Hotfile is liable for the “staggering” amount of infringement it financially benefitted from, and it is not protected by the DMCA safe harbor for online service providers.
That is the conclusion of U.S. District Court Judge Kathleen Williams of the Southern District Court of Florida. Williams made the ruling August 28, but the decision had been under seal until this past Friday.
What makes this case notable is that it was brought by the five major motion picture studios (Disney Enterprises, Inc., Twentieth Century Fox Film Corporation, Universal City Studios Productions LLLP, Columbia Pictures lndustries, Inc., and Warner Bros. Entertainment Inc.) against a fairly popular filelocker service (at least at the time the suit was filed). This appears to be the first time an infringement lawsuit from major players in the creative industries had been brought against a filelocker service specifically — previous efforts have aimed at P2P services and the like. Add to that the fact that the suit was filed in Florida, outside the 2nd and 9th Circuits where the bulk of copyright litigation occurs, and you have a situation where the court is interpreting many DMCA provisions on a blank slate.
Oh, and the decision also comes in at just under 100 pages.
I had previously discussed the issues in front of the court and both parties’ arguments. That was over a year ago, which is an unusually long time between hearing and order on a summary judgment motion. Part of that may have been the sheer complexity of the issues; part of that may have been the aggressive litigation posture taken by both sides. The court diplomatically refers to the many “robust pleadings” filed without leave of court and notes, certainly an understatement, that “the parties do not agree on much.”
Whatever the case may be, the issue boils down to a familiar one. Hotfile, an “off-shore technology company” provides online file storage. As is often the case, infringing works are among the files that were stored and shared by Hotfile users. The film studios argued that Hotfile should be held liable for such infringement because it contributes to, encourages, or benefits from such infringement.
As the court explains, Hotfile’s storage locker service allows registered users to upload any file they want from their computer to Hotfile’s servers to be stored. The service automatically generates a link where the file can be accessed. The files or links are not otherwise private; any member of the public can access them so long as they know the link. Hotfile operated an affiliate program which paid users when they directed others to Hotfile file locations, encouraging affiliates to catalog and broadcast Hotfile links. Hotfile also provided premium service, which gave users additional file space and faster download speeds for a monthly fee.
The court begins with a careful, comprehensive discussion of the facts (nearly 35 pages worth). As we’ll see in a moment, the question of whether Hotfile had adopted and reasonably implemented a repeat infringer policy will play a key role in determining whether the service is protected by the DMCA safe harbor, but I do want to highlight the court’s discussion regarding the policy that was in place — some of these numbers are staggering. The evidence reveals that when the studios had filed their complaint, Hotfile had received a total of 10 million takedown notices for infringing content, yet had only terminated 43 users – 33 of those as a result of a court order from prior litigation. At the same time, nearly 25 thousand Hotfile users had accumulated more than three infringement notices; 61 of those users had over 300 notices each. After the litigation began, Hotfile adopted a “revamped” repeat infringement policy, and the results were dramatic: 444 of its 500 highest paid affiliates were terminated for repeated infringement.
The DMCA safe harbor and the repeat infringer policy
Turning to the legal issues, the court begins with the DMCA safe harbor. 117 U.S.C. § 512. As it notes, if Hotfile qualifies for the safe harbor, it is immunized from any liability for infringement. If it doesn’t, the court must then separately consider if Hotfile is liable. The court cites to the relevant legislative history of the DMCA to explain the law’s motivations in striking a balance between protecting creators’ rights and promoting the growth of online services.
The DMCA applies to online service providers, which Hotfile clearly is, and one of the safe harbors covers storage at the direction of the user, which the court says covers the activities at issue here. 2I note that the court glosses over one of the studios’ (admittedly far-reaching) arguments that the infringement in question was not within the scope of the storage safe harbor. The studios had argued that Hotfile’s practice of routinely deleting files that have never been downloaded was the antithesis of “storage” and thus not protected under the safe harbor.
The DMCA safe harbor requires that a service provider
has adopted and reasonably implemented, and informs subscribers and account holders of the service provider’s system or network of, a policy that provides for the termination in appropriate circumstances of subscribers and account holders of the service provider’s system or network who are repeat infringers. 317 U.S.C. § 512(i)(1)(A).
The court here turned to the legislative history and other courts to flesh out the details of this provision, since the terms are not defined in the statute. It determined that a policy is “reasonably implemented” if it terminates users, under appropriate circumstances, who “repeatedly or blatantly infringe copyright.” At a minimum, this means that a policy “must be capable of tracking infringers.” Hotfile, said the court, “effectively did nothing to tie notices to repeat infringers.” This, in addition to the breathtaking scope of infringement that Hotfile essentially ignored took the service provider outside the protection of the DMCA safe harbor. The court concluded:
Here, the scale of activity – the notices of infringement and complaints from copyright holders — indicated to Hotfile that a substantial number of blatant repeat infringers made the system a conduit for infringing activity. Yet Hotfile did not act on receipt of DMCA notices and failed to devise any actual policy of dealing with those offenders, even if it publicly asserted otherwise. It has presented no evidence to show that the small number of removals that did occur were for any reason other than threatened Iitigation or by court order. lndeed, it has been unable to point to a single specific user who was terminated pursuant to its policy of manual review and exercise of ”discretion.” Documents and statistics indicate that there was never any realistic threat of termination to Hotfile’s users, whose activities were protected by the company’s indifference to infringement notices. In sum, regardless of official policies forbidding infringement, Hotfile did not significantly address the problem of repeat infringers. This renders Hotfile’s policy legally insufficient under Section 512(i).
