By Nikolaos Theodorakis
On June 11 the European Commission announced the launch of an antitrust investigation pertaining to Amazon’s business policies in the distribution of electronic books (“e-books”). Amazon relies greatly on the success of its e-book enterprise, yet the Commission questions whether Amazon does so to the detriment of its consumers. The investigation adds to the pressure on the online retailer in Europe, since it is already being investigated for low tax rates it pays in Luxembourg.
The e-books sector has sparked investigations on antitrust concerns in the past. In 2011, the European Commission launched proceedings due to concerns that certain international publishing houses (Penguin Random House, Hachette Livres, Simon & Schuster, HarperCollins and Georg von Holtzbrinck Verlagsgruppe), in collaboration with Apple, could have colluded to limit retail price competition for e-books in the EEA, in violation of applicable antitrust rules. Amazon prodded the US Justice Department to file an antitrust suit for conspiracy to fix e-book prices.
Following this investigation, the companies offered a number of commitments in December 2012 and July 2013 to address the concerns. The European Commission has recently pursued tax, antitrust and similar investigations into the businesses of Apple, Google and Facebook. The Commission is also pursuing an antitrust investigation on whether large tech companies have impeded competition in Europe’s online shopping industry. President Obama has, however, warned on the danger of Europe turning to protectionism to the detriment of the U.S. technological sector.
Scope of the investigation
Amazon is the largest distributor of e-books in Europe, taking advantage of the vast growth of the industry in the past years. Commission’s investigation will focus on e-books in English and German.
Specifically, the Commission has targeted provisions that are included in Amazon’s contracts with publishers. The disputed clauses allegedly require publishers to inform Amazon about: (i) more favourable or alternative terms offered to Amazon’s competitors, or (ii) offer Amazon similar terms and conditions than to its competitors, or (iii) through other means ensure that Amazon receives terms at least as good as the ones of its competitors. Since Amazon is the most powerful e-book retailer, these clauses might prevent an innovator from breaking through.
The Commission fears that such clauses hamper healthy competition since they make it more difficult for other e-book distributors to compete with Amazon through the development of comparatively advantageous strategies. Consequently, restrictive measures of this nature reduce the choice that consumers have. Such behaviour could also count as abuse of a dominant market position, and restrictive business practices. Ultimately, these provisions hinder the level playing field and potentially decrease competition between different e-book distributors.
As is the case in relevant investigations, the legal basis of potential antitrust is based on Articles 101 and 102 of the Treaty on the Functioning of the European Union (TFEU), which prohibit anticompetitive agreements and the abuse of dominant market positions. These provisions are implemented in the EU’s Antitrust Regulation (Council Regulation No 1/2003), which is also applied by national competition authorities.
Article 11(6) of the Antitrust Regulation supports that the initiation of proceedings by the Commission, and relieves the competition authorities of the Member States of their competence to apply EU competition rules to the practices in question. Article 16(1) of the said Regulation provides that national courts must avoid handing down decisions that could conflict with a decision contemplated by the Commission in already launched proceedings.
The investigation will focus on Amazon’s contracts with publishers and in particular the use of the “Most Favored Nation” clauses (“MFN”). According to these clauses, the seller must provide to the buyer the lowest price offered to any rival purchaser. Since Amazon maintains the right to be informed about the terms given by other retailers, it is only natural that it will always be able to provide the lowest price in the market. This might constitute anti-competitive behaviour, and abuse of its dominant position in the market.
The Commission has informed Amazon about the open proceedings in the case. There is no legal deadline to complete inquiries for this anti-competitive claim, and the duration of an antitrust investigation varies depending on the laboriousness of a case, the willingness of the firms to cooperate, and the exhaustiveness of defence tools that they might wish to use. For instance, the Google antitrust case lasted four years before the official charges were even levied.
Amazon has declared its willingness to cooperate fully during the process. The European Commission’s e-books investigation is at an early stage and could be dropped or end in a settlement without a formal finding of wrongdoing. If formal charges are eventually brought against Amazon, and Amazon does not successfully refute them, the company could face a fine as much as 10 percent of its more recent annual global sales.
By Marie-Andrée Weiss
On May 18, 2015, the Ninth Circuit held en banc that actress Cindy Lee Garcia does not have a copyright interest in her performance in the Innocence of Muslims movie and that Google can thus not be asked to remove it from all its platforms. The case is Garcia v. Google, No. 12-57302. Many amici curiae filed briefs, which can all be found here.
