Copyright

When the Copyright Office released its report on Section 512 of the DMCA on May 21, 2020, it largely concluded that the law was not working as intended, and that "Congress' original intended balance has been tilted askew."[ref]Section 512 of Title 17 – A Report of the Register of Copyrights at 1[/ref] This conclusion did not sit well with the websites, ISP's and OSP's. And at 198 pages and 1041 footnotes, it's a lot to unpack. In the intervening month, plenty has been written about the report. But let's dive a little deeper and get to the parts that didn't get into the headlines or the bullet points.

92,000 Public Comments or 86,000 Bot Comments?

Chief amongst the complaints was that somehow the copyright office had not really taken into account "the public interest," in reaching its conclusions. For its part, the Electronic Frontier Foundation complained that "[t]he Copyright Office's report speaks of the views of online service providers and rightsholders, while paying only lip service to the millions of Internet users that don't identify with either group."[ref]Internet Users of All Kinds Should Be Concerned by a New Copyright Office Report[/ref] Well, not quite. The Copyright Office was nothing but thorough. It first sent out a general Notice of Inquiry (NOI) on December 21, 2015, inviting public comments. It received over 92,000 comments.[ref]Report of the Register of Copyrights at 12[/ref] It held two days of public roundtables in two separate cities: May 2^nd and 3^rd of 2016 in New York, and May 12^th and 13^th 2016 in San Francisco. This apparently was not enough. Another NOI was issued on November 8, 2016, seeking additional comments on 16 specific questions, and announced another day of roundtable discussions that were held in Washington, D.C. on April 8, 2019.[ref]Id. at 13[/ref] Your humble correspondent is well aware of all of these events. Not only did I file written comments, but I attended both the New York roundtable and the follow-up roundtable held in Washington, D.C. The Copyright Office was kind enough to sit me right next to the representative from Google. The EFF was also well aware of these events. Corynne McSherry of the EFF was a roundtable participant in Washington, D.C.[ref]Id. at Appendix B[/ref] According to the Report, "the Office received more than 92,000 written responses from a wide range of parties, including rightsholders OSPs, technology companies, libraries, legal scholars, public interest groups, and individual members of the public."[ref]Id. at 12[/ref] But as they say, let's take a "deeper dive" into what these numbers really show. For this, you have to read the footnotes. Around 94% of all the comments received were what the CO calls "template forms," a more polite word for spam comments. These comments basically reproduced "template text provided by a third party,"[ref]Id. at footnote 36[/ref] and contained "less than 40% original text."[ref]Id.[/ref] Put simply, cut-paste, then click-submit. So, it's not that the public was excluded, or ignored, it's more that 94% of the "public comments" were most likely generated by bots, in an effort to seem like a massive public comment. If all of this sounds vaguely familiar, similar tactics were deployed in Europe against various EU commissioners, filling their inboxes full of thousands of identically worded diatribes against the proposed Article 13, which sought greater copyright protections. EU commissioner Volker Reick, in an article translated and posted by The Trichordist, described it thusly: "[T]hose intent on scuppering the directive had not yet exhausted their firepower. What came now was the hour of the bots, the automatically generated emails, the automatically placed phone calls and the miraculous multiplication of protest, or rather its simulation. In the week before the plenary vote in the EU parliament on whether the adoption of the report by the JURI Committee and the negotiation mandate given to Voss should be allowed to stand, the inboxes of EU parliamentarians were flooded with automatically generated emails. Some EU parliamentarians reported having received 60,000 emails. In total, 6 million emails appear to have been dispatched to EU parliamentarians in this fashion. Compare that number to the handful of protesters in Berlin… Almost all the emails were identical in content, phrasing and formatting, and many even came from one and the same sender, presumably following the logic that more is better."[ref]Article 13: Anatomy of a Political Hack- Guest Post by Volker Rieck[/ref] So, what probative weight should the CO give over 86,000 "template text provided by a third party" comments? I'll let the EFF figure that one out. Plus, who do you think is sending them?

Whatever Happened to "Red Flag Knowledge"? No One Seems to Know.

