaintiff’s motion for summary judgment on liability (CD Cali. Dec. 21, 2009).
↑4 Which make the claims that SOPA would lead to increased litigation, “shakedowns” of new services, or a boon for trial lawyers absurd: copyright holders already can sue a site for copyright infringement, and if successful, can collect monetary damages. A cause of action that is much narrower than copyright infringement, for which only limited injunctive relief is available, isn’t exactly attractive to someone looking to make a buck.

How the Stop Online Piracy Act will hit what it aims at

By , October 31, 2011.

Creators are often told they have to learn how to “compete with free” because of online piracy. To some extent, this is true, but certainly not a new idea — piracy existed way before the Internet became so prominent in our lives. 1See, for example, Martin Luther’s Warning to the Printers, Wittenberg (1541): “Avarice now strikes / and plays this knavish trick on our printers whereby others are instantly reprinting [our translation] / and are thus depriving us of our work and expenses to their profit, / which is a downright public robbery / and will surely be punished by God / and which is unworthy of any honest Christian.” Creators have been adapting, and continue to adapt, to the realities of digital technology — through a combination of new business models, technology, and carefully constructed legislation.

But what creators shouldn’t have to do is compete with paid. That is, the law should provide little leeway for those who profit off the infringement of the core exclusive rights protected by copyright law for centuries — rights that the US Constitution empowers Congress to secure to authors in order to promote the progress of the useful arts and sciences for all of society.

Most online firms recognize the value of copyright and creative works — rogue sites, however, believe that all of that value should go directly into their own bank accounts. Recognizing the ease of doing this and the harm it causes to creators, consumers, and the economy in general, Congress has pursued legislation addressing the problem of rogue sites: the PROTECT IP Act was introduced in the Senate last May, and the Stop Online Piracy Act (SOPA) in the House this past week.

Some of the early criticism of the House bill has centered around Section 103 of SOPA, which would provide a system for notifying advertising and payment providers of Internet sites that are dedicated to piracy and using their services.

The EFF predicts “we’re going to see a flurry of notices anyway – as we’ve learned from the DMCA takedown process, content owners are more than happy to send bogus complaints.” Larry Downes, writing on behalf of TechFreedom, claims the bill is “drafted to ensure maximum litigation.” And Gary Shapiro, president of the Consumer Electronics Association, notes “the potential for abuse by the notoriously litigious content industry is clear.”

If you read the bill, however, it is difficult to see these predictions coming true. Section 103 is appropriately called a “Market-based system to protect U.S. customers and prevent U.S. funding of sites dedicated to theft of U.S. property” — the market will naturally mitigate against any flurry of notices. The robust procedures envisioned in the bill, along with the penalties for abusing the notice process, also make it an unlikely candidate for abuse.

The Market-based System

The notification system under SOPA is similar to the one provided by the DMCA where content owners can notify service providers of infringing material uploaded by users. By comparing the two notification processes, we can see that, contrary to early criticisms, the notification process under SOPA is less likely to be abused than the DMCA.

A service provider has an incentive to remove content when it receives a DMCA takedown notice. Expeditiously removing content once it is notified preserves its safe harbor from any liability it might face. There is also little downside to removing content for the service provider. The DMCA shields the service provider from any legal claims from a user whose content was removed. 217 USC § 512(g). And, in most cases, any subsequent business repercussions of content takedown are slight — at worst, a user of a free service takes his business elsewhere.

While an ad provider or payment processor may, in some cases, be liable for contributing to copyright infringement, 3See Warner Bros. v. Triton Media, Consent Judgment, No. CV 10-6318-GW (CD Cali, Oct 27, 2010), defendant liable for “providing advertising consulting and referrals for, and/or providing other material assistance to” infringing websites. this requires a rare set of circumstances. 4See Perfect 10 v. Visa International, 494 F.3d 788 (9th Cir. 2007), payment processor not liable for copyright infringement for processing credit card transactions on infringing sites. Most ad and payment service providers, then, do not have the same additional legal incentives to comply with SOPA takedowns as online service providers have to comply with DMCA takedowns.

They also don’t have business incentives to comply with bad faith notices under SOPA. Each site they are ordered to block is, presumably, a paying customer or revenue source. I doubt many businesses would be happy to rubber-stamp any and every order that cuts into their bottom line without some way of making sure the site at issue is one that genuinely falls within the scope of this law.

Due Process

The bill provides a fair process for resolving disputes between content owners and ad and payment providers when they disagree about whether a particular site is dedicated to the theft of US property.

When an ad or payment provider receives a notice under SOPA, one of three things may happen: (1) the identified site files a counter-notice, (2) no counter-notice is filed and the provider complies with the notice and prevents its service from being used by the infringing site, or (3) no counter-notice is filed but the provider fails to comply with the original notice.

In the first and third scenarios, absolutely nothing happens to either the identified site or provider unless the copyright owner chooses to pursue further. In those cases, a copyright owner is limited to filing an action against the identified site’s operator, or the site itself if the operator is not locatable or resides outside the US. 5This dual in personam/in rem style cause of action is reminiscent of the one in the Anticybersquatting Consumer Protection Act, see Caesars World v. Caesars-Palace.com, 112 F.Supp.2d 502 (ED Va 2000) for a discussion of the constitutional propriety of the Act.

When an action is commenced, the court may issue an injunction — governed by the Federal Rules of Civil Procedure, the same rules that govern injunctions in any federal civil lawsuit — against the site owner to cease infringing activity. The copyright owner may then serve court orders against ad and payment providers providing services to the identified site.

When one of these providers gets such an order, it must use “technically feasible and reasonable measures” to prevent its services from being used by the identified site, but once it has put these measures into place, it has no duty to monitor.

If the provider still refuses to comply, the copyright owner may, upon a showing of probable cause that the provider has refused to comply, request further action by the court. Under the statute, the court would issue an order to show cause to the provider. If the provider cannot show cause why it has failed to comply, the court may order it to comply, or, in the case of a knowing and willing refusal to comply, it may “impose an appropriate monetary sanction”.

These are the only legal remedies available to a copyright owner under this section of SOPA. Copyright owners cannot shut down a site, they cannot collect monetary damages (any monetary sanctions go directly to the court). Success under this market-based system is getting an ad or payment provider to prevent its services from being used by a site dedicated to the theft of US property to profit off that theft.

Little room for mistakes

Under the DMCA, a content owner who “knowingly materially misrepresents” that content is infringing is liable for any damages, attorney fees, and costs incurred by the user as a result of the content being taken down. 617 USC § 512(f). One of the few cases that dealt with this part of the DMCA involved the infamous takedown of the dancing baby video on YouTube. In 2008, the Northern District Court of California concluded that a takedown notice requires a “good faith consideration of whether a particular use is fair use”; otherwise, a content owner may be liable for misrepresentation. 7Lenz v. Universal Music Corp., 572 F.Supp.2d 1150, 1156 (ND Cali 2008).

Even so, as the court pointed out, the amount of damages a content owner would face for such a misprepresentation would likely be “nominal.” Having a video removed from YouTube for a few days may be annoying, but provable financial damage? Probably not much. 8See Ben Sheffner’s discussion of the 512(f) damages provision in Lenz for more on this.

SOPA provides a similar provision against misrepresentation in notices — but while the language mirrors that in the DMCA, there are two practical realities that make it different.

First, a good faith effort to determine that an entire site is “dedicated to theft of US property” under the definition of the bill requires considerable more effort than determining whether a single file is infringing. The notification itself requires substantially more investigation than a DMCA notice. Under the DMCA, a copyright owner need only identify what work is being infringed and the content that is infringing. 917 USC § 512(c)(3). Under SOPA, the copyright owner must show, among other things, “specific facts to support the claim that the Internet site, or portion thereof, is dedicated to theft of U.S. property” and “clearly show that immediate and irreparable injury, loss, or damage will result” to the copyright owner in the absence of timely action; “Information reasonably sufficient to establish that the payment network provider or Internet advertising service is providing payment processing or Internet advertising services for such site”; and identification of evidence that indicates the site is US-directed.

Second, the risk of making a material misrepresentation is much higher. The operator of a site whose sources of income have been threatened is far likelier to push back than a user whose video was taken down. And unlike the nominal damages present in a DMCA takedown, the loss of ad revenues and credit card transactions because of a bad faith takedown could add up.

No law is immune from abuse, and copyright law shouldn’t be an exception. But the penalties for erroneous notices under SOPA have teeth, the procedures and remedies involved don’t encourage over-litigiousness. These provisions strike a proper balance that should prove to be effective against rogue sites and protective of good faith actors.