Garcia had signed up to appear in a low-budget movie, Desert Warrior, and was led to believe the film was about ancient Egyptians. This movie was never completed, but Garcia’s five-second performance was later incorporated by its writer and producer in his anti-Islamic The Innocence of Muslims movie. Garcia’s original lines had been dubbed so that she appeared to ask: “Is your Mohammed a child molester?” The movie was uploaded on YouTube and led to violent protests in the Middle East. It may even have been at the origin of the 2012 attack on the United States Consulate in Benghazi. A fatwa was issued against all people having participated in the movie, and Garcia received death threats.
Garcia filed eight Digital Millennium Copyright Act (DMCA) takedown notices, but Google refused to take the movie down. Garcia then claimed that she had a copyright interest in her performance and sought a preliminary injunction to have Google remove the movie from YouTube. The United States District Court for the Central District of California refused to grant the temporary injunction in November 2012, as Garcia’s copyright claim was not likely to succeed. A three-judge panel of the Ninth Circuit reversed, and issued an injunction to Google to take down all copies of The Innocence of Muslims from YouTube and its other platforms. This decision was later amended to state that the injunction did not encompass movies which did not include Garcia’s performance. The panel’s decision was vacated by the Night Circuit which granted a rehearing en banc. The en banc court affirmed the district court’s decision.
No copyright in a movie performance
Article 7(1)(c) of the 1961 Rome Convention for the Protection of Performers, Producers of Phonograms and Broadcasting Organizations gives performers, including actors, the right to prevent the unauthorized reproduction of a fixation of their performance, if the original fixation had been made without their consent or if the reproduction was made “for purposes different from those for which the performers gave their consent.“ However, the United States did not accede to the Convention and does not recognize otherwise that performers have an individual right in their performance.
Indeed, the Copyright Office refused in March 2014 to register Garcia’s performance because its “longstanding practices do not allow a copyright claim by an individual actor or actress in his or her performance contained within a motion picture. The rationale behind this position is clear: an actor or actress in a motion picture is either a joint author in the entire work, or, as most often is the case, is not an author at all by virtue of a work for hire agreement.”
The three judge panel had found that “[a]n actor’s performance, when fixed, is copyrightable if it evinces “some minimal degree of creativity… `no matter how crude, humble or obvious’ it might be” (at 1263). The panel had further noted that pantomimes and choreographic works may be protected under Section 102 of the Copyright Act and had concluded that “[i]t’s clear that Garcia’s performance meets these minimum requirements.”
But the Ninth Circuit en banc did not agree with this statement, because the Copyright Act only protects original works of authorship fixed in any tangible medium, a fixation which must be done, under 17 U.S.C. §101, by or under the authority of the author. Garcia had not fixed her performance, which was instead fixed by the director and his crew. The en banc court also warned that granting a copyright in a movie performance would “turn cast of thousands into a new mantra: copyright of thousands” (p. 20). It quoted the Community For Creative Non Violence v. Reid case, where the Supreme Court explained that “the author is the party who actually creates the work, that is, the person who translates an idea into a fixed, tangible expression entitled to copyright protection” (p. 22, quoting Reid at 737).
In a rather emotional dissent, Judge Kozinski stated that “Garcia’s dramatic performance met all of the requirements for copyright protection: It was copyrightable subject matter, it was original and it was fixed at the moment it was recorded” (p. 33). He argued that if one considers that “Garcia’s scene is not a work, then every take of every scene of say, Lord of the Rings is not a work, and thus not protected by copyright unless and until the clips become part of the final movie” (p.35). For Judge Kozinski, “a performer need not operate the recording equipment to be an author of his own performance” (p.37).
Could Garcia have been successful by filing different claims?
Garcia’s goal was less to have her performance protected by copyright than to have The Innocence of Muslims taken down from the Web, as its wide dissemination and the ensuing fatwa had caused her “severe emotional distress, the destruction of her career and reputation and credible death threats” (p24). However, the en banc court stated that “[t]his relief is not easily achieved under copyright law… [and] the protection of privacy is not a function of the copyright law… Likewise, authors cannot seek emotional distress damages under the Copyright Act, because such damages are unrelated to the value and marketability of their works” (p.25).
While the en banc court was “sympathetic to her plight … the claim against Google is grounded in copyright law, not privacy, emotional distress, or tort law” (p.8). The court further noted that the “difficulty with Garcia’s claim is that there is a mismatch between her substantive copyright claim and the dangers she hopes to remedy through an injunction. Garcia seeks a preliminary injunction under copyright law, not privacy, fraud, false light or any other tort-based cause of action. Hence, Garcia’s harm must stem from copyright- namely, harm to her legal interest as an author “ (p. 24).
