Category: Intellectual Property

Castle: Mass NOI Loophole Still Being Exploited

Attorney and blogger Chris Castle continues his reporting on major players like Amazon “innovating” the hell out of a loophole in Section 115 of the Copyright Act.  The filing of Notices of Intent (NOI) was designed for low-volume use but is not be exploited by Big Tech to avoid, or at least defer, paying royalties to songwriters. “Assuming the filing was made correctly, the user can then allege that the user is entitled to all of the benefits of the compulsory license without the obligation to pay royalties until the song owner catches them,” writes Castle.

See full article at Music Tech Policy here.

Malware Suggests Search Plays a Major Role in Piracy

Image by stefanocar75

Copyright holders have long insisted that search results play a substantial role in driving users toward pirate sites.  Google and piracy advocates have generally countered that search does not drive much traffic to illegal sites because the people who consistently use infringing sites know what they’re doing and will go directly to the content they’re seeking.  This is a reasonable assumption to make about the population of committed infringers out there, but one fact that refutes this premise is the extraordinary volume of malware (a 1-in-3 chance) on infringing sites.  Because malware isn’t there to catch the experienced visitor—it’s there to catch the unsuspecting individuals who may not even realize they’re using illegal sites when they first visit.

For those who don’t know how it works, it goes like this:  A user is interested in watching Moonlight.  If he types “Moonlight” into Google Search, the second-tier results will be links that read “Watch Moonlight Online for Free,” all of which are directed to infringing sites.  If the user actually types “Watch Moonlight” into the search field, then the first-tier results will be infringing links. And quite often, Google will automatically suggest words that prompt the user toward an infringing site. For instance, if the user logically adds the word “movie” (because moonlight is a word and not just a title), then Google will complete the thought with “online,” which then yields top results with links to “watch moonlight movie online” via an infringing site.

Google and the piracy apologists are almost certainly correct that many avid visitors to infringing sites are fairly sophisticated users; they have VPNs, ad-blockers, security software, etc. to avoid detection and malware.  But if these were the only kind of visitors landing on these sites, then the underground market in malware-based trade would not be nearly so robust as it is.

As described in this 2015 post about a report called Digital Bait, commissioned by Digital Citizens Alliance and conducted by RiskIQ, a sophisticated “crimeware economy” exists on the Darknet, where criminals buy and sell goods and services used exclusively for preying on users. To use a blunt example, if a teenage girl visits an infringing site, she has up to a 30% chance of contracting malware. That malware may be a Remote Access Trojan (RAT), which gives fairly unsophisticated hackers control of her computer, including her webcam.  Then, her IP address may be sold in this black market to people who want to spy on teenage girls in their bedrooms. In many cases, a user doesn’t even have to consume the infringing content in order to infect a device. The promise of “free content” may be draw the user into a dead-end malware trap.

If all traffic to pirate sites truly comprised only the knowledgeable users, then the criminals would not have a financial incentive to deploy so much malware on sites that infringe, or promise to infringe, copyrighted content.  The very existence of prevalent malware is an indication that a substantial number of users who have no idea what they’re doing are visiting these sites, which logically leads to the conclusion that search must play a significant role in driving users toward these sites and into the hands of criminals.

Notice that, in this context, we don’t even need to address the subject of copyright infringement, let alone get bogged down in all the tedious rhetoric about free speech.  If Google’s top search results are indeed putting users in harm’s way, this is a consumer protection issue for the Fair Trade Administration and/or State Attorneys General.  And, in fact, Digital Citizens Alliance, after releasing its 2016 report Enabling Malware, began presenting its findings to the AGs.

Yes, it is likely true that once a user—even a fairly unsophisticated teenager—is aware of sites where free content is available, he will probably revisit those sites directly without going through a search engine. But even this kind of anecdotal assumption does not mean the role of search is insignificant, not least because the illegal nature of pirate sites means that they have a tendency to disappear and reappear as authorities in various regions shut them down.  A 2013 study indicated that 19% of the traffic to infringing sites could be directly attributable to search, and if that number were wrong by half, it would still represent billions of visits per year.

