by Lee Gesmer | Nov 8, 2012 | Copyright
Before the Internet made file sharing ubiquitous, liability for “indirect” copyright infringement was something of a legal backwater.* Massive file sharing of audio, image and video files has changed that. Where a website actually hosts a copyrighted file uploaded by a user, the legal rights of the parties are relatively clear: the uploader (and subsequent downloaders) are liable for “direct” infringement. The legal rights of the website owner are governed by the Digital Millennium Copyright Act (DMCA).**
*A notable exception being the Supreme Court’s pre-Internet decision in Sony Corp. of America v. Universal City Studios, Inc., which rejected a claim of contributory infringement directed at the VCR, since Sony did not encourage copyright infringement, and the VCR was capable of commercially significant noninfringing uses.
**Caveat: the courts are still working at interpreting and applying the DMCA. The most recent appellate decisions are Viacom v. Youtube (2nd. Cir. 2012) and UMG v. VEOH (9th Cir. 2011).
However, there are many situations where the DMCA does not apply because the illegal file is not being hosted by the defendant (that is, the file is not resident on a server owned or controlled by the defendant). In those cases, where there are countless uploaders and downloaders (many of which cannot easily be identified), copyright owners will sometimes sue the website owner. A single case, if successful, has the potential to inhibit access to thousands of illegal files. However, because the defendants in these cases never reproduced or published the works itself, copyright owners must argue “indirect” (as opposed to “direct”) infringement. Indirect infringement is an overlapping and often confusing pastiche of doctrines, which includes “vicarious” and “contributory” copyright infringement, as well as “inducement.”*
*For example, in the Grokster case, where the defendant provided a tool that enabled others to commit direct infringement (peer-to-peer file sharing), the Supreme Court found Grokster liable for inducing infringement—actively encouraging infringement by use of its file sharing program—despite the fact that Grokster did not host the infringing files.
The “first generation” line of cases represented by Grokster was relatively easy for copyright owners, since the defendants naively encouraged users to use their site or product to access or exchange infringing works. However, these “shooting fish in a barrel” lawsuits* are likely coming to an end as people creating these programs get wise to the law and stop sending each other incriminating emails that copyright owners can use to prove illegal motivation. The new, more difficult generation of cases is illustrated by the Seventh Circuit’s recent decision in Flava Works v. myVidster, where the issue was whether embedding a copyrighted video constituted contributory copyright infringement.**
*Napster and Aimster are part of this line of cases. A number of cases where the key evidence is the words of the defendant are still working their way through the courts. A recent example, decided by a California federal district court this summer, is David v. CBS. The court denied CBS’s (actually download.com) motion to dismiss a claim of copyright infringement relating to links to third-party P2P software based on, among other things, download.com’s public statements comparing the P2P software to other P2P programs known for copyright infringement (i.e., Napster and Limewire).
**Another example of the new generation of online copyright cases is Capital Records v. Redigi, discussed here.
myVidster is a “social video bookmarking service”—users of myVidster can embed the thumbnail of a video on myVidster which, when accessed (clicked), plays the video in a “frame” on the myVidster site. However—of critical importance—the video remains on the server to which it was originally uploaded; it is not hosted on the myVidster servers. As the opinion states, “myVidster doesn’t touch the data stream, which flows directly from one computer to another, neither being owned or operated by myVidster.”
As it so happens, some of the videos are copyrighted by Flava. Flava, upset that myVidster was facilitating access to its copyrighted videos, sued Vidster for copyright infringement. However, because myVidster did not host the files (and hence was not a “direct” infringer), Flava was reduced to arguing contributory infringement. This presented a problem for Flava—myVidster must have read a few of the cases cited above, or hired a savvy lawyer, because there was no evidence it was encouraging the use of its site to host copyrighted files. And, the decisions don’t reference any inculpatory internal emails that Flava could use to prove illegal intent. So, Flava did not have as evidence the kind of “illegal talk” that has been key in many earlier contributory infringement cases.