Garcia had originally filed a claim in Los Angeles Superior Court claiming invasion of privacy, false light, violation of her right to publicity, slander and intentional infliction of emotional distress. She voluntarily dismissed her state court suit to file instead a copyright infringement suit in the federal court, but she has since revived her state claims against Youssef. Indeed, she may be able to prove that the director’s conduct has caused her damages.
Copyright and First Amendment
The en banc Court noted that, “[u]nfortunately for Garcia … a “right to be forgotten” although recently affirmed by the Court of Justice for the European Union, is not recognized in the United States” (p. 26). This case, Google Spain SL v. Agencia Española de Protección de Datos, requires search engines to remove links to personal data upon demand of the individual whose personal information is thus displayed. The allusion to the right to be forgotten is interesting as the U.S. is generally viewing it as a threat to freedom of expression.
In this case, Garcia asked an Internet intermediary to take down speech. Justice Kozinski wrote in 2014 as part of the three-judge panel that the words which Garcia seems to utter in the dubbed version of her performance were “fighting words to many faithful Muslims” (Garcia v. Google, at 1262), a choice of words which may have been an attempt to present her dubbed performance as unprotected speech as it is so blasphemous that it triggers violence.
But for the en banc court, this “appeal teaches a simple lesson – a weak copyright claim cannot justify censorship in the guise of authorship” (p. 7) and “Garcia seeks to impose speech restrictions under copyright laws meant to foster rather than repress free expression”(p.8). The en banc court found that the panel’s injunction “censored and suppressed a politically significant film- based upon a dubious and unprecedented theory of copyright” and found it to be “a classic prior restraint of speech” (p. 29).
This case could have had a different outcome in Europe, which recognizes ‘neighboring rights.’ However, these rights cannot trump the rights of the author, and, in this case, Basile is the author of the offensive movie and chose to publish it on YouTube. France recognizes that interpreting artists have a moral right over their performances which may give them the right to prevent publication: the Versailles Court of Appeals held in 2004 that interpreters have moral rights, which are “non-negotiable and stem from each contract, and are outside the contractual scope” and that they “may hinder the unacceptable commercial policy” set by the other contracting party. Also, the Innocence of Muslims movie could be considered unprotected hate speech in Europe, and thus the injunction to take it down could have been successful, which would have made the whole discussion about Garcia’s neighboring rights unnecessary.
By Marie-Andrée Weiss
The movie Mr. Holmes, produced by Miramax, is set to be released in the U.S. on July 17. It is based on the Mitch Cullin novel A Slight Trick of the Mind and features Sherlock Holmes in his twilight years, after he retired to a Sussex farm to tend to his bees. The Conan Doyle Estate (CDE) is suing Miramax, Penguin, the editor of A Slight Trick of the Mind, and Mitch Cullin himself for copyright and trademark infringement. The case is Conan Doyle Estate Ltd. v. Miramax, LLC, No. 1:15-CV-432.
While the first fifty of the Sherlock Holmes short stories and novels are now in the public domain, ten are still protected by copyright in the United States. CDE was unsuccessful last year in its claim that the Sherlock Holmes and Doctor Watson characters are still protected by copyright, as the Seventh Circuit held in Leslie Klinger v. Conan Doyle Estate, Ltd. that they were now in the public domain (see here for a comment about the case in a previous TTLF newsletter).
CDE had argued in Klinger that Sherlock Holmes and Dr. Watson were “complex” characters who evolved during the stories and novels in which they appeared, and that, therefore, they should only become part of the public domain when the copyright of the last story in which they appear expires. This claim did not convince the Seventh Circuit, but Judge Posner noted in Klinger that the characters indeed changed over the years, writing that “[o]nly in the late stories . . . do we learn that Holmes’s attitude toward dogs has changed—he has grown to like them—and that Watson has been married twice. These additional features, being (we may assume) “original” in the generous sense that the word bears in copyright law, are protected by the unexpired copyrights on the late stories.“ Therefore, while the Sherlock Holmes character is no longer protected by copyright, some elements of the character first appearing in stories still protected by copyright may still be protected themselves. CDE claims the Cullin’s novel and the movie are copying these “highly original and protectable” elements.
Development of characters and copyright
According to the complaint, two of the public domain short stories refer to Sherlock Holmes’ retirement on the Sussex Downs, where he takes to bee-farming, and so these particular elements are no longer protected by copyright (Complaint p. 2). The complaint, however, distinguishes the “copyrighted mature Holmes” of the short stories protected by copyright from “the more clinical and purely rational Holmes” of the public domain stories (Complaint p. 12).