An additional point of contention by the EFF is that the report seeks to "second guess" decisions made by Courts. "In short, the Office has taken upon itself to second-guess the many judges actually tasked with interpreting the law because it does not like their conclusions. Rather than describe the state of the law today and advise Congress as an information resource, it argues for what the law should be per the viewpoint of a discrete special interest." [ref]Internet Users of All Kinds Should Be Concerned by a New Copyright Office Report[/ref] Funny. This is what the EFF does every day of the week. And I seem to recall that the Department of Justice, amongst other government agencies, frequently makes recommendations to Congress on legislative changes. But, I digress. Case in point is the whole issue of "red flag" knowledge. According to the report: "OSPs that seek protection under the section 512(c) and (d) safe harbors must also act expeditiously to remove or disable access to material when they have ‘actual knowledge' of infringement or, in the absence of such actual knowledge, when they have ‘aware[ness] of facts or circumstances from which infringing activity is apparent'—the ‘awareness' standard often referred to as ‘red flag' knowledge." [ref]Report of the Register of Copyrights at 26[/ref] In a series of Court decisions, the Courts have effectively removed "red flag knowledge" from the statute, as this author contended in a previous blog post.[ref]Forget What Congress Wrote, Forget What the Copyright Office Wrote; Court of Appeals Re-Writes the Copyright Act[/ref] This was complained about repeatedly by rightsholders at both roundtables. The CO agrees: "The Office believes a standard that requires an OSP to have knowledge of a specific infringement in order to be charged with red flag knowledge has created outcomes that Congress likely did not anticipate.  The Copyright Office reads the current interpretations of red flag knowledge as effectively removing the standard from the statute in some cases, while carving an exceptionally narrow path in others that almost requires a user to "fess up" before the OSP will have a duty to act.  OSPs are correct that Congress likely did not intend to adopt a general awareness standard for red flag knowledge, since such a standard would consume many OSPs Congress otherwise sought to protect. Yet courts have set too high a bar for red flag knowledge, leaving an exceptionally narrow space for facts or circumstances that do not qualify as actual knowledge but will still spur an OSP to act expeditiously to remove infringing content."[ref]Report of the Register of Copyrights at 123[/ref] This is not, as the EFF contends, telling Congress what the law "should be." This is the CO telling Congress that the Courts are not giving effect what is already in the law. And consider this telling point of the Report: "Significantly, Congress set up the actual knowledge standard as distinct from red flag knowledge… [b]ut as interpreted now, there is little space between the two.  To that point, at the Washington, D.C., public roundtable in April 2019, participants repeatedly were asked for an example of an activity that would be less than actual knowledge but would create red flag knowledge under the current judicial interpretations of the statute.  Stakeholders were unable to identify a single activity."[ref]Id. (Emphasis added)[/ref] I remember this quite well. I was on one of the panels that this was asked at. I remember feeling that the air had kind of gone out of the room. Remember, I've got the guy from Google sitting next to me. If he can't figure out what "red flag" knowledge is, then something is seriously wrong and needs to be fixed.

Can We Get a Unified Takedown System?