Faced with this situation, Flava played the hand it had been dealt as best it could. It notified myVidster that it was providing links to copyrighted Flava files (in the form of DMCA take down notices), and demanded that myVidster take down links to Flava-owned videos. myVidster declined.
This was enough for the district court, which found MyVidster’s refusal to comply to be “the epitome of willful blindness.” Moreover, the District Court found that MyVidster’s actions (i.e., hosting a site that provided links and embedded videos) materially contributed to the copyright infringement. The district court issued a preliminary injunction, ordering myVidster to take a number of steps intended to exclude Flava videos from myVidster, including honoring DMCA takedown notices and actively filtering to identify and exclude Flava videos.
However, Flava could not hold on to this ruling before a Seventh Circuit panel headed by renowned Judge Richard Posner. As noted, since myVidster did not upload the video to the various sites, the question was not one of “direct” infringement, but rather contributory infringement. Capturing the legal concept of contributory infringement is not a simple matter, as the Seventh Circuit explained in a Posnerian passage:
A typical, and typically unhelpful, definition of “contributory infringer” is “one who, with knowledge of the infringing activity, induces, causes or materially contributes to the infringing conduct of another.” . . . Such a one “may be held liable as a ‘contributory’ infringer.” . . . But does “may be held liable” mean that a person who fits the definition of “contributory infringer” may nevertheless not be a contributory infringer after all? And what exactly does “materially contribute” mean? And how does one materially contribute to something without causing or inducing it? And how does “cause” differ from “induce”?
The opinion proceeds to the conclusion that myVidster is not a contributory infringer (at least on the limited record associated with a preliminary injunction, which was the procedural posture of this case), for several reasons.
First, Judge Posner reasoned that although the initial uploaders are “direct infringers” myVidster is separated from this infringement by the linkers (or “bookmarkers”), and therefore is too remote from the infringement to be a contributory infringer. The bookmarkers are not infringers (only the uploaders are), and “the facilitator of conduct that doesn’t infringe copyright is not a contributory infringer.”
Second, Flava provided myVidster with DMCA takedown notices, but since myVidster is not hosting infringing videos, myVidster was justified in disregarding these notices. The DMCA provides a safe harbor to websites that host copyright protected works posted by users (user generated content), but in the words of the Seventh Circuit, “A noninfringer doesn’t need a safe harbor.”
Third, as noted, based on the preliminary injunction record myVidster is not inviting people to post copyrighted videos, and therefore is not inducing infringement, as was the case in the “first generation” file sharing cases referenced above.* This is absolutely essential for defendants in cases of this sort, whether they host the infringing works (as in Viacom v. Youtube), or provide links to infringing works, as myVidster does. In this case Flava argued that myVidster was aware that its site was serving as a clearinghouse for infringing copies, and that its willful blindness to this fact caused it to cross the line. However, the Seventh Circuit declined to hold that willful blindness led to contributory infringement where myVidster did not encourage links to copyrighted works.
*Although, it might be noted, using a “ster” suffix might not be the best choice for an online service of this kind. (Think Grokster, Napster, Aimster . . . .).
Fourth, not only is myVidster not copying the Flava copyright-protected works, but it is not performing them publicly. As noted above, myVidster doesn’t touch the datastream when a video is accessed and therefore, technically speaking (the Seventh Circuit found), myVidster is not “performing” the work. The server hosting the video and the myVidster end-user that initiates display of the video may be publicly performing the video, but myVidster is not. The Seventh Circuit’s analysis on this issue is analgous to the “server test” adopted by the Ninth Circuit in Perfect 10 v. Amazon. That is, that infringement of the public display right can occur only when the copyrighted work resides on the defendant’s own server, not when it resides on another site’s server and is framed or displayed in-line.