CDE claims that the Cullin novel and the Miramax movie are using elements of Sherlock Holmes’ character which were originally created by Sir Arthur Conan Doyle in the short stories still protected by copyright, such as his love of nature and dogs or his ability to express love. For instance, the complaint states that, in The Adventure of the Lion’s Mane, a story still protected by copyright, “Conan Doyle changed Sherlock Holmes. Holmes now loves the quiet of nature” (Complaint p. 10.) For CDE, Conan Doyle continued to develop the Sherlock Holmes character in the ten stories still protected by copyright (Complaint p. 2). For instance, he “develop[ed] a gentler demeanor, emotional warmth, and the ability to show love in his later years” (Complaint p. 8). Conan Doyle “changed Holmes in later life by giving him a gentleness and kindness Holmes did not possess in public domain stories” (Complaint p. 12). His knowledge of medicine was also created in these stories, as was his embrace of modern technologies, such as the microscope, and their use to solve crimes.
CDE further alleges that Cullin’s work is infringing because, in A Glass Armonicist, a story within a story in Cullin’s novel, Holmes is shown to be respectful and solicitous toward Dr. Watson, which the Complaint states is “[a] more subtle but important copying” from two protected stories, The Three Garridebs and The Adventure’s of the Lion’s Mane (Complaint p. 15). The Complaint cites a passage of The Lion’s Mane where the detective recalls how deeply moved he was when he first saw a very beautiful woman who had been the object of affection of a man who had died under mysterious circumstances, and affirms that “Cullin similarly has Holmes react from his heart rather than just his mind to the woman at the center of “The Glass Armonicist” (Complaint p. 15). The Complaint does not, however, cite any parallel allegedly infringing passage.
These two claims are likely to be unsuccessful. As Sherlock Holmes is in the public domain, writers are free to write stories featuring him, and may provide him with new feelings. Respect and solicitousness for an old friend, emotions when meeting a beautiful woman, are not particularly original feelings. Indeed, most people would find them normal and, while Conan Doyle chose to provide these feelings to the famously cold detective in his twilight years, their inner normalcy may prevent finding them original enough to prove infringement of the protected “Sherlock Holmes as a retiree”
The setting of both works
Conan Doyle has Holmes describing his “little Sussex home” in the second paragraph of The Adventure of the Lion’s Mane. The Complaint gives as an example of alleged copying by both the novel and the movie their featuring of “the lonely villa, the chalk cliffs in the distance, the path to the beach, and Holmes’ walks on it.” This particular example is not likely to be found protectable by the court: if a particular character has a farm in Sussex, it may very well be close to the Channel, and if it is, the house could very well be “commanding a great view of the Channel,” as written in The Lion’s Mane. Such a property would also be likely to have a path leading to the sea, and this path could be “long, tortuous . . . steep and slippery” as described in The Lion’s Mane, and as often observed in the area. Such a setting would probably be considered scènes à faire for a novel set in Sussex, in a house close to the sea.
The style of both works
The complaint points out that the Glass Armonicist is supposedly written by Holmes himself, using the first person, not the voice of Doctor Watson, and that Conan Doyle had his hero narrate his adventures in only two stories, The Lion’s Mane and The Adventure of the Blanched Soldier, which are still protected by copyright. The complaint reproduces a passage of Blanched Soldier and an allegedly infringing passage of Cullin’s novel. Both passages have Holmes crediting Watson for having written their adventures in a way interesting to the readers, instead of sticking to the facts, as Holmes would have preferred him to do. Both passages conclude that the writer must present the story in a way which is interesting to the reader (Complaint p. 13).
The Complaint also compares a passage of Blanched Soldier with a passage from Cullin’s novel which CDE claims is infringing. In both extracts, Holmes is shown sitting with his back to the window while receiving a visitor, who is himself illuminated by light. In both passages, the visitor does not know how to start his conversation with Holmes, and the detective finally initiates it by one of his famous deductions. In the original story, Holmes deduces that the visitor is from South Africa, while in Cullin’s work Holmes deduces that the visitor is seeking advice regarding his wife. In both works, the visitor is surprised by Holmes’statement.
Sherlock Holmes as a Trademark
CDE also claims trademark infringement and unfair competition because the movie, Mr. Watson, “uses a title confusingly similar to [the Estate]’s trademark SHERLOCK HOLMES.” CDE owns the SHERLOCK HOLMES trademark for “Organisation of exhibitions for cultural, educational, and entertainment purposes” which was registered last February in the principal register and the SHERLOCK HOLMES trademark for “Electronic gaming machines.” It has applied for registration of the SHERLOCK HOLMES trademark for entertainment services, for motion pictures, and for printed matter. This application has, however, has been suspended by the USPTO for five years pending registration of a similar mark in the same class. Indeed, many SHERLOCK HOLMES trademarks are currently owned by various entities, reflecting the fact that the name of a character in the public domain may indeed be registered as a trademark.