I was able to be heard on a pet-peeve of mine, namely the passive-aggressive system that Google and others take to discourage the sending of takedown notices.[ref]How to Send a Takedown Notice to Google in 46 (or more) Easy Steps![/ref] "[Y]ou are asked a series of questions to justify your takedown before you can even file your takedown, including [a question about whether you are] the subject of the photograph, in which, if you say yes, you get this bright red warning saying if you're the subject of the photograph, you're most likely not the copyright owner, as if the selfie had never been invented . . . . Google will not let you file a takedown notice unless you create a Google account, which requires you to agree to Google's terms of service.  This includes a choice of jurisdiction and venue in Google's favor.  And this is before you even get to the takedown page."[ref]Report of the Register of Copyrights at 153, footnote 842[/ref] I suggested at the time, a unified takedown form ought to be put in place that would always be acceptable. It seems that the CO agrees. "The Office also agrees with many rightsholders that the bespoke nature of each OSP's webform, combined with DMCA pages that are not readily accessible from the homepage and do not always contain direct contact information for the OSP's designated agent, results in significant increases in the time and effort that must be invested by a rightsholder to submit a takedown notice. The Office acknowledges that the use of nonstandard and handwritten takedown notices requires OSPs to expend additional time and effort to process them, and that some users seek redress for issues other than copyright through a takedown notice under section 512.  To help alleviate some of this burden in the near term, the Copyright Office will publish standard notice and counter-notice forms and will develop additional user education materials regarding the types of notices that are appropriate under section 512."[ref]Id. at 159[/ref] That's a good start. "There is an opportunity, however, for Congress to address the growing disconnect between a statute written for communications prevalent in the late 1990s and current technological expectations and capacities while simultaneously creating a more standardized system that benefits OSPs, rightsholders, and the public.  Congress may want to consider a mechanism for ‘future-proofing' the standard notice requirements under section 512 by shifting enumeration of notification methods from a statutory mandate towards a regulatory process.  For example, since section 512 requires only that an OSP's agent information be ‘publicly available' on its website, there is currently no standardized practice for the location or content of user notifications regarding the takedown process.  Further, many websites require navigating through multiple links (such as ‘Help' or ‘Terms of Service') before reaching the web-based submission form or designated agent contact information.  Congress could modify the language of section 512(c)(2) to provide that the designated agent's information be not just ‘on its website in a location accessible to the public,' but also ‘prominently displayed,' and delegate to the Copyright Office the power to set regulations regarding minimal placement and content standards."[ref]Id. at 158-159[/ref] Now, we're getting somewhere.

On June 11, 2020, the Second Circuit handed down a decision in the case of Greene v. Paramount [ref]2020 WL 3095916 Second Circuit 2020 (marked not for publication)[/ref] on a question the Court had to admit it had not previously "spoken definitively on:" can a person be libeled by what is clearly labeled a work of fiction? At issue is the movie "The Wolf of Wall Street." The movie is based on the book of the same name written by Jordan Belfort, detailing his (very real) exploits at the brokerage firm Stratton Oakmont. "Belfort wrote about various criminal and other unsavory activities at Stratton Oakmont in the 1990s. Belfort was charged with and pled guilty to, inter alia, securities fraud and money laundering, for which he served time in prison and was ordered to pay over $100 million in restitution."[ref]Id. at 1[/ref] The Plaintiff, Andrew Greene, worked at Stratton Oakmont. In fact "Greene is discussed extensively in the Book both under his full name, Andrew Greene, as well as his nickname ‘Wigwam' (a reference to his toupee). The Book describes Greene as engaging in illegal conduct at Stratton Oakmont. Despite his awareness of the contents of the Book, Greene never sought any legal redress with respect to its depiction of him."[ref]Id.[/ref] The first problem for the Plaintiff is that this does not carry over into the film "The Wolf of Wall Street." There is no character named Andrew Greene. The character that Greene is upset about is named "Nicky Kosskoff," who in fact, is the name of the husband of one of the film's producers.[ref]Id.[/ref] Thus, the Plaintiff is faced with a daunting burden of proof. In order to prevail he must prove the libelous statement were "of and concerning" him.[ref]Id. at 2[/ref] Faced with the fact that there is no character of the same name, the character must have traits or take actions that would lead anyone who knows the Plaintiff to recognize that the character is in fact, the Plaintiff. Yet, in order to prove libel, the Plaintiff must then proceed to prove the opposite, that what the character does is so terrible and awful that the character could not possibly be him, and is therefore defamatory. Or put more succinctly: "That's me! But it couldn't possibly be me!" The similarities are scarce. Both the character Nicky Kosskoff and Plaintiff Andrew Greene worked at Stratton Oakmont, and both wore toupees. Otherwise, there are few similarities. The screenwriter claimed that Nicky Kosskoff was in fact a composite for three different people. Compounding the Plaintiff's problems is that he admits that he is a public figure. This, under the landmark case of New York Times v. Sullivan,[ref]376 U.S. 254, 279-80, 286, 84 S.Ct. 710 (1964).[/ref] requires the heightened showing of "malice," that either the publisher knew the statement was false and published it anyway, or published it despite seeing obvious reason to doubt the truth of the statement. In affirming the dismissal with prejudice, the Court points to the disclaimer shown as the first slide of the end credits. "While this story is based on actual events, certain characters, characterizations, incidents, locations and dialogue were fictionalized or invented for purposes of dramatization. With respect to such fictionalization or invention, any similarity to the name or to the actual character or history of any person, living or dead, or any product or entity or actual incident, is entirely for dramatic purpose and not intended to reflect on an actual character, history, product or entity."[ref]2020 WL 3095916 at 2[/ref] But this is a bit of a slight of hand. The majority of the audience appeal of "The Wolf of Wall Street" (or "The Irishman," or "Goodfellas") is that these are true stories about real people. To then run and hide behind "this is a work of fiction" is not quite playing on a level field. And it doesn't always work. A jury in Georgia awarded $100,000 to a woman who claimed she had been defamed by a work of fiction titled "The Red Hat Club."[ref]When Art Imitates Life: Suing for Defamation in Fiction[/ref] There, unlike Andrew Greene, the Plaintiff was able to show over two dozen "specific similarities" between her and the character "Su-Su," who much to the dismay of the Plaintiff, is portrayed as an promiscuous alcoholic.[ref]Id.[/ref] It is often said that truth is stranger than fiction. But then again, the truth is not actionable, and the fiction is.