Taking the Profit out of Piracy

The provisions of Section 103 of the Stop Online Piracy Act will certainly continue to evolve during the legislative process; if SOPA and the PROTECT IP Act are successfully voted on, they will need to be reconciled before becoming law.

But these provisions represent a good start for creators who have long noted the ease (and injustice) of profiting from online piracy and escaping liability. They are aimed in theory at rogue sites, and in practice will be used against only rogue sites. Safeguards are built into the bill to limit abuse.

Web services who are acting legitimately and legally should welcome rogue sites legislation because effective protection of creative labor is vital to a functioning online marketplace, and a functioning online marketplace benefits us all.

References

References
↑1 See, for example, Martin Luther’s Warning to the Printers, Wittenberg (1541): “Avarice now strikes / and plays this knavish trick on our printers whereby others are instantly reprinting [our translation] / and are thus depriving us of our work and expenses to their profit, / which is a downright public robbery / and will surely be punished by God / and which is unworthy of any honest Christian.”
↑2 17 USC § 512(g).
↑3 See Warner Bros. v. Triton Media, Consent Judgment, No. CV 10-6318-GW (CD Cali, Oct 27, 2010), defendant liable for “providing advertising consulting and referrals for, and/or providing other material assistance to” infringing websites.
↑4 See Perfect 10 v. Visa International, 494 F.3d 788 (9th Cir. 2007), payment processor not liable for copyright infringement for processing credit card transactions on infringing sites.
↑5 This dual in personam/in rem style cause of action is reminiscent of the one in the Anticybersquatting Consumer Protection Act, see Caesars World v. Caesars-Palace.com, 112 F.Supp.2d 502 (ED Va 2000) for a discussion of the constitutional propriety of the Act.
↑6 17 USC § 512(f).
↑7 Lenz v. Universal Music Corp., 572 F.Supp.2d 1150, 1156 (ND Cali 2008).
↑8 See Ben Sheffner’s discussion of the 512(f) damages provision in Lenz for more on this.
↑9 17 USC § 512(c)(3).

Polis Seeks to Make Copyright More Complicated and Uncertain

By , September 28, 2011.

Rep. Jared Polis thinks something must be done about copyright.

The freshman second-term Colorado Congressman, who sold his online greeting card company bluemountain.com to Excite@Home for $780 million in 1999, introduced legislation on September 14th that seeks to simplify copyright law regarding sound recordings made before 1972.

H.R.2933 — The Sound Recording Simplification Act — is short, consisting of its title and two lines. All it does is delete 17 USC § 301(c). In a statement announcing the bill, Polis explained:

The variation in protection from state to state hinders the ability to have a consistent business approach regarding rights ownership. The scope of protection and what would constitute acceptable use are inconsistent due to the lack of detailed precedent. Additionally, state law does not have provisions that account for modern technology, such as streaming music websites, and songs as data are increasingly easy to transmit.

But upon closer look, Polis’s bill would fail completely to address these issues. At best, it would change nothing. At worst, it would lead to more uncertainty rather than less, and give rise to a host of unforseen consequences.

Federal and state protection for sound recordings

Recorded music consists of two distinct copyrighted works: the underlying musical composition and the sound recording of that composition. US copyright law has expressly provided protection for musical compositions since 1831, but it wasn’t until 1972 that protection was extended to sound recordings. 1The New York Court of Appeals provides an impressively thorough history of sound recordings, copyright, and preemption — beginning in 15th century England — in Capitol Records v. Naxos, 830 NE 2d 250 (2005). Note, however, that federal copyright protection did not retroactively extend to sound recordings made before the passage of the 1972 legislation.

Though the federal government did not protect sound recording copyrights before 1972, states did. By the early 70s, roughly half of state legislatures had adopted criminal provisions aimed at curbing music piracy.

Prior to the Copyright Act of 1976, the federal government and the states concurrently exercised copyright power — for all types of works, not just sound recordings. 2See Goldstein v. California, 412 US 546, 560 (1973). That all changed with the passage of the 1976 Act. Congress gave the federal government exclusive authority over copyright; “all legal or equitable rights that are equivalent to any of the exclusive rights within the general scope of copyright” under any state’s statutes or common law are preempted by US copyright law. 317 USC § 301(a).

Preemption and pre-1972 sound recordings

But a problem was anticipated: what would become of sound recordings made before 1972?

One way to interpret the preemption provision would be that it had no effect on pre-1972 sound recordings. Since they weren’t covered under the scope of federal protection, state laws protecting them weren’t preempted.

But the provision could be interpreted another way. In a 1975 hearing concerning the revisions that would lead to the 1976 Act, the Department of Justice raised concerns about this possibility:

This language could be read as abrogating the anti-piracy laws now existing in 29 states relating to pre-February 15, 1972, sound recordings on the grounds that these statutes proscribe activities violating rights equivalent to * * * the exclusive rights within the general scope of copyright. * * *” Certainly such a result cannot have been intended for it would likely effect the immediate resurgence of piracy of pre-February 15, 1972, sound recordings.

In other words, a court could strike down state piracy laws if it interpreted the statute this way, putting every sound recording made before 1972 into the public domain.

So Congress added § 301(c) to make it absolutely clear that this is not what it intended. At the same time, it recognized that state piracy laws did not provide protection for only a “limited time” as federal copyright law did — sound recordings gave perpetual protection. So it also put in a sunset provision to the clause, 75 years from the first day sound recordings were covered under federal law (the Copyright Term Extension Act increased that time by 20 years). 4H.R. Rep. No. 94-1476, pg 133.

Today, §301(c) provides that:

With respect to sound recordings fixed before February 15, 1972, any rights or remedies under the common law or statutes of any State shall not be annulled or limited by this title until February 15, 2067. The preemptive provisions of subsection (a) shall apply to any such rights and remedies pertaining to any cause of action arising from undertakings commenced on and after February 15, 2067. Notwithstanding the provisions of section 303, no sound recording fixed before February 15, 1972, shall be subject to copyright under this title before, on, or after February 15, 2067.

This is the provision that Polis’s bill would delete. The only thing this would accomplish is reinstating the uncertainty that led to its addition 35 years ago. It would not bring pre-1972 sound recordings under federal copyright protection. And, if the DOJ’s original concerns are unfounded and courts interpret the statute to mean that state piracy laws aimed at pre-1972 sound recordings are not preempted, than Polis just gave those recordings perpetual protection.

I highly doubt Rep. Polis is looking to extend copyright protection forever.

Haste makes waste

In Polis’s rush to do something about pre-1972 sound recordings, he not only ends up doing nothing, he steamrolls over the legitimate concerns of those working on solutions.

In 2009, hearing concerns that current law addressing pre-1972 recordings can cause problems, Congress directed the US Copyright Office to conduct a study on the “desirability of and means for” bringing those recordings under federal copyright law. It specifically asked the Office to look at “the effect of federal coverage on the preservation of such sound recordings, the effect on public access to those recordings, and the economic impact of federal coverage on rights holders.”

The Copyright Office began seeking input for its study last November. 5Federal Copyright Protection of Sound Recordings Fixed Before February 15, 1972, Notice of Inquiry, 75 F.R. 67777 (Nov. 3, 2010). In this first round of comment seeking, it raised thirty potential questions about implementing federal protection for these recordings.

The questions cover preservation and access issues; economic issues, especially relating to value of recordings and ownership of rights concerns; and implementation issues, like term length and due process concerns.

To give just a few examples, the Copyright Office asks:

4. Would bringing pre-1972 sound recordings under Federal law—without amending the current exceptions—affect the ability of such institutions to provide access to those recordings? Would it improve the ability of libraries and archives to make these works available to researchers and scholars; and if so, in what way? What about educational institutions, museums, and other cultural institutions?

16. Under Federal law the owner of the sound recording will generally be, in the first instance, the performer(s) whose performance is recorded, the producer of the recording, or both. Do State laws attribute ownership differently? If so, might that lead to complications?

23. If the requirements of due process make necessary some minimum period of protection, are there exceptions that might be adopted to make those recordings that have no commercial value available for use sooner?

27. Could the incorporation of pre-1972 sound recordings potentially affect in any way the rights in the underlying works (such as musical works); and if so, in what way?

Polis’s bill addresses none of these questions. To his credit, this is quite an accomplishment — copyright legislation that everyone, from rightsholders, to downstream copyright users, to libraries, museums, and educational institutions, would agree is a bad idea.