CDE has allowed writers to create new Sherlock Holmes stories, such as Anthony Horowitz’s Moriarty. It has also authorized Warner Brother’s Sherlock Holmes movies, and the BBC series Sherlock which is set in contemporary times. Other writers have used Sherlock Holmes as a retired detective character. For instance, Michael Chabon’s novel, The Final Solution, features an octogenarian detective, smoking a pipe and wearing tweeds, tending to his bees in Sussex. The novel is, however, set during World War II, and the main character is not named, but only referred to as ‘the old man.’ All of these works, whether authorized or not, have contributed to the fame of the Baker Street detective; the “Elementary my Dear Watson” phrase was not created by Conan Doyle, but by Hollywood. We may soon have more clues on whether some aspects of the characters created by Sir Arthur Conan Doyle are still protected by copyright.
By Marie-Andrée Weiss
Richard Prince is an appropriation artist whose name became known by every U.S. copyright attorney, after he won a fair use case which became a cause célèbre. The Second Circuit found in Cariou v. Prince that Prince’s use of Alain Cariou’s Yes Rasta photographs to create his Canal Zone series was transformative enough to be fair use.
Perhaps feeling his oats, Prince used images found on Instagram to create his New Portraits series which was presented at the Gagosian Gallery in October 2014. Prince selected several images posted on Instagram by celebrities, or by “Instagramers” famous for their personal style, and posted comments under the images using his own Instagram account, @richardprince4. Some comments were only a line of emojis, other were nonsensical. The artist then took screenshots of the original postings and his comments and printed them on canvas using an inkjet printer.
The exhibition did not generate much comment when it was privately presented at the Gagosian Gallery in October 2014, but when New Portraits was shown to the public at the Frieze Gallery in New York last May, it generated much comment about whether Prince had the right to reproduce the images, including by the subjects and the authors of the photographs.
Is New Portraits protected by fair use?
Prince may very well be trying to elicit more comments, including legal ones, over his appropriation art: the Gagosian Gallery page about the exhibition did not provide much information about the exhibition, but merely featured the pithy warning “Subject to copyright. Gallery approval must be granted prior to reproduction.” Is this phrase business as usual for an art gallery wishing to protect its rights, and those of its artist, or is it Prince’s taunting of the original Instagram’s users,: come and get me, and we’ll fight in court?
Photographer Donald Graham took the gauntlet. He posted on Instagram in October 2014 a photograph of the New Portraits exhibition, which showed Prince’s work reproducing his 1997 “Rastafarian Smoking a Joint, Jamaica” photograph, hung between other paintings, and wrote: “Appropriated Exhibit. The only way you’d know my work was a part of this display is . . . well, that’s just it, you wouldn’t know. #PrinceofAppropriation.”
The Graham photograph had been posted by another Instagram user, apparently without authorization, not by Graham himself. Prince added this comment under the original post: “Canal Zinian da lam jam” and then reproduced the Instagram post, complete with the @richardprince4’s comment, which could be interpreted as a tease, as Graham’s photograph is a black and white image photograph of a Rastafarian, just as Cariou’s photographs used in the Canal Zone series. Graham sent Prince a cease and desist letter over the use of his work, but no suit has been filed.
It remains to be seen if New Portraits will again lead to a copyright infringement suit allowing Prince to test the limits of fair use. Would this defense be successful? In Cariou v. Prince, the Second Circuit put great weigh on whether Prince’s paintings were transformative enough to be fair use, noting that “Prince altered [the Cariou photographs] significantly, by among other things painting “lozenges” over their subjects’ facial features and using only portion of some of the images” (Cariou at 699). By contrast, Prince did not “alter significantly” the original Instagram posts to create his New Portraits works, as he reproduced them in their entirety adding only a comment or two. Adding a comment on an Instagram post is certainly not transformative enough to qualify as fair use.
However, as we know from Marcel Duchamp and his Ready-mades there may be more to art than what meet the naked eye. Richard Prince is an appropriation artist, and thus savvy at deconstructing and reconstructing images. In New Portraits, Prince reproduced the entire photographs and did not add collages and paint as he did in Canal Zone. However, it can be argued that he took these images out of their original context, social media, and then “re-contextualized” them by making them part of an exhibition presented at prestigious art galleries and making them command six figure prices. He may thus comment on social media, on the contemporary art market, on both, or on noting at all. It should be noted that, while the Southern District of New York (SDNY) had found Prince’s use of Cariou’s work not to be fair as it did not comment on the original works, the Second Circuit rejected this holding, finding that “[t]he law imposes no requirement that a work comment on the original or its author to be considered transformative” (Cariou, at 706).
