Tag: Section 512

Activists Promote Revisionist History of DMCA

Rusty boat lying on girders on shore with light cloudy ocean background
Photo by eyevision Pond5

Earlier this month, Rolling Stone published an article by Steve Knopper called Inside YouTube’s War With the Music Industry.  I would characterize the article as more of a glimpse than an inside view; but setting that aside, the article contained a quote about the DMCA that caught my attention.

Knopper focuses on the fact that several major stars like Taylor Swift, Paul McCartney, and Beck recently signed an open letter to Congress demanding revision to Section 512 of the DMCA.  The article is actually too short and broad to really get into the nitty-gritty, but anyone who follows the issue knows that the crux of all artists’ complaints with the DMCA is that it has unintentionally enabled platforms like YouTube to distribute works without a negotiated license and to monetize mass infringement of works without providing rights holders an effective means of control—as the law was meant to do.

Toward the end of the article is a quote from Corynne McSherry, legal director of the Electronic Frontier Foundation.  Keeping in mind that her statement is brief and can be taken out of context, McSherry said, “I don’t think copyright owners appreciate what they got [with the DMCA]. In 1997, if you wanted to get music taken offline, you had to go to court.”

I don’t know if McSherry added any nuance to that statement, but as presented, it connotes a revisionist history that feeds a general misunderstanding of DMCA and how it got here.  Both tech-industry advocates and mainstream reporters are apt to continue the narrative that Section 512 is principally a takedown provision that happens to have a liability shield (safe harbor). In fact, it’s the other way around, and the difference is actually rather important.

Why the DMCA is seen as flawed for rights holders.

Drafting of Title I of the DMCA began in December of 1996 in order to implement the WIPO Copyright Treaty adopted that same month.  The treaty was a special agreement under the Berne Convention to which the US has been a signatory since 1989.  Title I is primarily known for Section 1201, which prohibits circumvention of technical protection measures (TPM) used to protect copyrighted digital goods. In the late 1990s, this would largely have applied to products like DVDs and CDs, but the DMCA was intended to anticipate copyright enforcement on the internet.  (1201 has its own criticisms, as noted in a recent post.)

While Title I was being debated, major ISPs, which at that time were telecoms led by Verizon and AT&T, petitioned Congress for a shield from monetary liability for copyright infringement—the shield we now call the “safe harbor.”  These service providers knew that users would continue to upload infringing content but argued that only the rights holders could know what was infringing and that the rights holders were in the better position to identify infringements than the ISPs were.  These service providers further argued that, if they were constantly vulnerable to litigation for contributory infringement, this would stifle the development of the still-fledgling internet.

It is worth noting that when the ISPs initially argued that the responsibility of “ferreting out” online copyright infringement should fall to the rights holders, Roy Neel of the United States Telephone Association highlighted the availability of “spiders” (now called “bots”) that owners could use to automate search for infringing files.  Today, this same automation, which has grown in scale in an effort to keep up with the scope and frequency of infringement, is regularly characterized by the internet industry and “digital rights” activists as an overreach that cannot help but chill speech online. The history that this automation was a key component of what the ISPs initially lobbied for is rewritten almost daily with statements like the one by McSherry.

In response to the ISPs’ request for the liability shield, major rights holders, including the MPAA and RIAA, negotiated conditions the service providers would have to meet in order maintain the “safe harbor” status—most prominently that the ISPs had to adopt systems for removing infringing material upon request (i.e. takedown). The result of these negotiations are the provisions in Title II of the DMCA, particularly Section 512, which is the part of the law the aforementioned music stars are asking Congress to amend.

So the request for safe harbor came first, and the takedown provisions were one part of the conditional compromise.  Rights holders had not petitioned Congress for a takedown provision in order to avoid the burden of litigation, which is what McSherry’s statement can imply to the casual reader.