On June 2,2020, District Judge Liam O'Grady denied various motions by Cox Communications trying to wiggle out of the $1 Billion verdict handed down in December 2019.[ref]Sony Music Entertainment et al V. Cox Communications, Inc., et al Case No. 1:18-cv-950-LO-JFA. No other citation is currently available.[/ref] These motions were Renewed Motion for Judgment as a Matter of Law and Motion for Remittitur (Reduction of Damages) or, in the Alternative, a New Trial. Cox won a partial victory in that the Judge ordered that the number of works at issue be recalculated. The bad news for Cox is that the Judge let stand the jury's award of statutory damages: $99,830.29 for each work infringed.[ref]Id. at 7[/ref] At issue was Cox's "repeat infringer policy," a requirement for Cox to maintain "safe harbor" under the Digital Millennium Copyright Act (DMCA). This means, in theory, that an ISP must have a policy in place for terminating the account of people who routinely use the ISP service to infringe copyrights. Cox had such a policy. It simply refused to enforce it in any meaningful manner. The team within Cox responsible for enforcing the policy was the "Cox Abuse Team" According to the Court: "CATS implemented a graduated response to address the reported infringing activity; the graduated response involved a thirteen strike policy, or ‘13 plus' given that the customer facing action generally began at the second notice."[ref]Id. at 5[/ref] If this sounds familiar, it's because this is the strategy that got Cox tagged for $25 million in a case brought by BMG Rights Management. Way back then, I summarized Cox's DMCA policy as follows:[ref]14 Strikes and You're Out! (Maybe): How Cox Communications Lost its DMCA Safe Harbor[/ref]

• If Cox receives a notice of infringement against a subscriber, it places a "strike" against the account, but does not send the notice to the subscriber.
• If Cox receives subsequent notices, strikes are placed on the account, and notice is sent to the subscriber.
• Cox takes no further action until the account has received 8 "strikes" against the account. Then the account holder is sent to a static "detention" page warning them against further infringement and instructing them to delete all infringing files and remove file sharing software. There is no real penalty and the subscriber can re-enable the account by simply clicking on the link on the web page.
• If the subscriber gets another warning notice, it is sent back to "detention," but the subscriber account can again be re-enabled easily.
• On the tenth notice, the subscriber is sent to "jail," i.e. a web page that they cannot exit. They must call Cox customer service to get their account re-enabled.
• On the eleventh notice, the subscriber is sent back to "jail," and must call Cox customer service, again.
• On the twelfth and thirteenth notices, the subscriber is sent back to "jail," and must talk to a higher level of Cox Customer service to get their service reinstated.
• On the fourteenth notice, Cox will review the account and "consider" termination, but termination is never the automatic remedy.