An interesting aspect of this decision is the Seventh Circuit’s conclusion that because MyVidster did not host the copyrighted works, it was not subject to DMCA take-down notices. This is a controversial aspect of the decision, since the DMCA could be read to make link sites subject to DMCA notice and takedown, an interpretation of the statute that Judge Posner characterized as “implausible,” given its far-reaching implications. myVidster’s decision to ignore these notices took guts (or was foolhardy) on the part of MyVidster, since a DMCA defense would have allowed it to hedge its bets. However, the Seventh Circuit agreed with MyVidster’s argument that because MyVidster did not host the works, it was not subject to the DMCA.
This case was a close call for myVidster. The decision could easily have gone the other way in another circuit, or even before a different Seventh Circuit panel. See Perfect-10, Inc. v. Amazon.com, Inc. (“Applying our test, Google could be held contributorily liable if it had knowledge that infringing Perfect 10 images were available using its search engine, could take simple measures to prevent further damage to Perfect 10’s copyrighted works, and failed to take such steps”). It rested, in the final analysis, on the fact that myVidster was two steps removed from the infringing acts (uploader – linker – myVidster), and that myVidster did not encourage infringement. However, absent a settlement the case is far from over—this was only a preliminary injunction decision, with full discovery and trial yet to come. Flava may yet develop an evidentiary record sufficient to swing the case in the other direction.
Most importantly, the case shows how finely nuanced the law has become with respect to online contributory infringement. Stay tuned.
Flava Works v. myVidster
by Lee Gesmer | Nov 5, 2012 | Corporate Law, Noncompete Agreements, Trade Secrets, Trademark
Oriental Financial Group, Inc. v. Cooperativa De Ahorro y Crédito Oriental (1st Cir. October 18, 2012) — In this case the First Circuit adopts the trademark law “progressive encroachment doctrine,” joining the 6th, 7th, 8th, 9th and 11th circuits. The progressive encroachment doctrine may be used as an offensive countermeasure to the affirmative defense of laches (delay in brining suit) where the trademark owner can show that “(1) during the period of the delay the plaintiff could reasonably conclude that it should not bring suit to challenge the allegedly infringing activity; (2) the defendant materially altered its infringing activities; and (3) suit was not unreasonably delayed after the alteration in infringing activity” (quoting Oriental Financial).
Harlan Laboratories, Inc. v. Gerald Campbell (D. Mass. October 25, 2012) — Applying Indiana law, Judge Patti Saris issues a preliminary injunction enforcing a one year non-compete agreement. However, the opinion makes liberal use of Massachusetts and First Circuit precedents.
Blake v. Professional Coin Grading Service (D. Mass. October 6, 2012) — In this case, which involves alleged trade secrets associated with a method to grade the “eye appeal” of coins, Judge William Young concluded that the “method” was not subject to trade secret protection due to the fact it had been publicly disseminated before being disclosed to the defendants. However, Judge Young ruled that the case could proceed based on the alleged misappropriation of a proposed marketing plan. In addition to his analysis of trade secret law, the case contains an extensive discussion of Lanham Act issues including “reverse confusion” (which is always confusing) as well as the application of Massachusetts law to the intellectual property issues raised in the case (conversion, breach of contract, the covenant of good faith and fair dealing, unjust enrichment and civil conspiracy).
In DeJesus v. Bertsch, Inc. (D. Mass. Oct. 16, 2012) Judge Young conducts a detailed analysis of corporate successor tort liability under the Massachusetts “de facto merger” and “mere continuation” exceptions. In this case he concludes that the defendant corporation is not subject to successor liability.
by Lee Gesmer | Oct 31, 2012 | Copyright, Litigation
In Third Degree Films v. Does 1-47 (D. Mass. October 2, 2012), Judge William Young took on the “copyright trolls” in the adult film industry as best he could, holding that the plaintiff (a publisher of copyright-protected adult films that are being shared on the Internet) cannot join 47 “John Doe” defendants in a single action — it must instead file 47 individual suits.