Meanwhile, in 1996, we were all still dialing into AOL; Napster was three years away from changing everything; YouTube would not launch until 2005; and users could only rather slowly view a single photograph online, let alone stream HD video. At the time Title II of the DMCA was drafted, neither Congress nor the rights holders—nor perhaps some of the online service providers—quite anticipated the massive growth in online infringement that was to come.  Plus, the parties were—perhaps naively—sanguine about the prospect that new technological protection measures would be adopted through collaboration between the ISPs and the rights holders—which happens to be exactly what Congress had instructed both parties to do. That collaboration never quite happened.

Meanwhile, many rights holders would argue today that subsequent court interpretations of DMCA have undermined the intent of Congress, giving ISPs so much latitude in the manner in which they conformed to its conditions that the law inadvertently creates an incentive to enable user infringement at the volume we see today. As Robert Levine points out in his book Free Ride, the DMCA’s lead architect, former commissioner of the USPTO Bruce Lehman, admitted at a conference in 2007 that the law had been a great disservice to the creative industries.

But are rights holders better off?

Is Corynne McSherry correct to say that copyright owners are fortunate to have the DMCA because its takedown provisions obviate the need for costly litigation?  It’s hard to imagine this is actually the case.  If we’re talking about major rights holders, like big studios or labels, consider the options …

Option 1 – Without DMCA

Several years of expensive litigation against a single provider for multiple infringements in which the rights holder would stand a decent chance of winning the case, where the rights holder may be afforded some injunctive relief, and whereby the case may act as a deterrent to other providers regarding infringement on their platforms.

Option 2 – With DMCA

A round-the-clock, expensive process of requesting individual files be removed one at a time from multiple platforms only to have the same files reappear while some service providers monetize these uses, become too big to sue, and remain free from liability forever.

I’m not in a position to speak for the major rights holders here, but I don’t know if McSherry really wants to ask them if they wouldn’t choose Option 1.  When Viacom sued YouTube in 2007 for a billion dollars’ worth of infringements and the case was finally decided in YouTube’s favor in 2013, it is not unreasonable to imagine a different outcome absent the DMCA safe harbor provision.  (Keep in mind that this particular post isn’t about anyone’s opinion of how things should be, only a consideration of McSherry’s assertion that rights holders are better off under Section 512 of the DMCA.)

For smaller, independent rights holders, they’re kind of hosed either way but maybe a bit better off with Option 1.  Absent the DMCA, they could not afford to sue the likes of Google anyway; but with the DMCA, the takedown provisions are useless as a means of enforcement because of the Whack-a-Mole problem. If, on the other hand, the majors had successfully sued a big provider like YouTube (i.e. had Viacom gone the other way) independents may have benefitted by proxy.

For one thing, if Viacom had won its case, it’s a pretty good bet YouTube would suddenly discover a remarkable capacity for mitigating infringement on its platform.  Secondarily, had Viacom set the opposite precedent, this would have empowered independent rights holders to either litigate or threaten to litigate as a means of enforcing their rights across the web.  So, McSherry’s note that if you wanted to get music offline in 1997, you had to go to court is only partly true and not necessarily the lesser alternative for rights holders, as she implies.

Yes, YouTube itself would have evolved differently—and a few billionaires might only be multi-millionaires today—but do the social benefits of web platforms have to come at the expense of permission and compensation-free exploitation of other people’s work?

Over the weekend, I watched a 1912 film on YouTube called Falling Leaves by Alice Guy Blaché—a file that silent film accompanist and historian Ben Model uploaded along with his own piano composition.  To me, this is a great example of the information-age ideal.  I can research an important work in film history, which has long been in the public domain, and discover a contemporary artist/historian at the same time. Great. Love it.

But was the progress of YouTube—whose growth was substantially subsidized by copyright infringement—the only history that could possibly have resulted in my seeing Guy-Blaché’s film and finding Model’s work in 2016? Not necessarily. So, I don’t know if McSherry really meant to play the alternate histories game here, because it is certainly possible to imagine the benefits of the internet evolving since 1997 without the hundreds of millions of infringements that occur every month. And that was certainly what Congress expected when Title II of the DMCA was being negotiated. So, now that it’s clear the intended consequences did not come to pass, maybe amending the law isn’t such a crazy idea.