I’m left scratching my head over how Rep. Polis can think expedience is better than comprehensive study for finding solutions, or how one can craft legislation that is so bad that it not only doesn’t solve whatever problem you’re trying to fix, it actually makes those problems worse.

Perhaps this is just business as usual for Polis, because a similar thing happened a few years ago with Colorado’s Amendment 41. The ballot measure, which Polis helped sponsor and back in 2006, was ostensibly targeted at curbing lobbyist influence, but ended up making “scholarships for children of public employees and performance awards for employees” illegal. 6John Straayer, Direct Democracy’s Disaster, National Conference of State Legislatures Magazine (March 2007). It led to a fire-fighters association indicating that it “might have to close its foundation, which had made scholarships available for fire-fighters’ children and provided funeral assistance for fire-fighters’ survivors.” When these problems became apparent, Polis and the amendment’s other sponsors said that “they didn’t really mean what it said.”

In the same way, the Sound Recording Simplification Act doesn’t mean what Polis thinks it means.

References

References
↑1 The New York Court of Appeals provides an impressively thorough history of sound recordings, copyright, and preemption — beginning in 15th century England — in Capitol Records v. Naxos, 830 NE 2d 250 (2005).
↑2 See Goldstein v. California, 412 US 546, 560 (1973).
↑3 17 USC § 301(a).
↑4 H.R. Rep. No. 94-1476, pg 133.
↑5 Federal Copyright Protection of Sound Recordings Fixed Before February 15, 1972, Notice of Inquiry, 75 F.R. 67777 (Nov. 3, 2010).
↑6 John Straayer, Direct Democracy’s Disaster, National Conference of State Legislatures Magazine (March 2007).

Copyright Uncertainty

By , August 15, 2011.

Is copyright law too uncertain?

That certainly seems to be the prevailing view. Any number of copyright critics point to the uncertainty of the law — especially for online users, who are in closer contact with copyright issues than any previous generation has been — as one of its biggest flaws.

As attorney Steven Horowitz notes in a forthcoming article, the conventional argument is that this uncertainty “deters lawful uses of copyrighted works.” But, as you may guess from the title of the article, Copyright’s Asymmetric Uncertainty, copyright is only uncertain depending on the perspective. Those issues important to copyright holders, says Horowitz, are clear; issues important to copyright users are not so clear.

I was struck by how closely Horowitz’s distinction between what is clear and what is unclear in copyright law mirrors the distinction between creative and consumptive infringement that I had I previously wrote about — though, upon reflection, it might be more accurate to label these two categories “creative appropriation” and “consumptive appropriation”; use of a copyrighted work either creatively or consumptively is not necessarily infringement.

While there is plenty of overlap, in general, the line between creative appropriation and creative misappropriation (or infringement), relies largely on substantive doctrines of copyright law: the idea/expression distinction, substantial similarity, and fair use. Broadly speaking, the new work is compared to the original work. The line between consumptive appropriation and infringement, however, is governed by more regulatory provisions; a comparison between the two works is unnecessary since this type of appropriation typically consists of verbatim or near verbatim copies of the original work.

This isn’t just arbitrary line-drawing. As I said in my original post on the subject, recognizing this distinction is important to discussions of copyright law because each category has its own separate issues. “Uncertainty” is one of those issues: while there may be confusion over the rules that govern creative uses of copyrighted works, that is not the case when dealing with outright piracy of existing works.

Horowitz explains copyright’s asymmetric uncertainty by first discussing what issues are most important to copyright holders:

Although there are many who would produce expressive works irrespective of copyright protection, those whose production depends on copyright care most about three features of the law. They want (1) a reliable entitlement, (2) prohibiting at least pure or close copying of their work, which is (3) protected by remedies sufficient both to deter copying and to compensate for any losses that result from it. These are the core protections that ensure meaningful incentives to produce copyrightable expression, for without them a copyright holder might have to compete against others who sell identical copies of his own work. Such competition against perfect substitutes would drive down the price toward the near-zero marginal cost of the copy, leaving him without the ability to profit from the work or to recoup the costs of its production. Uncertainty in these three salient features of the law would spell trouble for copyright holders, but fortunately the features are predictable in practice.

In other words, copyright holders are most concerned with consumptive appropriation of their works. That shouldn’t come as a surprise.

That’s not to say copyright holders are not concerned with creative appropriation of their works — I think it goes without saying that the level of this concern varies widely from person to person. Some copyright holders strictly police against most creative appropriations, some may evaluate each use on a case-by-case basis — perhaps asserting their rights against commercial uses but not worrying about noncommercial uses — and some may welcome (or even encourage) such uses.

But most copyright holders’ primary concern is about control over how and where their complete works are copied and distributed.

Horowitz next looks at the issues of importance to copyright users:

Most affected by copyright’s uncertainty are the potential users of copyrighted works, those who want to use a work in ways not intended or expected by the copyright holder. For example, users may mix pieces of prior works together or write new stories using characters from old ones. The intent of the user is not to produce or sell verbatim copies of a preexisting work but to put some portion of a work to a new purpose. The potential user is most concerned with whether his expressive activity will constitute infringement and if so whether the remedies for infringement will be proportional to the harm caused—in other words whether his liability has a reasonable ceiling. Copyright makes answering both of these questions difficult. A potential user cannot predict with confidence whether a contemplated use will be deemed infringing or whether the damages will be manageable or devastating.

Most of copyright’s uncertainty falls within the scope of creative appropriation, not consumptive appropriation. One would think, then, that it wouldn’t be an issue when discussing piracy.

But what happens is that the uncertainty criticism is used as an across-the-board attack on copyright — as though piracy cannot be addressed until courts or Congress figure out, for example, a fair use rule that is both applicable to a wide variety of situations and easy enough for a fifth grader to understand. Any uncertainty that exists primarily in the area of creative appropriation shouldn’t be used as an argument against reasonable efforts to curb the harms of consumptive misappropriation.

The PROTECT IP Act applies to sites that have “no significant use other than engaging in, enabling, or facilitating the reproduction, distribution, or public performance of copyrighted works, in complete or substantially complete form, in a manner that constitutes copyright infringement.” Similarly, the Commercial Felony Streaming Act only applies to criminal infringement, where the willfulness requirement forecloses its application to most creative infringements. Doctrines like fair use, substantial similarity, and the idea/express distinction play little role in the application of proposed laws like these.

Copyright Legislation 2.0

By , June 13, 2011.

Imagine if the web had not progressed past the technology available around the mid 1990s, when it made its way into the mainstream.

No Flash or JavaScript, no CSS, PHP, or XML, no widgets or APIs. Just plain vanilla HTML (and maybe some server-side scripting if you knew what you were doing).

It would be hard to imagine a web like this today. Today’s web allows a myriad of ways for people to engage in communication, commerce, social networking, entertainment, and learning. This is possible because the technology behind the web continued to progress, rather than being frozen in place.

Freezing the technology in place would make little sense. Yet, when it comes to the legal framework that protects copyright and content creators, there are some who call foul whenever new legislation is proposed, who believe it makes perfect sense for the law to be frozen in place while technology rapidly advances.

The underlying idea seems to be that unchecked, wide-scale copyright infringement is just how things are going to be from now on. Content creators need to get busy adapting or get busy dying.

But why shouldn’t the law continue to adapt as well?

Currently, two bills are pending in Congress — S.978 and the PROTECT IP Act. Both address specific problems. S.978 standardizes the criminal penalties for streaming large amounts of copyrighted works with the penalties of copying and distributing large amounts of copyrighted works, since the harm from both types of piracy is the same. The PROTECT IP Act targets websites whose sole purpose is infringement. These sites are easy to set up, and the ability to profit from advertising or subscriptions provides an incentive to create them. Copyright owners find the DMCA takedown provisions ineffective against such sites and civil lawsuits are difficult to bring against such sites because of their nature. Legitimate advertising and financial transaction providers are also in a bind with these sites since they expose themselves to liability whenever they cut service off to a user.

In other words, laws like these take into consideration the evolution of the online world. As the internet matures and new business models for providing content develop, new problems emerge. The same is true in most areas of life.

Consider automobiles. Although driving has been the primary means of transportation for decades, laws continue to be tweaked for safety, environmental, and other considerations. Just ten years ago, very few states had so-called “move over” laws. But today, every state except Hawaii and D.C. has enacted such laws. The goal with these laws is not to completely eliminate traffic accidents, but to address specific situations where safety problems have arisen.