While this, on its own, was enough to kick Hotfile out of the safe harbor, the court continued to provide observations and conclusions on two other DMCA requirements. The first provides that service providers designate an agent to receive infringement notices and register the agent with the U.S. Copyright Office. Hotfile, however, did not register an agent until December 2009 and had not identified the agent on its website as required by the statute until May 2010. Additionally, the court suggested that Hotfile was still not in compliance with this requirement since it “to date, has not provided a proper mailing address for its registered agent insofar as it lists only a post office box,” contrary to Copyright Office regulations.
The second requirement is that a service provider cannot have either actual or “red flag” knowledge of infringement. But, as the court notes, the DMCA places limitations on this requirement. Under the statute, a service provider does not initially have a duty to affirmatively monitor for infringement. Courts have also determined that the statute requires the knowledge be of specific infringing activity rather than a general awareness of infringement, a determination that the court here adopts. At the same time, willful blindness is a form of knowledge, and courts have struggled with incorporating willful blindness into the DMCA consistent with its limitations on the knowledge prong — the court here notes the complexity of this inquiry but doesn’t attempt to clarify it.
Instead, it concludes that whether Hotfile had the requisite knowledge that would disqualify it from the DMCA safe harbor was a question that would need to be decided by a jury. This conclusion is a little odd procedurally, as the court had already granted summary judgment denying Hotfile’s DMCA defense, meaning the question of knowledge would never reach a jury, but perhaps Judge Williams wrote it anticipating an appeal.
Secondary liability for copyright infringement
Next the court turns to the question of infringement itself. As it notes, the DMCA safe harbor and secondary liability are independent inquiries (even though there is plenty of overlap between some of the factors and the evidence that is relevant to each). And while it has been a challenge for courts to apply common law secondary liability principles to new technologies, they still play an important role since copyright protections would be rendered meaningless if creators were limited to going after each individual infringers using such services.
Inducement and contributory infringement are discussed first. The court marches through some of the most relevant precedent here, including, among others, the Supreme Court’s Grokster decision, the recent Lime Group decision, and the Ninth Circuit’s Perfect 10. The court extracts from this precedent its own standard for examining contributory infringement, one that provides three separate avenues for establishing secondary liability:
First, while it may be unclear whether Grokster introduced a new category of Iiability based on inducement or whether it spoke to pre-existing notions of contributory liability, it is evident that a defendant will be liable for actually expressing an intention to foster infringement. If that intent is express or can otherwise be said to be ‘’unmistakable,’’ the Sony/Betamax defense will not apply and the defendant will be liable for aII acts of direct infringement committed using its system, as was the case in Grokster. Similarly, as explained in Amazon.com, where traditional principles permit a court to impute intent – for instance, where the defendant knows of specific infringing content available on its system yet fails to remove it – that defendant may be Iiable, by operation of Iaw, just as if he had actually intended to infringe under Grokster. Finally, contributory infringement may be found based on a material contribution theory in instances where a defendant did not express an intention to foster infringement but provided the means for infringement of distributed a commercial product that was subsequently used to infringe. Under that theory, the Sony/Betamax rule provides a backstop of liability, immunizing a defendant who demonstrates that noninfringing uses of the system are substantial.
Here, the evidence left a number of unresolved questions about whether Hotfile is liable under any one of these avenues, and the court concludes that “while Hotfile may have difficulty explaining its ‘innocence’ to a jury, the genuine issues of material fact must be resolved by a jury at trial.”
Less complex is the vicarious liability inquiry. The standard used by the court here is two pronged: one is vicariously liability if he (1) profits from direct infringement while (2) declining to exercise a right to stop or limit it.
The profit, or financial benefit, from infringement does not, the court observe, need to be a direct one, or even a significant one. It is enough that there is a causal relationship between infringement and benefit to satisfy this prong. And here, the court found that the evidence did establish a connection between infringement on Hotfile and its sales of premium subscriptions, concluding that “it is undeniable that it financially benefitted from [infringement] by attracting some users.”
The second prong was just as easily met. The infringing files were stored on Hotfile’s own servers, and it contractually retained a right to control uploaded files in its terms of service. As a matter of law, then, Hotfile is vicariously liable for copyright infringement.
Two final issues were addressed in the court’s order. First, the court held that Anton Titov, one of the primary corporate officers of Hotfile, is, as a matter of law, personally liable for any infringement along with Hotfile itself. And second, the court found that there was enough evidence to support Hotfile’s counterclaim against one of the studios involved, Warner. The evidence suggested that Warner intentionally sent takedown notices for files that it knew it had no right to remove.
While the decision has been a long time coming, and while it will take a long time to digest — it’s very thorough — this is likely only the beginning. Hotfile has already appealed the decision to the Eleventh Circuit.
|↑1||17 U.S.C. § 512.|
|↑2||I note that the court glosses over one of the studios’ (admittedly far-reaching) arguments that the infringement in question was not within the scope of the storage safe harbor. The studios had argued that Hotfile’s practice of routinely deleting files that have never been downloaded was the antithesis of “storage” and thus not protected under the safe harbor.|
|↑3||17 U.S.C. § 512(i)(1)(A).|