To support revision to the DMCA, click the link in the sidebar to sign the Take Down/Stay Down petition.

Astroturf Organizations Typically Hysterical on DMCA

As the deadline approached for public comments to the Copyright Office in anticipation of its review of Section 512 of the DMCA, TorrentFreak reported yesterday morning that 50,000 “citizens” chimed in to protest DMCA “abuse,” apparently enough to “crash” the government’s servers.  Assuming the crash did occur, it’s probably an endorsement for Copyright Office modernization, but to the matter at hand, if there are 50,000 actual, non-attorney citizens who understand DMCA, I’ll eat my hat and the box it came in. This is more SOPA-fying, scare-mongering bullshit, and I really wonder how many times people are going to fall for it.

The TF article quotes this statement by Tiffani Cheng of the Google-funded organization Fight for the Future: “The DMCA affects all Internet users and they should have an opportunity to express their concerns with the ways content is censored from the Internet, causing damage to free speech that can’t be undone.”

To describe DMCA as a tool for censorship is a gross exaggeration that enables major OSPs (Online Service Providers) to use individuals as human shields to cover their profit interest in keeping DMCA ineffective for rights holders. It’s not that DMCA abuse does not occur, but the comparatively few incidents in which an individual or entity purposely misuses takedown should not be allowed to mask the enterprise-scale motives for major OSPs to support, promote, or even condone mass infringement.  That was never the intent of the DMCA.

Millions of copyright stakeholders know first hand that the OSPs have been incentivized by the terms of the DMCA to fabricate an illusion of ignorance with regard to obvious cases of infringement hosted on their platforms, promoted by their search engines, or supported by the access they provide.  Simply put, in order to retain the safe harbor (i.e. neutral) status, which service providers consider essential to their existence, they are supposed to meet certain obligations according to the statutes.  In many cases, large providers either fail to meet these conditions outright (as we saw in Cox v BMG) or they push the boundaries of reason and good faith when it comes to what’s called “red flag” knowledge of infringing or other illegal activity making use of their services.

For instance, among the conditions an OSP must meet to retain safe harbor under DMCA is that it may not benefit financially from infringement.  So, when a user uploads a whole TV episode, let’s say, to YouTube (which nobody disputes is infringing) and YouTube generates ad impressions during the period when the file is online before the rights holder takes it down, that’s revenue.  Why is that transaction not a clear violation of the statutory conditions, which would appear to make YouTube liable for the infringement rather than the neutral party it claims to be?

For obvious reasons, OSPs do not want to change the status quo.  And to be clear, rights holders are not looking to end safe harbor protections or to seek new means of taking down more material that is non-infringing; they have enough challenges just trying to keep a lid on the large volume of undisputedly infringing content.  And make no mistake, the major OSPs could give a damn about your free speech or your remix videos beyond the extent to which defending those things makes them money. (One could make far more compelling arguments that these service providers stifle speech through manipulation of their algorithms than all the DMCA abuse that’s ever been cataloged.) So, with regard to DMCA, these service providers would like to perpetuate the game that earns them revenue and grows their market share without having to bother with the legitimate rights of creators.

Even a Google-funded report released last week on DMCA notice and takedown procedures, conducted by researchers at Berkeley and Columbia Law, indicates that the majority of errors and abuses of the DMCA takedown process occurs among smaller and mid-sized rights holders, OSPs, and plain bad actors.  I may write a more detailed discussion of that fairly large report in a future post. But I mention it here because not even research—at least the anecdotal portion of it—slanted in favor of the internet industry appears to really support the assertion that DMCA takedown is widely abused as a tool to censor your “tweets and videos.” This is a typically hysterical claim that sounds sillier with each passing day that trillions of online expressions are exchanged without incident.  Meanwhile,  the  DMCA remains an inadequate tool for most rights holders of all sizes to mitigate large scale infringement and outright piracy of their works.  And these uses are still not free speech.