The issue here is part of a larger controversy, the “porn film copyright shakedown.” The way this works is as follows. Copyright holders file Doe suits, which identify defendants only by IP address (all the plaintiff knows at that point). They then subpoena the ISPs and identify the owner of the IP address. Having identified the owners, they tell them that, absent a quick settlement (typically under $5,000), they will name them in the suit and serve them. Most people, rather than suffer the embarrassment (or what Judge Young calls the “reputational cost”) of having court records show that they downloaded films with titles like “Big Butt Oil Orgy 2,” settle out-of-court. Judge Young describes this process as “misusing the subpoena powers of the court, seeking the identities of the Doe defendants solely to facilitate demand letters and coerce settlement, rather than ultimately serve process and litigate the claims.”*
*As one court put it, a defendant – “whether guilty of copyright infringement or not — would then have to decide whether to pay money to retain legal assistance to fight the claim that he or she illegally downloaded sexually explicit materials, or pay the money demanded. This creates great potential for a coercive and unjust ‘settlement.'” SBO Pictures, Inc. v. Does 1-3036.
Why do the plaintiffs in these cases name tens, hundreds, sometimes thousands of Doe defendants in one suit? Money, money, money. By filing claims against multiple John Does, as Third Degree tried to do in this case, the plaintiff avoids a separate filing fee for each defendant (currently $350 per complaint). If Third Degree were required to file 47 separate cases, it would cost $16,450. Filing one case — $350. Assuming an average settlement of $2,500, joinder results in a gross of $117,500, while separate suits would yield a gross of only $101,050. Some porn film copyright cases name thousands of defendants. At $350 a pop, this can start to add up and the number of defendants climbs. In one case, for example, the plaintiff sought (unsuccessfully) to join 3,036 separate defendants.
No doubt, it’s more profitable for a copyright holder to be able to file a single case against multiple defendants and save money on filing costs, but that begs the legal question: why should this be permissible? Why shouldn’t porn film plaintiffs be required to file separate cases, like everyone else? The answer lies in the federal rules of civil procedure, which allow multiple defendants to be joined in a single case where the claims arise “out of the same transaction, occurrence, or series of transactions or occurrences.” (Fed. R. Civ. P. 20(a)(1)). How do the copyright plaintiffs claim the benefit of this rule?
The answer lies in a variation of the BitTorrent peer-to-peer file-sharing technology known as “BitTorrent swarm” or segmented file transfer. Simplified, a BitTorrent swarm distributes parts of a file among mutiple users. The file is downloaded from various sources simultaneously and assembled on the destination computer. Because movie files are very large (as compared with mp3 files, for example), this provides a faster, more efficient way to distribute these files.*
* BitTorrent swarms in the context of mass copyright filings are discussed on Slashdot here. For a semi-technical discussion in a court filing see the John Doe filing in Malibu Media v. John Does 1-5, here.
Although the 47 alleged downloaders in the Third Degree Films case may never have met each other or know each other’s identities, they all downloaded the same adult film and (according to Third Degree) were part of the same swarm. Perhaps any given member of the “swarm” interacts with electronically another participant, perhaps not. The larger the swarm (and a BitTorrent swarm can include thousands of users), the less likely that any one defendant will share part of the file with another. However, unbeknownst to them, using BitTorrent swarm technology made them susceptible to the “swarm joinder theory.”