In the same way, the goal of legislation like the PROTECT IP Act and S.978 is not to completely eradicate online piracy, or allow copyright owners to “go back to the way things were.” Piracy is part of the copyright landscape, and it will always exist in some form or another.

The goal is rather to allow creators and legitimate intermediaries to develop sustainable business models that allow both widespread dissemination of content and the ability to be remunerated for those who invest their time and money to create. Obviously, one of the big challenges facing creators is figuring out these business models, but that doesn’t mean the law shouldn’t also play a role. The alternative would be like using a static web site in a Web 2.0 world.

Fears of Felony Streaming Bill Overblown

By , June 07, 2011.

The Senate Judiciary Committee is set to consider legislation that would make unauthorized public performance of a copyrighted work a felony.

Currently, only unauthorized reproduction and distribution can incur felony charges under criminal copyright infringement provisions — violation of other exclusive rights, like public performance, is considered a misdemeanor.

Bill S. 978 (referred by some as the “Commercial Felony Streaming Act” or “Commercial Felony and Streaming Act”), introduced May 12, 2011, provides for a maximum 5 year prison sentence if “the offense consists of 10 or more public performances by electronic means, during any 180-day period, of 1 or more copyrighted works; and the total retail value of the performances, or the total economic value of such public performances to the infringer or to the copyright owner, would exceed $2,500; or the total fair market value of licenses to offer performances of those works would exceed $5,000.”

The MPAA explains the reasoning behind the proposed legislation:

While existing law provides felony penalties for willful copyright infringement, such penalties only apply to defendants who illegally reproduce or distribute the copyrighted work. Due to technological advances since enactment of these penalties, copyrighted content can now be illegally streamed, not just downloaded, online. However, it is unclear whether Internet streaming constitutes distribution of copyrighted works, and therefore eligible to be prosecuted as a felony.

Internet streaming of copyrighted works clearly implicates the public performance right, and thus can be subject to criminal liability currently if done willfully and “for purposes of commercial advantage or private financial gain.” Due to the lack clarity surrounding the felony offense, however, prosecutors are reluctant to pursue cases against even the most egregious, illegal Internet streaming services. Indeed, prosecutors tend to focus their limited resources and manpower on prosecuting felony offenses that might result in significant jail time because such prosecutions have the greatest deterrent effect.

Newly appointed Register of Copyrights Maria Pallante testified to the Judiciary Committee in support of the legislative goals of the Act earlier this month. Echoing the MPAA’s explanation for the need of the bill, she added:

One might ask why it is not sufficient to prosecute streaming as a misdemeanor.  The fact is, as a practical matter, prosecutors have little incentive to file charges for a mere misdemeanor. This means that, compared to similar infringing conduct involving the large-scale making or distributing of copies (e.g. DVDs of a movie), streaming is not only a lesser crime on the books, it is a crime that may never be punished at all.  As a matter of policy, the public performance right should enjoy the same measure of protection from criminals as the reproduction and distribution rights; prosecutors should have the option of seeking felony penalties for such activity, when appropriate.

Though the proposed legislation makes only a nominal change in the law and is broadly supported, it has attracted some concern that it could cover a wide range of common activities online — some have even suggested that if the bill passes, the government will start throwing people in jail for embedding infringing YouTube videos.

What counts as a “public performance” online?

The US Copyright Act gives creators of “literary, musical, dramatic, and choreographic works” the exclusive right “to perform the copyrighted work publicly.” 117 USC § 106(4). It also states that “to ‘perform’ a work means to recite, render, play, dance, or act it, either directly or by means of any device or process or, in the case of a motion picture or other audiovisual work, to show its images in any sequence or to make the sounds accompanying it audible.” 217 USC § 101. Finally, the Act states that “to perform or display a work ‘publicly’ means—(1) to perform or display it at a place open to the public or at any place where a substantial number of persons outside of a normal circle of a family and its social acquaintances is gathered; or (2) to transmit or otherwise communicate a performance or display of the work to a place specified by clause (1) or to the public, by means of any device or process, whether the members of the public capable of receiving the performance or display receive it in the same place or in separate places and at the same time or at different times.”

Though the application of the public performance right continues to evolve in the online arena, the law generally distinguishes between downloading and streaming, with downloading constituting a reproduction and streaming constituting a performance. 3See, for example, US v. ASCAP, 485 F.Supp.2d 438 (SDNY 2007).

The Second Circuit has taken a broad view of who can be directly liable for infringement of the public performance right online. The court has adopted the conclusion of the Southern District in New York that “Congress intended the definitions of ‘public’ and ‘performance’ to encompass each step in the process by which a protected work wends its way to its audience.” 4David v. Showtime/The Movie Channel, Inc., 697 F. Supp. 752, 759 (1988), quoted in NFL v. PrimeTime 24 Joint Venture, 211 F.3d 10, 13 (2nd Cir. 2000). In other words, the fact that one service provider has permission to publicly perform a certain work doesn’t by itself give users or services permission to “re-publicly perform” that work — the same is true in the offline world, which is why public establishments like bars and restaurants need a license to play music for their customers even when that music comes from a radio playing music from stations that themselves are licensed to publicly perform works. 5See Broadcast Music, Inc. v. Claire’s Boutiques, 949 F.2d 1482 (7th Circuit 1991). Other circuits seem to embrace this interpretation as well. 6For example, the 7th Circuit cites to David in Video Views v. Studio 21, 925 F.2d 1010, 1020 (1991) to support its statement that “the Copyright Act contemplates a broad interpretation of the concept of ‘public performance.'”

Perhaps the broadest interpretation of what is a public performance was made by a district court in Texas in Live Nation Motor Sports v. Davis. 7No. 3:06-CV-276-L. (ND Texas 2007). There, the court found a defendant liable for copyright infringement for merely linking to (rather than embedding) the plaintiff’s webcast without permission. Though this case rightfully raised concerns, its interpretation doesn’t seem to have been adopted by any other court.

Some have argued that the Ninth Circuit uses a different test for determining liability for direct infringement of the public performance right, based on that court’s decision in Perfect 10 v. Amazon. 8For example, Evan Brown, Enthusiast website owner enjoined from streaming webcasts of racing events, Internet Cases blog (Dec 22, 2006); Jason Lunardi, Guerrilla Video: Potential Copyright Liability for Websites that Index Links to Unauthorized Streaming Content, 19 Fordham Intellectual Property Media and Entertainment Law Journal 1077, 1113-14 (2009). The court there used what it called the “server test” — direct infringement of a public display right only occurs when the content is copied on the service provider’s own server — not, for example, when it is displayed through inline linking or framing (though such a server provider may still be indirectly liable).

But I think the Ninth’s holding in Perfect 10 is only limited to the public display right and doesn’t extend to the public performance right. The Copyright Act defines the display right as showing a “copy” of a work, and the court based its conclusion largely on this inclusion of the word “copy” in the definition. 917 USC § 101. It said, “Instead of communicating a copy of the image, Google provides HTML instructions that direct a user’s browser to a website publisher’s computer that stores the full-size photographic image. Providing these HTML instructions is not equivalent to showing a copy.” 10508 F. 3d 1146, 1160 (2007).

Unlike the definition of “display”, the definition of “performance” in the Copyright Act doesn’t require a “copy”. Thus, the question of where the actual content is hosted is irrelevant to whether a public performance has occurred. As far as I can tell, no court has applied the “server test” in the public performance context, so I think it’s reasonable to say there’s an open question about whether the Ninth Circuit takes a different approach than the Second in determining who is liable for violating the performance right online.

Will S.978 Put You in Jail for Embedding Infringing Videos?

Despite the breadth of the public performance right, civil lawsuits against individuals alleged to have infringed it online are rare — Live Nation Motor Sports is the exception rather than the rule. The worry that S.978 will lead to prisons overflowing with people for sharing online videos that happen to be infringing is overblown.

The standard for establishing criminal copyright liability is much higher than civil liability. Prosecutions for criminal copyright infringement under existing law are rare. According to the Administrative Office of the US Courts, less than 50 people are charged with a criminal copyright offense every year. 11Caseload Statistics 2010. There’s no reason to think that this number will change drastically because of S.978.

Other factors support the idea that most internet users have no reason to worry about this bill. The Department of Justice’s Prosecuting IP Crimes Manual lists several considerations for US Attorneys to keep in mind when deciding whether to bring charges. Among the considerations specific to IP crimes:

  • Federal criminal prosecution is most appropriate in the most egregious cases.
  • Limited federal resources should not be diverted to prosecute an inconsequential case or a case in which the violation is only technical.
  • Federal prosecution is most appropriate when the questions of intellectual property law are most settled. Victims have a broad range of civil remedies that include restitution, damages, punitive or quasi-punitive damages, injunctions, court costs, and attorneys’ fees.
  • The sources or manufacturers of infringing goods and services are generally more worthy of prosecution than distributors.