Copyright Office to Review Safe Harbor in DMCA

(Republishing as the April 1 deadline for comments to the US Copyright Office approaches.)

Remember Bill Clinton?  If you don’t, he’s that guy who was just in New Hampshire campaigning for his wife Hilary, who’s running for president. Anyway, Bill Clinton was president so damn long ago that when he was first sworn into office, most of us didn’t even have email.  I mean in real history terms, the Clinton administration was just yesterday; but in Internet history terms, it was like forever ago. After all, most of the general public (which is to say folks other than university professors and members of the military-industrial complex) all suddenly jumped online during Bill Clinton’s first term in office. You might even say we became collectively and instantly an America Online.  Ah, the 90s. Good times. Or should I say YAHOO!?  I don’t know. Anyway, the Web looked very different. If you don’t believe me, you can Ask Jeeves.

During President Clinton’s second term, in October of 1998, he signed into law the Digital Millennium Copyright Act, a.k.a. the DMCA, which defines the conditions, responsibilities, and limitations for both ISPs and copyright holders pertaining to various remedies for online copyright infringement.  In order to balance interests and protect free speech online, Section 512 of the DMCA provides the conditions by which ISPs are granted safe harbor, shielding them from liability that might stem from copyright infringements perpetrated by users of their sites or services.  Section 512 also specifies the conditions an ISP must meet in order to maintain safe harbor protection, and among these is the establishment and maintenance of processes for removing infringing material and for canceling the accounts or denying access to repeat infringers.  The former would generally apply to platforms that host User Generated Content (UGC) while the latter would typically apply to an access provider like TimeWarner, Cox, et al.  But as I say, the Web looked very different when the DMCA was written.

As the capabilities and platforms have expanded—for instance from effectively no video in 1998 to YouTube’s stated 300 hours of high-quality video being uploaded to its platform every minute in 2014—it was inevitable that the incidences of repeat infringement and repeat infringers would expand in kind.  And they did. Rights holders presently send out hundreds of millions of DMCA takedown notices each year—just to YouTube alone—and only a small fraction of these (fewer  than 1%*) are either sent in error or are intentionally abusing the takedown system for purposes such as stifling criticism or other forms of non-copyright-related complaints.  Additionally, most wrongful takedowns are remedied by the counter-notice procedure provided for in the DMCA, which restores content that has been removed; and any further action from either party requires legal procedures outside the DMCA, including litigation, which is not all that common relative to the volume of content we’re talking about.

Meanwhile, the safe harbor provisions in Section 512 of DMCA were never intended to provide a blanket shield for ISPs while they profit incidentally from the high volume of infringements committed by users; but that’s more or less what’s been happening.  If at any given moment, there are 100 million infringing videos on YouTube, each generating only one view per video before they are removed by the notice-and takedown-procedure, that’s 100 million ad impressions generated while YouTube complies with the DMCA.  YouTube gets the ad revenue while infringing videos go up and down in a constant ebb and flow, and the safe harbor provisions shield the platform from any liability—or even responsibility—to further mitigate mass infringement.  Safe harbors were not intended to provide an incentive for allowing (if not fostering) mass infringement by users, which is why the Copyright Office announced on December 31 that it will begin a review of Section 512 of the DMCA on the premise that the provisions may well be antiquated relative to the realities of the contemporary Web.  In announcing the review and call for comments, the Copyright Office states:

“While Congress understood that it would be essential to address online infringement as the internet continued to grow, it may have been difficult to anticipate the online world as we now know it, where each day users upload hundreds of millions of photos, videos and other items, and service providers receive over a million notices of alleged infringement. The growth of the internet has highlighted issues concerning section 512 that appear ripe for study.… Among other issues, the Office will consider the costs and burdens of the notice-and-takedown process on large- and small-scale copyright owners, online service providers, and the general public. The Office will also review how successfully section 512 addresses online infringement and protects against improper takedown notices.”