Whether a plaintiff can name many defendants in a single suit based on swarm joinder is highly controversial. Judge Young issued an opinion on this issue a year ago in Liberty Media Holdings v. Swarm Sharing Hash File. In that case he permitted the “swarm participants” to be joined in one case, concluding that the Doe defendants’ behavior satisfied the “same transaction or occurrence” requirement. In the current case Judge Young stands by his reasoning in Liberty Media. He discusses the technological complexities of this issue, and concludes that even indirect interactions between swarm defendants may constitute “shared, overlapping facts” sufficient to establish a “series of transactions or occurrences.”*
*Judge Young noted it gave him “pause that district courts are so divided over whether file sharing via the BitTorrent protocol constitutes a series of transactions or occurrences . . . The inquiry is so fact intensive, and the BitTorrent protocol so technologically complex, that no principled conclusions have emerged from the abundance of recent case law and this Court is not entirely comfortable hanging its hat on its own understanding of the process.” Indeed, some of the discussion of whether an initial seeder indirectly uploaded pieces of a work to every peer in the swarm is surprisingly abstruse.
However, judges are not always bound by the letter of the law, and when it comes to multi-defendant joinder, Judge Young decided to take full advantage of his discretion. In this case, Judge Young expressed “serious concerns regarding the propriety of joinder of tens, hundreds, or thousands of Doe defendants” in adult film mass copyright infringement cases. He noted that each case would require a mini-trial (for example, one defendant suggested that her tenant, who occupies the other half of her two family house, must be responsible for the download). Combining 47 defendants with different defenses into one case was likely to create a “procedural albatross. Judge Young also took a dim view of the use of joinder to create a “low-cost, low-risk revenue model for the adult film companies.”
There have been many adult film (and conventional film) “copyright troll” cases in the last couple of years, and Judge Young’s decision collects court decisions in many of them. It appears that this phenomenon may be close to running its course. Many courts now see the issue the way Judge Young sees it, and are denying joinder in BitTorrent cases. Some plaintiffs have been sanctioned for abusing the joinder process. Bellwether trials are pending to test the extent to which an IP address can be used as the basis for legal wrongdoing, and whether the plaintiffs in these cases are correctly representing the technological properties of BitTorrent. Class action suits have been filed against Third Degree Films and other adult film companies based on their litigation practices. The easy money has been made, people are getting wise to the risk associated with these downloads and the courts have wised-up. The end may be in sight.
by Lee Gesmer | Oct 16, 2012 | Contracts
It’s easy to create an enforceable online “click-wrap” agreement. But, as two recent cases remind us, it’s also easy to do it wrong. Two recent cases are a reminder of this.
In the first case, In re Zappos.com Security Breach Litigation, Zappos was sued in connection with a large data security breach. Responding to the predictable class action lawsuit, Zappos argued that the plaintiffs were required to arbitrate under Zappos’ online user agreement. However, Zappos didn’t have a ‘user agreement,” it only had terms and conditions. And, it did not require purchasers to “click through” to indicate acceptance of those terms. The terms, which included the arbitration requirement, were under a link users were not even required to access while making a purchase, much less consent to. Quoting the court:
we cannot conclude that Plaintiffs ever viewed, let alone manifested assent to, the Terms of Use. The Terms of Use is inconspicuous, buried in the middle to bottom of every Zappos.com webpage among many other links, and the website never directs a user to the Terms of Use. No reasonable user would have reason to click on the Terms of Use, . . .
. . . The arbitration provision found in the Zappos.com Terms of Use purportedly binds all users of the website by virtue of their browsing. However, the advent of the Internet has not changed the basic requirements of a contract, and there is no agreement where there is no acceptance, no meeting of the minds, and no manifestation of assent. A party cannot assent to terms of which it has no knowledge or constructive notice, and a highly inconspicuous hyperlink buried among a sea of links does not provide such notice. Because Plaintiffs did not assent to the terms, no contract exists, and they cannot be compelled to arbitrate.
To make those terms enforceable Zappos needed to require purchasers to affirmatively signal their assent by “clicking” their agreement to be bound. It failed to do this, and therefore the terms were not binding. Zappos might have been better off in arbitration, but it’s stuck in court.