Consumers have a right to know the nature of their vulnerabilities when using any product or service, and they have a right to demand that U.S. companies take every reasonable step to mitigate exposure to risk.  To date, Google has refused to take even the obvious step of demoting known infringing sites in their search results, let alone to alter the way in which auto-complete may drive consumers toward these sites.

Google does now feature the legal channels for consuming media, including their own services like YouTube and Google Play, which is a good step but not likely sufficient to protect consumers as hackers become more sophisticated and more ambitious.  In fact, one likely consequence of advertisers becoming more effective at keeping their brands off pirate sites is that the criminals will depend more on the “crimeware economy” to make money through infringing content as a means to deliver malware.

Google is getting a lot of pushback lately—from the EU’s anti-trust decision, from the advertisers, and from the Canadian Supreme Court this week in the Equustek case. (More on that shortly.)  I would not be surprised if the State AGs and other consumer-protection agencies begin to take a more active interest in the relationship between search, piracy, and malware.

Packingham Ruling Likely Not Instructive to DMCA

Photo source by spaxiax

Another Supreme Court First Amendment decision this past Monday was a source of excitement for parties who consistently argue that copyright enforcement in cyberspace cannot help but infringe First Amendment rights.  I’ll say at the outset that I fully agree with the decision in this case but very much doubt any proposal that the opinion in Packingham v. North Carolina contains the DMCA implications certain copyright critics are hoping it does.

The case itself has nothing to do with copyright. In fact, Packingham begins with the highly-sensitive issue of registered sex-offenders and a North Carolina law designed to prohibit anyone who is so registered from using social media sites.  The eight justices (minus Gorsuch) held unanimously that the State law was far too broad in scope to achieve its purpose–that it infringes the speech rights of people who have already served their sentences by unduly denying them access to the contemporary equivalent of a library, public forum, employment resource, etc.

“With one broad stroke, North Carolina bars access to what for many are the principal sources for knowing current events, checking ads for employment, speaking and listening in the modern public square, and otherwise exploring the vast realms of human thought and knowledge. Foreclosing access to social media altogether thus prevents users from engaging in the legitimate exercise of First Amendment rights. Even convicted criminals—and in some instances especially convicted criminals—might receive legitimate benefits from these means for access to the world of ideas, particularly if they seek to reform and to pursue lawful and rewarding lives.”

It is just that kind of language that “digital rights” activists love to hear in general, but more specifically, some of the usual suspects have theorized that this decision may have constitutional implications for DMCA Section 512(i), which stipulates that, to remain shielded from liability for third-party copyright infringement, internet service providers must have policies in place that will ultimately result in account termination for “repeat infringers.”  Critics of online copyright enforcement are hopeful that the Supreme Court’s decision in Packingham establishes the precedent to advocate that restricting web access under any circumstances is always a First Amendment violation, which would gut one of the few remedies in DMCA that may actually be effective.

Reading Between the Rhetoric

The opinion, written by Justice Kennedy, does contain quite a few buoyant pull-quotes that internet advocates are going to save for future cases—if not make into hats, mugs, and bumper stickers.  “While we now may be coming to the realization that the Cyber Age is a revolution of historic proportions, we cannot appreciate yet its full dimensions and vast potential to alter how we think, express ourselves, and define who we want to be,” the opinion says. While there can be little doubt that Packingham affirms that cutting a citizen’s access to the internet begs careful scrutiny in context to the First Amendment, it may not be quite so fatal to DMCA 512(i) as is being suggested.

For all of its praise of the web, the opinion also contains caveats such as, “…this opinion should not be interpreted as barring a State from enacting more specific laws than the one at issue. Specific criminal acts are not protected speech even if speech is the means for their commission.”  It is a matter of settled law that copyright infringement is not protected speech; so qualifiers like this quote would seem to reject the prospect that citing Packingham will be particularly instructive to the constitutionality of Section 512(i).

Further, Justice Alito, joined by Justice Thomas, agrees with the decision, though not with the opinion.  Specifically, Alito sharply criticizes the broad rhetoric employed by his colleagues, concerned that these platitudes may foreclose a State’s right to tailor a far more narrowly-defined law that would seek to keep sexual predators away from access to children via social media. Observing that the internet unquestionably creates new opportunities for committing criminal acts, Alito seeks to temper the slightly techno-geeky ebullience of the opinion with comments like the following:

“The Court is unable to resist musings that seem to equate the entirety of the internet with public streets and parks. And this language is bound to be interpreted by some to mean that the States are largely powerless to restrict even the most dangerous sexual predators from visiting any internet sites, including, for example, teenage dating sites and sites designed to permit minors to discuss personal problems with their peers. I am troubled by the implications of the Court’s unnecessary rhetoric.”