There are many other considerations that limit the application of criminal copyright infringement laws to only the most egregious pirates. No one need worry about facing jail time for sharing videos online should S.978 pass.

References

References
↑1 17 USC § 106(4).
↑2 17 USC § 101.
↑3 See, for example, US v. ASCAP, 485 F.Supp.2d 438 (SDNY 2007).
↑4 David v. Showtime/The Movie Channel, Inc., 697 F. Supp. 752, 759 (1988), quoted in NFL v. PrimeTime 24 Joint Venture, 211 F.3d 10, 13 (2nd Cir. 2000).
↑5 See Broadcast Music, Inc. v. Claire’s Boutiques, 949 F.2d 1482 (7th Circuit 1991).
↑6 For example, the 7th Circuit cites to David in Video Views v. Studio 21, 925 F.2d 1010, 1020 (1991) to support its statement that “the Copyright Act contemplates a broad interpretation of the concept of ‘public performance.'”
↑7 No. 3:06-CV-276-L. (ND Texas 2007).
↑8 For example, Evan Brown, Enthusiast website owner enjoined from streaming webcasts of racing events, Internet Cases blog (Dec 22, 2006); Jason Lunardi, Guerrilla Video: Potential Copyright Liability for Websites that Index Links to Unauthorized Streaming Content, 19 Fordham Intellectual Property Media and Entertainment Law Journal 1077, 1113-14 (2009).
↑9 17 USC § 101.
↑10 508 F. 3d 1146, 1160 (2007).
↑11 Caseload Statistics 2010.

COICA: First Amendment

By , November 29, 2010.

Last week, I’ve been taking a close look at the proposed Combating Online Infringement and Counterfeiting Act (COICA), which was unanimously approved by the Senate Judiciary Committee on November 18th. On Monday, I examined what the bill does, and on Wednesday I addressed due process criticisms.

Today I want to take a look at the most serious criticism: that COICA amounts to “censorship.” It’s a little tough, actually, to find discussion of COICA online that doesn’t mention censorship. Much of the criticism centers around the characterization of the COICA proceedings as a “prior restraint.” Others have expressed concerns that the bill would be “overbroad.”

Prior Restraint

The First Amendment says “Congress shall make no law … abridging the freedom of speech.” The Supreme Court has held that:

[P]rior restraints on speech and publication are the most serious and the least tolerable infringement on First Amendment rights. A criminal penalty or a judgment in a defamation case is subject to the whole panoply of protections afforded by deferring the impact of the judgment until all avenues of appellate review have been exhausted. Only after judgment has become final, correct or otherwise, does the law’s sanction become fully operative. A prior restraint, by contrast and by definition, has an immediate and irreversible sanction. If it can be said that a threat of criminal or civil sanctions after publication “chills” speech, prior restraint “freezes” it at least for the time. 1Nebraska Press Assn. v. Stuart, 427 US 539, 559 (1976).

The general rule regarding prior restraints, as restated by Mark Lemley and Eugene Volokh, is that “Speech may not be restrained until a final judicial determination that it’s unprotected by the First Amendment.” 2Freedom of Speech and Injunctions in Intellectual Property Cases, 48 Duke Law Journal 147 (1998). In practice, this means that courts are reluctant to grant preliminary injunctions in cases involving speech (permanent injunctions, however, are fine, since they come after a final judicial determination that the speech at issue is unprotected). But when it comes to copyright infringement cases, as Lemley and Volokh point out,

[P]reliminary injunctions are granted pretty much as a matter of course, even when the defendant has engaged in creative adaptation, not just literal copying. How can this be? True, the Supreme Court has held that copyright law is a constitutionally permissible speech restriction; though copyright law restricts what we can write or record or perform, the First Amendment doesn’t protect copyright-infringing speech against such a restraint.

COICA provides that a domain name may only be blocked after an in rem proceeding is commenced and a court issues an injunction — preliminary or otherwise. As such, under current law, COICA’s provisions are no more a prior restraint than the preliminary injunctions routinely ordered in infringement cases. Now, one may argue that preliminary injunctions shouldn’t be available in copyright infringement cases as a whole, that despite case law to the contrary, they amount to constitutionally prohibited prior restraints. This is the argument made by Lemley and Volokh. However, even accepting such an argument, there is one exception: “piracy and other ‘easy’ cases.” 3This is one example of why I think it’s important to draw distinctions between “creative” and “consumptive” uses of copyrighted works. Lemley and Volokh explain that “most obvious cases of copyright infringement won’t implicate the prior restraint rule at all”:

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. Harper & Row, as we read it, divides uses of another’s work into two areas: on the one hand, use of another’s idea or facts, or fair use of another’s expression, which is constitutionally protected; on the other hand, use of another’s expression that’s not a fair use, which falls within the “copyright exception” to the First Amendment. So long as Congress and the courts restrict speech that’s within this copyright exception, there’s no First Amendment problem. And while the First Amendment imposes certain limits on copyright law — the requirement of copying, the idea-expression dichotomy, and some aspects of the fair use doctrine — it does not constitutionalize every nuance of the Copyright Act.

COICA’s provisions are targeted at these obvious cases of copyright infringement. The language of the bill reflects this: blocking only applies to sites “dedicated to infringing activities,” which include sites “primarily designed” to offer infringing works “in complete or substantially complete form” and only when “such activities are the central activities” of the site (emphasis added).

In Center for Democracy & Technology v. Pappert (CDT v. Pappert), a District Court case cited by several law professors in their letter in opposition to COICA, a Pennsylvania state law allowing websites to be blocked for containing child pornography was struck down as unconstitutional because it amounted to a prior restraint. 4F. Supp. 2d 606 (ED Pa, 2004). The case has been used because of the similarities between the PA law and COICA in blocking internet sites based on content. However, the similarities end there. The Pennsylvania law provided substantially different procedures than COICA does, and it was specifically those procedures that led to the court’s conclusion.

As the court in CDT v. Pappert explains:

The Act and Informal Notice process are not prior restraints in the traditional sense. They do not prevent speech from reaching the market place but remove material already available on the Internet from circulation. However, they are administrative prior restraints as that term has been interpreted by the Supreme Court. According to the Court, “only a judicial determination in an adversary proceeding ensures the necessary sensitivity to freedom of expression, only a procedure requiring a judicial determination suffices to impose a valid final restraint.” Thus, if material protected by the First Amendment is removed from circulation without these procedural protections, the seizure is invalid as a prior restraint. The Court used the term to describe a Rhode Island Commission’s practice of sending letters to book distributors that asked the distributors to remove books from circulation in Bantam Books v. Sullivan, 372 U.S. 58 (1963) and a procedure that allowed courts to order pre-trial seizure of films alleged to be obscene in Fort Wayne Books, Inc. v. Indiana, 489 U.S. 46, 51-52 (1989).

Based on the decision in Bantam Books and Fort Wayne Books, this Court concludes the procedural protections provided by the Act are inadequate. These cases require a court to make a final determination that material is child pornography after an adversary hearing before the material is completely removed from circulation. Under the Act, a judge is only required to make a finding of probable cause, he can make this determination ex parte, and there is no requirement that the publisher or distributor receive notice or an opportunity to be heard. [Citations removed] 5CDT at 656-57.

Unlike the Pennsylvania law, COICA requires the government to prove by a preponderance of the evidence that a particular site is dedicated to infringing activities, an in rem proceeding is not an ex parte proceeding, and the registrant of the domain name does receive notice and an opportunity to be heard. 6See my previous post, COICA: Due Process. Thus, any comparisons between COICA and the law in CDT v. Pappert are inapposite and superficial at best.

Overbreadth

Besides the claim that COICA would act as a prior restraint, critics contend that the bill also violates the First Amendment because it would “suppress vast amounts of protected speech containing no infringing content whatsoever.” 7Law Professors’ Letter in Opposition to S. 3804 (Combating Online Infringements and Counterfeits Act). In other words, even though copyright infringement isn’t protected under the First Amendment, other speech that is protected will be affected when the domain name for the entire site is blocked.