Does New Portraits violate the New York right of publicity?
Even if the fair use defense could indeed protect Prince, he could have to face right of publicity claims. Most of the works from the New Portraits series were screenshots of photographs representing the original Instagram users. Although not simple “selfies”—they are composed with great care–they do represent the likeness of these Instagram users. Could they claim that Prince has violated their right to publicity? Most U.S. states, including New York, where the exhibition was presented, have right to publicity laws. Sections 50 and 51 of New York Civil Rights Law prohibits the use of personal likeness “for the purpose of trade” without written consent.
It has been reported that some of these works have sold for around $100 000, so they have indeed been traded. However, Prince used the likeness of Instagram users for artistic purposes. Could this be used as a defense by Prince if he is sued for a right of publicity violation? This question is so far purely rhetorical, as none of the Instagram users have yet filed a right of publicity suit against Prince. One of them, DoeDeere, who is a professional makeup artist and cosmetics entrepreneur who regularly posts images of herself wearing stunning make up and wigs, stated on Instagram: “No, I did not give my permission and yes, the controversial artist Richard Prince put it up anyway. It’s already sold ($90K I’ve been told) during the VIP preview. No, I’m not gonna go after him. And nope, I have no idea who ended up with it!”
Indeed, filing a right to publicity suit against Prince would not be a slam dunk, especially in the Second Circuit. The SDNY held in 2002 in Hoepker v. Kruger, that the unauthorized use of the likeness of one of the plaintiffs by collage artist Barbara Kruger to create a work was protected by the First Amendment, regardless of whether the court applies the New York standard, which views art as speech trumping privacy whether the use is transformative or not, or the California standard, which requires the use to be transformative (Hoepker, at 350).
New Portraits is shown in London until August 2015. As the U.K. has no right to publicity, Prince does not have to fear that crossing the Atlantic may lead to a suit. Prince recently wrote on the Gagosian Gallery London exhibition page a piece about the show, explaining how he had recently discovered the pleasure of taking pictures with an iPhone. He “asked [his] daughter more about Tumblr. Are those your photos? Where did you get that one? Did you need permission?” Whether one chooses to believe that Prince is naïve enough to ask his teenage daughter copyright questions, or if this phrase is an artful prick, the question of whether Prince indeed needed permission under U.S. law to use these images is worthy of an answer. Richard Prince concluded ”What’s yours is mine.” Could New Portraits be less a traditional art show than a performance piece appropriating a copyright infringement suit, or a command performance where the parties, judges, and attorneys are being provoked to step into the arena at the command of Prince?
By Marie-Andrée Weiss
What is retail tracking?
This is the first FTC complaint against a retail tracking company. According to the complaint, Nomi “uses mobile device tracking technology to provide analytics services to brick and mortar retailers . . . [and] has been collecting information from consumer’s mobile devices . . . since January 2013.”
While online retailers may easily track their visitors’ digital trail, brick and mortar retailers used to have to resort to asking “are you looking for something in particular?” to find out about their client’s interests, only to be often rebuffed by “just looking…” They also could instruct their staff to report observations about clients’ expressed interests and peruse over sales reports to define and refine their marketing strategy. But tracking companies can now provide retailers precise data on consumer’s behavior.
The complaint explained how sensors placed by Nomi in its clients’ stores detect the media access control (MAC) addresses which mobile devices broadcast when searching for WiFi networks. Nomi also collects MACs from the stores’ WiFi access points. The information thus collected by Nomi is used to compile analytics reports about the percentage of customers passing by the store versus entering it, the average duration of their visit, the type of mobile devices they use, the percentage of repeat consumers within a particular period of time, and the number of customers that have also visited another of the retailer’s location. This information allows retailers to measure the impact of in-store promotions or displays and to adjust their layouts and offerings accordingly.
Nomi provided an opt-out option on its own site. However, consumers had to provide all of their mobile devices’ MAC addresses, a rather cumbersome process, especially since consumers did not know which retailers were using Nomi tracking services and could thus spend time opting out of a service which may never even track them.
According of the terms of the consent order, Nomi agreed not to misrepresent “the extent to which, consumers can exercise control over the collection, use, disclosure, or sharing of information collected from or about them or their computers or devices, or… the extent to which consumers will be provided notice about how data from or about a particular consumer, computer, or device is collected, used, disclosed, or shared.”
Do retail tracking systems identify consumers?