The second case, Schnabel v. Trilegiant (2d Cir., Sept. 7, 2012), is not an online license case per se, but it makes the same point. In this case the web site, a membership organization, emailed (or, if the email bounced, snail-mailed) terms and conditions to new members after they had already joined. The terms contained an arbitration clause. Could the plaintiff be required to arbitrate based on this “after the fact” mailing? Not under the circumstances in this case. The Second Circuit held:
a reasonable person would not be expected to connect an email that the recipient may not actually see until long after enrolling in a service (if ever) with the contractual relationship he or she may have with the service provider, especially where the enrollment required as little effort as it did for the plaintiffs here. In this context the email would not have raised a red flag vivid enough to cause a reasonable person to anticipate the imposition of a legally significant alteration to the terms of conditions of the relationship with [the defendant].
Much of the case law on online contract formation between vendors and purchasers comes from the Second Circuit. See: Specht v. Netscape Communications Corp., 306 F. 3d 17 (2nd Cir. 2002); Register. com, Inc. v. Verio, Inc., 356 F. 3d 393 (2nd Cir. 2004).
You might not know it from these two cases, but this is not rocket science: require customers to affirmatively consent to online terms before they perform the transaction, and the agreement likely will be enforceable.
Oh, one more thing – don’t state that the terms can be changed at any time, at the whim of the site owner, without obtaining new assent by the customer. A lot of terms contain a provision to this effect, but as the court noted in the Zappos case, where the site owner reserves the unilateral right to revise the online terms the contract becomes “illusory, and therefore unenforceable.” The district court in Zappos cites many of the cases applying this principle to online agreements. For a 2009 blog post discussing this aspect of online agreements, see here.
by Lee Gesmer | Sep 13, 2012 | Trademark
One of the thorniest issues in trademark law is whether and when trademark law will protect the use of a single color. After all, there are an infinite number of colors, and it would hardly be fair if one company could obtain a theoretically perpetual right to exclude others from using a color. So, the law makes it difficult to achieve this.
Cases involving color marks are rare, but the Second Circuit released an important decision last week in Christian Louboutin S.A. v. Yves Saint Laurent Am., Inc., (2nd Cir. 2012). The court held that Louboutin’s trademark, consisting of a red, lacquered outsole on a high fashion woman’s shoe (the “Red Sole Mark”), has acquired limited “secondary meaning” as a distinctive symbol that identifies the Louboutin brand, but (oddly) only where the red outsole contrasts with the color of the remainder of the shoe.
The heart of the decision is the court’s functionality analysis. Trademark law recognizes two types of functionality, “utilitarian” functionality, and “aesthetic” functionality. Utilitarian functionality occurs when a product feature is essential to the use or purpose of the article, or if it affects the cost or quality of the article. The court had no difficulty finding that the red outsole was not precluded from trademark protection by reason of utilitarian functionality, since the color serves no utilitarian purpose.
However, that did not end the inquiry. The court had to determine whether Louboutin’s Red Sole Mark was barred under the doctrine of “aesthetic” functionality.
Enforceable
In doing so, the court noted the “counterintuitiveness” of this inquiry: “how can the purely aesthetic be deemed functional, one might ask?” The court answered its own question by noting that “the doctrine of aesthetic functionality bars protection of a mark that is “necessary to compete in the [relevant] market.” A mark is “aesthetically functional, and therefore ineligible for protection under the Lanham Act, where protection of the mark significantly undermines competitors’ ability to compete in the relevant market.” The Red Sole Mark easily passed muster under this test.
Unenforceable
The district court had applied a per se rule that assumed functionality (and therefore unenforceability) for color marks in the fashion industry. The Second Circuit disagreed, stating that Supreme Court precedent “requires an individualized, fact-based inquiry into the nature of the trademark, and cannot be read to sanction an industry-based per se rule.” After reviewing the record, and Louboutin’s extensive evidence supporting secondary meaning for the Red Sole Mark, the court concluded that the color mark could be asserted as a trademark only when used as a red outsole contrasting with the remainder of the shoe; it could not be used when both the sole and the remainder of the shoe were red.