Tellingly enough, the "Cox Abuse Team" got renamed the "Customer Safety Team"[ref]Sony Music Entertainment et al V. Cox Communications, Inc at 6[/ref] reflecting the shift from routing out bad actors as required by the DMCA, to protecting the bottom line of Cox Communications. One email read "This customer will likely fail again, but let's give him one more change [sic]. He pays 317.16 a month." Another stated "we want to hang on to every subscriber we can."[ref]Id. at 20[/ref] But the most blunt was one email that stated "F*** the DMCA."[ref]Id. at 7[/ref] This policy shift was also reflected in Cox limiting the number of notices any one party could send, and reducing the staffing of the CST from nine to four.[ref]Id. at footnote 7[/ref] At trial, one team member admitted "we told each copyright owner to limit [the notices] or give us money to hire people."[ref]Id. at 6-7[/ref] Never mind, of course, that enforcing the repeat infringer policy is YOUR responsibility under the DMCA, and a cost that YOU need to absorb. In the order, the Judge finds that there was sufficient evidence to support the jury's finding of direct infringement, vicarious liability and contributory liability. The only real question comes in what is the proper number of works "at issue" for the award of statutory damages. Cox argues that the correct approach is to simply multiply the award times the number of discrete copyright registrations.[ref]Id. at 23 footnote 15[/ref] The Court notes that this is not correct under 4^th Circuit and most all other circuits.[ref]Id.[/ref] However, the Court notes that this issue ignores the impact of compilations and derivative works. In the First, Seventh, Ninth, Eleventh and D.C. Circuits, the test for how many copyrights there are in a compilation for calculating statutory damages is whether the copyrights that make up a compilation have economic value standing apart from the compilation. This is the so-called "Independent Economic Value Test."[ref]Id. at 25[/ref] Opposed to that is the Second Circuit, which treats the compilation as one copyright no matter how many individual parts are contained within it.[ref]Id. at 27-28[/ref] Indeed, with photographs, the individual components could number in the thousands, in one case 9,000.[ref]Id. at 30[/ref] However, if the components were issued individually, say as a musical single, as well as a compilation, a musical album, more than one damage award could be supported.[ref]Id. at 28 citing EMI Christian Music Group v. MP3 Tunes 844 F.3d 79 (2d Cir. 2017)[/ref] Here, the Court sides with the Second Circuit, but finds that since all the songs at issue were offered individually through various services, this does not reduce the number of songs at issue.[ref]Id. at 36[/ref] Derivative works are a different concern. The Plaintiffs include 16 record company plaintiffs, and 37 music publisher plaintiffs.[ref]Id. at 2[/ref] This question arises because the song is one copyright, the sound recording is a derivative work of that same copyright, yet it is entitled to its own separate copyright.[ref]17 USC 101[/ref] Are these two works? Or one? The Court seems to feel there should only be one award of statutory damages as the derivative work (the sound recording) wholly encompasses the underlying work (the song).[ref]Sony Music Entertainment et al V. Cox Communications, Inc at 41[/ref] It cites to the language of Section 504(c) that "all parts of a compilation or derivative work constitute one work." The Court puts forth a hypothetical where an infringer infringes a song where the song copyright and sound recording copyright are owned by the same person. This results in one award. Suppose the same infringer infringes a song in which there are ten licensed samples of ten sound recording and ten musical compositions. (Believe me this is very possible). So now the infringer is facing 20 damage awards? The Court refuses to go this far.[ref]Id. at 49[/ref] So now the damage award itself is not going to be reduced, but the multiplier, the number of "works" at issue will be. The Court instructs the parties to offer their calculations on how many songs have been subsumed into the sound recordings. This does not appear to me to be a simple task. Consider that there might be multiple "covers" of a song. One song might have been subsumed by a sound recording also at issue. But that does not mean that another "cover" might exist that has not been sued on. This would invoke the ownership of the song itself, apart from the sound recording. Stay tuned. Not only for the future order, but the inevitable appeal.