Each MAC is a 12-digit identifier, which the FTC considers to be a persistent unique identifier, even though Nomi cryptographically hashes it, because when a particular MAC is hashed, the resulted hashed MAC is always the same. When one hashes a document or information to encrypt it, an algorithm transforms a string of characters, the input, into another string of characters, the hash value. In our case, each unique 12-digit identifier input are encrypted into a unique hash value, which can be therefore used as identifier.
In his dissenting statement, Commissioner Wright argued that Nomi did not track individual consumers, but merely recorded whether they are unique or repeat visitors to a store, without knowing their “identity.” But Chairwoman Ramirez cited in her statement about the proposed consent order an article written last year by Jonathan Mayer, from Stanford University, which stated that “[h]ashing [MAC addresses] is… no defense against re-identification” and explained how he had built such a re-identification system in less than an hour. Ashkan Soltani, the FTC Chief Technologist, noted in a post that the use of a persistent identifier presents privacy issues since tracking pattern of movement in itself is often enough to uniquely identify an individual.”
Commissioner Wright also argued in his dissenting statement that the FTC should not have issued a complaint against Nomi, as “aggressive prosecution of this sort will inevitably deter industry participants like Nomi from engaging in voluntary practices that promote consumer choices and transparency [ and…] sends a dangerous message to firms weighing the costs and benefits of voluntarily providing information and choice to consumers.” For Commissioner Wright, the market has already responded to consumers expressing their preference, and he alluded in a footnote to several instances where retailers pulled out their tracking programs after consumers voiced their concerns.
But these instances may also be interpreted as signs that consumers are very concerned about being tracked in stores, and thus must be provided with effective ways to opt out, after having been put on notice of such programs. Ashkan Soltani cited in his post a recent OpinionLab survey which found that 8 out of 10 shoppers do not want retailers to track them using their smart phones, adding that “[t]he privacy issues are further exacerbated by the fact that most consumers are not aware that their device information may be captured as they walk by a store or visit an airport.” As such, defining privacy policies may very well drive innovation by incentivizing the creation of products and services respecting consumers’ privacy.
The FTC offered the public the opportunity to file comments about the case, and provided an Analysis to Aid Public Comment. The Information Technology and Innovation Foundation (ITIF), a think tank, while stating it did not condone Nomi’s mistake, argued in its comment that “innovation, by its very nature, involves risks and mistakes . . . .Certainly, companies should not face punitive measures for actions that were taken in good faith and did not cause consumer harm. This would create perverse incentives for companies to slow down the pace of innovation” (ITIF comments, p. 3).
Whether or not the FTC was too quick to act, this case signals the need to provide start-ups and entrepreneurs with the privacy framework they need to create products and services respecting consumers’ privacy. Since most consumers wish to guard their privacy, privacy protection can be an effective marketing tool to attract consumers and generate sales.
By Milos Novovic
One of the key characteristics of the modern Internet is its dynamic nature: over the past decade, the Internet became a domain in which user-generated content rules supreme, allowing people to express themselves, connect with each other and share their data with unprecedented ease. Blogs, wikis, social media websites and other interactive online platforms came to be the new face of the Internet: yet, their very existence heavily relies on the laws shielding the online intermediaries – such as these platforms – from the legal responsibility for user-posted content, provided that certain conditions are met. In the European Union, this principle has been long enshrined in the Directive 2000/31/EC on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market (“E-Commerce Directive”, “ECD”); still, it has been called into question by the recent ruling of the European Court of Human Rights (“ECtHR”) in the Delfi AS v. Estonia (64569/09) case. While the case deals with the liability of online intermediaries in the context of defamation and freedom of expression, it might have wide-reaching consequences for intellectual property laws and the online environment as a whole.
Delfi is one of the largest Internet news portals in Estonia. In 2006, Delfi published an article on possible destruction of ice roads – roads which connect mainland and islands over the frozen sea – by a company called SLK, owned by a person known as L. While the article itself was undisputedly well-balanced and contained no defamatory statements or hate speech, it drew a lot of attention from readers, who posted 185 comments. Some of these comments were manifestly insulting to L, and some could be perceived as direct threats of violence against him.
At the time, Delfi had no team of professional moderators reviewing user-submitted comments before they were published. They did, however, implement an automated software filter which prevented users from posting comments with certain obscene words, as well as a notice-and-takedown system, allowing users to report offensive comments, upon which they were promptly removed. L’s lawyers asked Delfi to remove the offensive comments and pay a fine, and while Delfi removed the offensive material on the same day, it refused to pay any compensation, prompting L to file a lawsuit before Estonian courts. Delfi’s claim that it was immune from liability under the Information Society Services Act, an Estonian law which transposes the ECD, was ultimately rejected, and L was awarded 320 euros in compensation. Delfi brought the matter to the ECtHR, claiming that the ruling of Estonian courts violates its right to freedom of expression under the Article 10 of the European Convention on Human Rights (ECHR). In a unanimous judgment, First Section Chamber of ECtHR concluded that no such violation has occurred, and that Estonian court ruling was a justified and proportionate restriction on Delfi’s right to freedom of expression. Delfi then brought the matter to the Grand Chamber, which affirmed the ruling.