Any law aimed at nonprotected speech, no matter how well written, can potentially reach protected speech. To reflect this reality without leaving lawmakers powerless to address harms of nonprotected speech or conduct, the Supreme Court has developed the “substantial overbreadth” doctrine: “particularly where conduct and not merely speech is involved, we believe that the overbreadth of a statute must not only be real, but substantial as well, judged in relation to the statute’s plainly legitimate sweep” — if it is, the law is void on its face. 8Broadrick v. Oklahoma, 413 US 601, 615 (1973).

A law that is not substantially overbroad on its face may nevertheless be found unconstitutionally overbroad if a substantial amount of protected speech is affected by the law as it is applied. Laws affecting speech must be narrowly tailored so that even if they are not substantially overbroad on their face, a defendant engaged in protected speech does not find himself within the laws scope. 9For simplicity’s sake I’m purposely avoiding discussion of the different levels of scrutiny that are involved in First Amendment analysis by courts.

A court may find a law not narrowly tailored if less restrictive means exist to accomplish the legitimate sweep of the law. 10See, for example, United States v. Playboy Entertainment Group, Inc., 529 US 803, 813 (2000). In CDT v. Pappert the District Court took the less restrictive means test to mean that “if a statute regulating speech provides distributors of speech with alternatives for compliance and the majority of distributors reasonably choose an alternative that has the effect of burdening protected speech, the statute is subject to scrutiny as a burden on speech.” 11CDT at 652.

Courts are reluctant to declare laws substantially overbroad on their face. “[The overbreadth doctrine] has been employed by the Court sparingly and only as a last resort. Facial overbreadth has not been invoked when a limiting construction has been or could be placed on the challenged statute.” 12Broadrick v. Oklahoma, 413 US 601, 613 (1973). The court in CDT v. Pappert, for example, held that the law allowing a website to be blocked for providing access to child pornography was not overbroad on its face; it did not prohibit nor chill protected speech. 13CDT at 648-49.

COICA is not substantially overbroad on its face as well. It targets only websites dedicated to infringing activities — conduct that is not protected by the First Amendment. Like the statute in CDT v. Pappert, it does not prohibit nor chill protected speech.

It is doubtful that COICA would also be overbroad as applied. The language of the bill — specifically the requirement that a website’s infringing activities must amount to the “central activities” of the site — limits the breadth of COICA to unprotected speech by definition. No doubt some protected speech would end up being blocked, but the First Amendment is not a shield for copyright infringement, and courts have little sympathy for defendants in situations like this who create an “all-or-nothing” predicament like this. 14See, for example, Dr. Seuss Enterprises, LP v. Penguin Books, 109 F. 3d 1394, 1406 (9th Cir, 1997).

That’s not to say that COICA would result in no cases where a substantial amount of protected speech may be blocked. No bill operates outside the constraints of the Constitution; the First Amendment is available to everyone no matter what law is at issue. It’s possible that a specific case could arise where a specific defendant’s freedom of speech is infringed by COICA. That doesn’t mean, however, that COICA itself is unconstitutional, and it certainly doesn’t mean COICA is a vehicle of censorship. 15See, for example, Virginia v. Hicks, 539 US 113, 123-24 (2003).

References

References
↑1 Nebraska Press Assn. v. Stuart, 427 US 539, 559 (1976).
↑2 Freedom of Speech and Injunctions in Intellectual Property Cases, 48 Duke Law Journal 147 (1998).
↑3 This is one example of why I think it’s important to draw distinctions between “creative” and “consumptive” uses of copyrighted works.
↑4 F. Supp. 2d 606 (ED Pa, 2004).
↑5 CDT at 656-57.
↑6 See my previous post, COICA: Due Process.
↑7 Law Professors’ Letter in Opposition to S. 3804 (Combating Online Infringements and Counterfeits Act).
↑8 Broadrick v. Oklahoma, 413 US 601, 615 (1973).
↑9 For simplicity’s sake I’m purposely avoiding discussion of the different levels of scrutiny that are involved in First Amendment analysis by courts.
↑10 See, for example, United States v. Playboy Entertainment Group, Inc., 529 US 803, 813 (2000).
↑11 CDT at 652.
↑12 Broadrick v. Oklahoma, 413 US 601, 613 (1973).
↑13 CDT at 648-49.
↑14 See, for example, Dr. Seuss Enterprises, LP v. Penguin Books, 109 F. 3d 1394, 1406 (9th Cir, 1997).
↑15 See, for example, Virginia v. Hicks, 539 US 113, 123-24 (2003).

COICA: Due Process

By , November 24, 2010.

On Monday, I discussed the Combating Online Infringement and Counterfeits Act, which had recently been unanimously approved by the Senate Judiciary Committee. Among the typical litany of complaints, critics claim that the bill lacks due process.

  • Public Knowledge has posted a letter from NetCoalition opposing COICA, raising several concerns revolving around due process.
  • The Competitive Enterprise Institute warns that the bill lacks “crucial safeguards to protect against the unwarranted suspension of Internet domain names.”
  • Wired magazine says, “In short, COICA would allow the federal government to censor the internet without due process.”
  • Legal scholar David Post writes at Volokh Conspiracy, “It would allow a court to effectively shut down a site operated out of Brazil, or France, without any adversary hearing (unless, I suppose, “the domain name” itself comes into court to argue the case) or any reasoned determination that the site actually is engaged in unlawful activity.”
  • Even notable IP blog IPKat has stated, “There is no provision in the Bill requiring a hearing, trial or defence from the party served with one of these orders.”

Does COICA lack due process? To answer this question, let’s take a closer look at what we mean by “due process” and then see if the bill lacks it.

Due Process of Law

Due process is an ancient and fundamental concept in the rule of law. In his book Due Process of Law Under the Federal Constitution (1906), Lucius Polk McGehee describes it like this:

DUE Process Of Law, as the meaning of the words has been developed in American decisions, implies the administration of equal laws according to established rules, not violative of the fundamental principles of private right, by a competent tribunal having jurisdiction of the case and proceeding upon notice and hearing. The phrase is and has long been exactly equivalent to and convertible with the older expression “the law of the land.”

“Notice” and a “hearing” are important components of due process of law. COICA provides for both of these. Let’s look at the hearing first.

Hearing

In order to block the domain name of a site “dedicated to infringing activities,” the Attorney General must commence an in rem proceeding in a federal district court. As I said on Monday, an in rem proceeding is a full-blown lawsuit, except the defendant is property rather than a person.

That means the Attorney General has to present evidence to a judge and establish — by a preponderance of the evidence 118 USC §983(c). — that the particular site is “dedicated to infringing activities” as defined in the bill. This is how any lawsuit works. The idea that COICA provides some kind of automatic process where the AG can blacklist any site at will, as some critics seem to suggest, is entirely off-base. Courts are not in the business of “rubber-stamping” legal claims, even those made by the government.

But if the defendant in an in rem proceeding is property, do we truly have an adversarial hearing? Contrary to David Post’s odd statement that the property itself would have to show up for an adversarial hearing, it is the owners of the property in an in rem proceeding who intervene in court to argue the case. Generally, anyone with a colorable ownership or possessory interest in the property can appear in an in rem proceeding. 2US v. 8 Gilcrease Lane, Quincy Florida 32351, 641 F. Supp. 2d 1, 4-6 (DC Dist, 2009).

COICA also provides that:

On application of the Attorney General following the commencement of an action pursuant to subsection (c), the court may issue a temporary restraining order, a preliminary injunction, or an injunction against the domain name used by an Internet site dedicated to infringing activities to cease and desist from undertaking any further activity in violation of this section, in accordance with rule 65 of the Federal Rules of Civil Procedure.

Again, the court would issue any of these orders the same way it would issue them in any other case — no due process infirmaries present. But some critics have suggested that the legal obligations these orders trigger cause due process concerns.

Once an injunction is issued under this provision of COICA for a non-US domain, the Attorney General can serve the order on internet service providers, financial transaction providers, and online advertising providers. Those providers are than required to block the domain or prevent it from using their services.

These requirements are very limited. Service providers need only take “technically feasible and reasonable steps” to comply with the order and are not required “(aa) to modify its network or other facilities to comply with such order; (bb) to take any steps with respect to domain name lookups not performed by its own domain name system server; or (cc) to continue to prevent access to a domain name to which access has been effectively disabled by other means.” The other providers are also only required to take “reasonable” measures. Any entity receiving one of these court orders is granted legal immunity from any cause of action arising out of the reasonable measures taken to comply with the order.