“The fatal problem for §14–202.5 [the North Carolina statute] is that its wide sweep precludes access to a large number of websites that are most unlikely to facilitate the commission of a sex crime against a child.”

So, for all the web-endorsing statements made in the opinion in this case, the copyright enforcement critics may be grasping at straws here for the simple reason that both the intent and application of Section 512(i) is narrowly tailored to restrict access to individuals who are specifically abusing that access to repeatedly commit a specific crime.  (Stay tuned for people to muddle this by contrasting sexual predation with copyright infringement, but that’s not a comparison anyone is making.)

An Analogy Regarding Users

With a little revision, the language employed by the Court to describe the internet might be transposed to describe the automobile at the dawn of its entrance into American society.  The essential right to travel by car to work, to the store, to a doctor, to take a vacation, or just to go anywhere without purpose, is affirmed in the American triad of inalienable natural rights described as the Pursuit of Happiness. It is a declaration that predates the First Amendment by fifteen years and a whole revolution.

Yet, despite all that, the Supreme Court is unlikely ever to rule that a State may not revoke a driver’s license under statutes narrowly tailored to the purpose of maintaining safe and legal transportation on the highways. Similarly, if a user were to lose some form of internet access as a result of ISP compliance with DMCA 512(i), this remedy is narrowly focused on that user’s consistent abuse of the system itself.

Actual Practice 

The main flaw in the view that Packingham implicates anything about the DMCA is that compliance with its conditions is not mandated. The DMCA is law, but its provisions are voluntarily adhered to in order to qualify for the “safe harbor” liability shield for infringements committed by users. As such, provisions like Section 512(i) are broadly written in order to give ISPs discretion to design their own policies, although that discretion—as we see in the Cox case—has been interpreted by some providers to mean that they can avoid enforcement of a repeat infringer policy until they lose the safe harbor.

In reality, a user has to try pretty damn hard to have an account canceled by a service provider for repeat copyright infringement.  At the very least, a user will receive several warnings that his chronic infringement will lead to account termination. In general, though, it is unclear that any major service providers actually enforce repeat infringer policies at all.  This, of course, does not stop the pundits from repeating unsubstantiated claims of rampant “DMCA abuse.”

What Packingham does suggest to me is that the opinion would serve a plaintiff quite well, if he were to sue an ISP or edge provider for wrongful termination. But the potential for “wrongful” termination implies that there can be legitimate reasons for termination, whether these are based on compliance with statutes or they are stated in the company’s own Terms of Service.

Service Providers are Not State Actors

“Even extensive regulation by the government does not transform the actions of the regulated entity into those of the government.” (Jackson v. Metropolitan Edison Co 1974)

As stated repeatedly on this blog, regardless of the many free speech invocations made by the internet industry, the First Amendment only prohibits the government from infringing speech.  Google, Facebook, Amazon, Twitter et al may do pretty much whatever they want, including proscribing any form of content they deem unfit for their platforms; and the First Amendment has nothing to say about it. And, of course, site owners do control content on their platforms as suits their business interests.

The counter-argument being made is that Packingham comes close to asserting that any type of user access is constitutionally protected by the First Amendment and, therefore, a provision like 512(i) must be unconstitutional.  While it is certainly true that SCOTUS articulated the significant relationship between the web and the First Amendment, I believe this opinion stops well short of closing that particular circle of reasoning.  In fact, my legal expert colleagues tell me that this logic doesn’t usually carry much weight in court as it would imply that the moment any of us obeys the law, we become “state actors.”

In the end, the Supreme Court said some nice things about the internet–all true, I think, and all good grist for the PR mills.  But it seems highly doubtful that this opinion says much, if anything, about DMCA or online copyright enforcement.  To the contrary, the more the Court analogizes the web to parks and other physical, public spaces, the more likely it seems they will view cyberspace as a place where the rule of law still applies.