With regard to access providers, the recent BMG v COX case offers insight into the conditions a provider must meet in order to maintain its safe harbor protection, and it also reveals the kind of shenanigans rights holders have been putting up with for years. Cox was successfully sued by BMG for contributory infringement based primarily on the fact that the plaintiff was able to demonstrate very clearly that the defendant did not have a reasonable procedure in place for canceling the accounts of repeat infringers as is mandated by Section 512 of the DMCA.  As I mentioned when I was a guest on the new podcast series Steal This Show, presented by TorrentFreak, pre-trial court documents indicate that Cox had what can best be described as a “13 strikes and you get a really strong warning” procedure for repeat infringers.  And perhaps even more damning, emails entered into evidence suggested that the access provider was effectively resetting the dial on repeat infringers to make them look like first-time offenders.  Naturally, there are more details, but suffice to say, neither the judge nor the jury found these processes to be adequate or reasonable procedures and, thus, Cox’s safe harbor defense was of no avail.

Still, at least some of the reporting and spin on that case will portray its outcome as an “expansion” of DMCA’s power to “stifle,” and it is worth noting that Judge O’Grady rejected amicus briefs from both the EFF and Public Knowledge, stated thus:

ORDER, having reviewed the proposed brief, the Court discerns no potential benefit to be gained from receiving the brief. This is not a situation where defendants lack competent representation. Nor have Public Knowledge and EFF persuaded the Court that they have a sufficiently special interest in the outcome of this litigation to warrant consideration of their viewpoint. Accordingly, the motion 405 for leave to file an amicus curiae brief is DENIED.

Given the amount of virtual ink and PR capital the Internet industry and its network of “activists” have devoted to excoriating the takedown provisions in DMCA, the general public—and even some reporters—can be forgiven for thinking that DMCA is synonymous with chronic and unrestricted removal of content without due process. In fact, it seems that even the reporting that wants to be objective (including the aforementioned TorrentFreak) can fall into the understandable but flawed habit of lumping all things DMCA together as though the law is little more than carte blanche for major rights holders to willfully bulldoze expression from the web.  In reality, of course, the DMCA is a more complex framework of remedies, counter-remedies, and defined boundaries for all parties. In fact, if you go looking for cases of DMCA takedown abuse, you’re far more likely to find smaller rights holders, foreign organizations, or non-copyright-industry parties (e.g. Ashley Madison) committing improper takedowns through DMCA.  Meanwhile, the usual “villains” in the public narrative—the major film studios and music labels, who send out the majority of notices to platforms like YouTube—generally know what they’re doing, which is why the incidence of error in their procedures is extremely low.

But as the Copyright Office proceeds with its review of Section 512 and various interests respond to the call for comments, it’s likely we’re going to see the Internet industry—and the EFF, Public Knowledge, et al—try to portray safe harbor provisions as sacrosanct. (After all, there’s a lot of free money at stake, and who doesn’t like free money?) And these entities can be expected to follow their well-worn playbook of presenting the unalienability of safe harbors as vital to the functioning of the Internet and free speech and the rights of man and the very air that we breathe, and so on.  And to be clear, nobody (me included) would advocate revoking safe harbors. These provisions are an important component of digital-age law and are predicated in part on protecting free speech.  I’m just saying to watch out for the hyperbole when it comes.  The Copyright Office is reviewing Section 512 to potentially recommend tweaking it to fulfill its intent, not to abandon its intent altogether.

In a way, the most frustrating aspect of DMCA for rights holders is that it is reminiscent of another Clinton legacy—that unfortunate absurdist theater called “It depends on what the definition of is is.”  Likewise, the proverbial game of Whack-a-Mole—as everyone describes counter-infringement procedures—is really a game of semantics that ISPs have been playing for more than a decade.  What is an infringement?  Who is a repeat infringer? What is a reasonably implemented policy?  And so on. But then, that’s law—a best attempt to use language to create salubrious policy.  And since Section 512 seems to unintentionally leak mass infringement like a sieve, it’s probably time for a rewrite.

*This estimate is based both on counter-notice data from MPAA takedown requests as well as anecdotal information stated in October 2015 amicus brief seeking new ruling in Rossi v MPAA.