In addition, the only means of enforcing these court orders is for the Attorney General to file for injunctive relief in court. Again, the party is afforded all due process of the law as anyone would have when sued for injunctive relief. COICA also provides for an additional defense for such parties: “A showing by the defending party in such action that it does not have the technical means to comply with this section, or that the order is inconsistent with this section, shall serve as a defense to such action to the extent of the inability to comply or such inconsistency.”

Finally, the bill provides that any owner or operator of the domain name, as well as any entity receiving a court order under COICA, may petition the court for modification of the order. The court may modify, suspend, or vacate the order based on evidence that “(A) the Internet site associated with the domain name subject to the order is no longer, or never was, dedicated to infringing activities; or (B) the interests of justice require that the order be modified, suspended, or vacated.”

Given all of the above, it’s difficult to see how the charge that any party affected by COICA won’t get the chance to be heard can hold up.

Notice

Due process requires the government to provide notice, or notification, to anyone whose life, liberty, or property is subject to deprivation in a legal proceeding. The notice must be sufficient; “reasonably calculated, under all the circumstances, to apprise interested parties of the pendency of the action and afford them an opportunity to present their objections.” 3Mennonite Bd. of Missions v. Adams, 462 US 791, 795 (1983).

COICA requires the Attorney General to both send “a notice of the alleged violation and intent to proceed under this subsection to the registrant of the domain name at the postal and e-mail address provided by the registrant to the registrar, if available” and publish “notice of the action as the court may direct promptly after filing the action.” These notice provisions are sufficient under existing law.

Some critics have raised the concern of a website whose domain name registrant is different from the owner or operator, leading to the problem that the owner of a website wouldn’t receive notice of a COICA action. It is difficult to see how this is relevant for due process. COICA provisions are limited to an in rem proceeding against the domain name itself, not the website. The universe of people aside from the registrant who can show a colorable ownership or possessory interest in the domain name to intervene in the in rem action would likely be very small.

Conclusion

I believe COICA satisfies due process under existing law. But you don’t have to take my word for it.

The Anticybersquatting Consumer Protection Act, passed in 1999, provides for in rem actions against domain names. That’s over ten years where courts have heard due process challenges, many of which are similar to the ones being raised concerning COICA. 4See, for example, Porsche Cars North America, Inc. v. Porsche. Net, 302 F. 3d 248 (4th Circ., 2002); Caesars World, Inc. v. Caesars-Palace. com, 112 F. Supp. 2d 502 (ED Va. 2000). The provisions in COICA reflect that experience and are written in a way to ensure the due process of law.

References

References
↑1 18 USC §983(c).
↑2 US v. 8 Gilcrease Lane, Quincy Florida 32351, 641 F. Supp. 2d 1, 4-6 (DC Dist, 2009).
↑3 Mennonite Bd. of Missions v. Adams, 462 US 791, 795 (1983).
↑4 See, for example, Porsche Cars North America, Inc. v. Porsche. Net, 302 F. 3d 248 (4th Circ., 2002); Caesars World, Inc. v. Caesars-Palace. com, 112 F. Supp. 2d 502 (ED Va. 2000).

Combating Online Infringement

By , November 22, 2010.

This past week, the Senate Judiciary Committee unanimously approved the Combating Online Infringement and Counterfeits Act (COICA, S.3804). The bill is targeted toward websites whose central activity is dedicated to infringement. It allows for seizure of the domain names of such sites when they are located in the US, and it requires service providers, advertising providers, and financial transaction providers to take reasonable measures to prevent their services from being used by such sites located outside the US.

COICA was introduced by Senator Patrick Leahy and cosponsored by seven Republican and eleven Democratic senators. Along with the unanimous approval by the Judiciary Committee, the bill has wide support among labor unions, business and trade organizations, and companies.

But as with any effort to promote the progress of the useful arts and sciences, the Judiciary Committee’s approval has set off a fresh round of cries that the “sky is falling” on free speech, internet freedom, and due process. Those who believe that copyright should benefit everyone except those who actually create the stuff that enriches our lives are ramping up their rhetoric of “blacklists” and “censorship.”

I wrote previously about COICA. Since it is moving closer to becoming law, I thought it merited some more attention to clear up some of the misunderstandings surrounding the bill. Specifically, I wanted to take a closer look at why COICA is needed and what it does.

Why COICA is Needed

The last bit of US legislation to substantively address online infringement was the Digital Millennium Copyright Act (DMCA), which was passed twelve years ago. The web was still in its infancy — Google was a month old when the bill passed. Since then, content industries have struggled to find ways to reduce internet piracy. The recording industry at first targeted intermediaries such as Napster and Grokster, moving next to its much-publicized lawsuits against individual filesharers, a campaign that ended nearly two years ago. More effective methods are needed.

Comic book artist and writer Colleen Doran provides a first-hand account of the effects online piracy has had on small creators and why legislation like COICA is needed. In The ‘Real’ Victims of Online Piracy, published in The Hill’s Congress Blog last Wednesday, she writes:

I spent the last two years working on a graphic novel called Gone to Amerikay, written by Derek McCulloch for DC Comics/Vertigo. It will have taken me 3,000 hours to draw it and months of research. Others have contributed long hours, hard work and creativity to this process. But due to shrinking financing caused by falling sales in the division, these people are no longer employed.

The minute this book is available, someone will take one copy and within 24 hours, that book will be available for free to anyone around the world who wants to read it. 3,000 hours of my life down the rabbit hole, with the frightening possibility that without a solid return on this investment, there will be no more major investments in future work.

[…]

Everyone gets paid – manufacturers of computers, iPads, electricity, bandwidth – everyone except the creators of content.

It costs big bucks to finance these pirate sites. Major advertisers and open source ad providers like Google pay them.

Congress is moving on a bill that promises to cut funding for pirates, and the usual suspects who have become accustomed to getting whatever they want online without paying for it are crying foul.

Both Bob Pisano, president of the MPAA, and the Global Intellectual Property Center have also recently written about the detriment of online piracy to creators and consumers and how COICA helps address the difficulties and challenges of enforcing copyright online.

Others have noted the specific need for encouraging advertising providers to block their services from rogue pirate sites, sites that often rely primarily on ad revenue for their profits. As Patrick Ross points out, advertising providers may be reluctant to block such sites since it would mean giving up market share or subject them to liability. COICA solves these problems by clearly identifying sites with primarily illegal purpose and extending legal immunity to service providers who take reasonable and technically feasible measures to block service to those sites.

The idea for a bill like COICA has been percolating for some time now. The Information Technology & Innovation Foundation identified a need for the types of policies enshrined in COICA in December 2009 in its report Steal These Policies: Strategies for Reducing Digital Piracy. From the executive summary:

Other approaches, however, such as blocking websites, may require governmental approval before industry can act. Toward this end, there is a need for a process by which the federal government, with the help of third parties, identifies websites and organizations around the world that are materially engaged in piracy so that ISPs and search engines can block them, advertising networks and other companies can refuse to place ads with them, and banks and credit card companies can refuse to process payments to them.

What COICA Does

The Combating Online Infringement and Counterfeits Act targets the worst of the worst infringing sites — what have been called “rogue websites.” These sites offer little except access to first-run movies, entire television series, or album after album of music. The bill refers to these sites as “dedicated to infringing activities.” “Infringing activities” includes:

offering or providing access to, without the authorization of the copyright owner or otherwise by operation of law, copies of, or public performance or display of, works protected by title 17, in complete or substantially complete form, by any means, including by means of download, transmission, or otherwise, including the provision of a link or aggregated links to other sites or Internet resources for obtaining such copies for accessing such performance or displays.

A site is “dedicated” to such activities if, “when taken together, such activities are central to the activity of the Internet site or sites accessed through a specific domain name.”

COICA would provide the US Attorney General with the authority to commence in rem proceedings against the domain name of a web site that is dedicated to infringing activities. Most civil litigation involves someone suing someone else — an in personam proceeding. A proceeding in rem, by contrast, is against specific property — often represented in court by the owner of the property as a third-party claimant. Aside from the difference in who’s suing whom, in rem and in personam proceedings are, by and large, the same: they are commenced in the same courts, subject to the same rules of evidence and rules of civil procedure, and end up in trial if not resolved before then.

Along with federal court rules and constitutional limitations that apply to all in rem proceedings, domain name forfeiture proceedings under COICA are governed by Title 18, §983 of the US Code, General Rules for Civil Forfeiture Proceedings. These provide, among other things, that the government has the burden of establishing by a preponderance of the evidence that the property is indeed subject to forfeiture — that is, the Attorney General has to prove to the judge that the web site is “dedicated to infringing activities.” They also explicitly provide for an “innocent owner” defense where the owner of the property had no knowledge that it was used for illegal purposes.

Upon commencement of the in rem proceeding, the court may issue a temporary restraining order, preliminary injunction, or permanent injunction against the domain name. These types of court orders are governed by existing law. To get a preliminary injunction, for example, the Attorney General “must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest.” 1Winter v. Natural Resources Defense Council, Inc., 129 S. Ct. 365, 374 (2008).

Any order granted against a domestic domain is then served on the domain name registrar or registry, which is required to suspend operation and lock the domain name. Orders granted against non-domestic domains are to be served on service providers, financial transaction providers, and online advertising providers. Those providers are then required to take reasonable measures to prevent their own services from being used by the web site associated with the domain name.

In my previous post on COICA, I mentioned some of the advantages such a proceeding has over other methods of enforcing copyright in the online realm. I also mentioned how COICA is not really a significant departure from existing law. It is more an optimization of existing law to address some of the most difficult challenges of online piracy.

Asset forfeiture has played an important role in law enforcement for centuries. Today, in the US, asset forfeiture is typically available either as part of sentencing for a criminal defendant or separately through an administrative or civil in rem proceeding. Civil forfeiture authority, however, is not inherent; it must be provided through legislation. 2LII Backgrounder on Forfeiture.

Forfeiture of infringing materials and property used to commit or facilitate the commission of copyright infringement is already provided for by law. This past summer, US authorities carried out Operation in Our Sites, an initiative that, among other things, resulted in the seizure of seven domain names for websites engaged in online piracy.

COICA is not the first law to expressly allow for civil forfeiture of domain names via in rem proceedings: the Anticybersquatting Consumer Protection Act (ACPA, 1999) provides for them as well. Provisions in COICA involving situs and venue mirror those found in ACPA.

Financial service providers and online advertising providers who provide services to sites dedicated to infringing activities already exists in the law. Last month, luxury goods manufacturer Gucci got a permanent injunction against two financial service providers that provided services to websites selling counterfeit goods. Not long after that, on October 29th, Warner Bros. and Disney won damages and an injunction against Triton Media, a company responsible for placing advertising on sites dedicated to copyright infringement.

Essentially, COICA sets forth an expressly defined procedure for reducing online piracy through civil forfeiture of domain names, as opposed to an ad hoc approach that could be cobbled together under existing law. In addition, it establishes clear safeguards to ensure the process is effective and legally sufficient. It gives clear guidelines to third-party operators like financial transaction providers and advertising providers so that they are assured of knowing what is expected of them as legitimate businesses while also providing immunity from liability resulting in technically feasible and reasonable measures to block sites operating outside of the law.

Later this week, I’ll take a closer look at some of the criticisms and concerns that have been raised over COICA.

References

Drawing Distinctions

By , November 11, 2010.

Earlier this month, I wrote about the Copyright Principles Project, an independent initiative that presented twenty-five recommendations for reforming copyright law for the digital age. One of the reasons I felt it was worth mentioning was because, unlike many other reform efforts, the Copyright Principles Project included participants from a wide variety of perspectives: academics, content industry representatives, and private practitioners.

One of the contributers to the Project was Jeremy Williams, Senior Vice President and Deputy General Counsel of Warner Bros. Entertainment. This past April, Williams joined Pamela Samuelson — the Project’s coordinator — and several other Project contributors for a two-day conference marking the 300th anniversary of the Statute of Anne, the first modern copyright law. Hosted by the Berkeley Center for Law & Technology, the Copyright @ 300 symposium used the anniversary as an opportunity to look back at “the law’s influence on the history and evolution of the Anglo-American copyright tradition” and look forward to copyright law’s challenges in the future.

Williams spoke to attendees at a seminar titled “Looking Forward: What Challenges Does Copyright Face in the Twenty-First Century.” The Berkeley Center has audio from the event available online. I recommend giving it a listen; Williams talks about the importance of copyright attitudes for the future and touches on a number of themes I’ve discussed here from a perspective that is seldom seen in online discussions and academic literature.

Copyright attitudes

Williams joshingly introduces himself as a “defender of those poor, down-trodden Fortune 500 companies.” The health of copyright in the future, he states, depends entirely on the attitudes of “Digital Natives.” It will not be long before the generation that never knew a world where everything is connected will be “lawmakers, norm-makers, and consumers.” Williams goes on to ask, “How do we promote respect for good copyright attitudes” when millions are simultaneously authors, distributors, and infringers.

This past Monday, I pointed out a recent paper that sheds some light on what these “Digital Natives” currently know and think about copyright law. Williams’s talk reflects a recognition of what the authors of that paper have found to be true. With those current attitudes in mind, he lays out what he considers should be the two main goals of copyright policy as it looks ahead:

  1. Empower individuals as creators. Policymakers and stakeholders should try to steer clear of talk about the “wrongness” of noncommercial copying. Extending this goal to small producers, content creators should strive to make licensing of their content for creative uses cheap and easy, and give wide berth for fair use.
  2. Move enforcement away from individuals and toward digital intermediaries.

Creative vs. Consumptive Uses

Next, Williams discussed the importance of distinguishing between creative and consumptive uses, a distinction I have previously made on this site. I wrote that the distinction is important to make because you’re dealing with an entirely different set of questions when you talk about fair use, remixes, and mashups then when you talk about filesharing. As it turns out, this is exactly the approach that Williams suggests is important for the future of copyright.

He says that promoting healthy copyright discussion requires a different approach for both types of uses: increased tolerance toward creative uses and increased focus on widespread, uncompensated dissemination of consumptive uses and the harm they cause.

From his own experience at Warner Bros., Williams identifies the primary creative uses that major creative companies deal with most often: fan uses and, as he puts it, ad hoc uses, like mash-ups. He suggests that companies like his should tolerate and support noncommercial uses — even tolerate some commercial uses. This is the approach he has taken, giving the okay to fan sites even when they use incidental advertising to support the site but drawing the line at fan merchandising. For third-party user-generated content sites like YouTube, he advocates “smart” technological filters that can identify uses of copyrighted works but allow creative uses like mash-ups.

Most importantly, Williams reiterates that these kinds of discussions should not revolve around copyright law. Rather, the focus should be on fostering mutually respectful behavior between creators and users.

Individual vs. Industrial Responsibility

The rest of Williams’s talk seemed to continue a meta-theme of drawing distinctions to further the health of copyright in the future. The next distinction he discussed was between individual responsibility for upholding copyright protection and “industrial” responsibility.

Basically, Williams says we should move away from placing responsibility on individuals as much as possible. People should be able to enjoy copyrighted works without thinking about copyright law.

The digital intermediaries — user-generated content and social networking sites — should instead be induced to help reduce the harm of widespread, uncompensated dissemination of copyrighted works, and copyright owners should be allowed to share in the value their works provide to these sites. Interestingly, Williams believes we should reduce the present focus on secondary liability doctrines. These doctrines suffer two disadvantages: because they require a showing of direct infringement, there is still a focus on individual behavior, and the factors involved in determining secondary liability — knowledge, inducement, etc. — are difficult to apply, leading to an uncertain and muddy legal landscape.

Williams’s recommendation? “Be direct about it.” Those who can protect copyright, should protect it. Those who can are likely to be the previously-mentioned social networking and user-generated content sites. Williams thinks the best way to implement this is through voluntary agreements and reasonable technological measures, with the law operating in the background as a way of inducing these sites to help solve problems in ways that are not commercially harmful to them.

Protest vs. Obliviousness

While you can find those people who embrace piracy as a “cause”, who believe they are making a “statement” or protesting against somebody or something when they share files without permission, the majority of people, especially digital natives, download simply because they can. Williams relates a conversation he had with his daughter. He asked her what her friends “think” when they choose to download music off p2p or torrent sites; she answered, “they don’t think.”

Unauthorized files are simply “there,” and unless you know otherwise, it is not unreasonable for someone to assume that if you find something online, you can download it. The future of copyright depends on addressing this obliviousness and helping the next generation understand the value copyright gives to creators.

Good Guys vs. Bad Guys

Williams wraps up his remarks by pointing out the one distinction that shouldn’t be made. Both sides in the copyright debates need to move past an “us vs. them” mentality, a “good guy-bad guy” way of thinking. Copyright is good for society and culture. Most people recognize the value of creativity and of artists. The emphasis should move away from the rules of copyright — ie, the notion that copyright exists solely to punish infringers — and toward the role of copyright.

« Older articles
Newer articles »