The Grand Chamber Decision
The Court began by restating the importance of user-generated expressive activity on the Internet and its importance for the freedom of expression. It noted at the outset, however, that the ease of access that the Internet offers also poses certain dangers in this context, as defamatory and otherwise “clearly unlawful” – a concept not defined – speech can be easily disseminated and might remain persistently available online. The Court therefore reasoned that there is a tension between Article 8 of the Convention, which protects the right to privacy (“personality rights”), and freedom of expression, guaranteed by the Article 10 – and deeming them both worthy of equal respect, sought to find the balance between the two. In so doing, the Court relied on its previous case-law, requiring an existence of interference with a fundamental freedom (undisputed in this case), that such interference is “prescribed by the law”, has a “legitimate aim”, and is “necessary in a democratic society”. Since the existence of a legitimate aim was not contested by the parties, the main part of the case focused on whether holding Delfi liable for the content of user-posted comments was “prescribed by law” and “necessary in a democratic society”.
Prescribed by law
The Court started by noting that the measure in question had to be not only based on the domestic law, but that such law also has to satisfy certain requirements: namely, be accessible and foreseeable as to its effects. The norm, as the Court stated, had to be formulated with sufficient precision to enable a citizen (although not with absolute certainty) to regulate his conduct – otherwise, it could not have been held to be law. Delfi argued that there was no domestic law according to which an intermediary was to be taken as a professional publisher of comments posted on its website, and that, having relied on the national legislation and ECD framework, it could not have foreseen a different outcome. The Estonian government claimed that there was no case-law on the basis of which Delfi could have presumed that they would not be liable, specifically as owners of an Internet news portal, and claimed that EU law actually supported domestic court ruling.
ECtHR then observed that the fundamental disagreement in the opinions as to which law was to be applied stemmed from the diverging views on the issue of how Delfi was to be classified. According to Delfi, it should have been classified as an intermediary as regards the third-party comments, whereas the Estonian Government argued that it was to be seen as a media publisher, including with regard to such comments. The Court reiterated that it is not its role to interpret national legislation to this effect, and as a matter of lawfulness held that, as a professional publisher, Delfi should have been familiar with the legislation or sought legal advice. In holding that Delfi was in a position to assess the risks related to its activities and that it must have been able to foresee, to a reasonable degree, the consequences, the Court was satisfied that restriction on Delfi’s freedom of expression was lawful.
This holding drew sharp criticism from the dissenting judges. They emphasized that liability of news portals under the ECD regime was far from being settled in 2006, citing the CJEU case of Papasavvas, which only clarified the matter in 2014. They wrote: “Only divine legal counsel could have been sufficiently certain that a portal operator would be liable for a comment it was not aware of, under a kind of strict liability that applied to publishers (editors) who operated in full knowledge of the whole publication. […] Vaguely worded, ambiguous and therefore unforeseeable laws have a chilling effect on freedom of expression. A troubling uncertainty persists here.”
Necessary in a democratic society
In determining the necessity of the restriction imposed, the Court applied the principles embodied in the pre-existing case law to the current case, and went on to analyze the context of the comments, the measures applied by Delfi in order to prevent or remove defamatory comments, the liability of the actual authors of the comments as an alternative to the Delfi’s liability, and the consequences of the domestic proceedings for Delfi’s business model.
The Court first affirmed that Delfi “must be considered to have exercised a substantial degree of control over the comments published on its portal”, relying on several facts – notably, the fact that only Delfi could remove the comments, as original users could not alter or delete them after publication.
In considering whether original posters of comments could be held liable instead of Delfi, the Court held that “the uncertain effectiveness of measures allowing the identity of the authors of the comments to be established, coupled with the lack of instruments put in place by the applicant company for the same purpose with a view to making it possible for a victim of hate speech to effectively bring a claim against the authors of the comments, are factors that support a finding that the Supreme Court based its judgment on relevant and sufficient grounds.” While the Court stressed the importance of anonymity online, it also underlined the perils it brings.
Analyzing whether measures taken by Delfi were sufficient to shield them from liability, the Court re-iterated its stance that Delfi exercised substantial degree of control over the comments, and that the mechanisms in place had failed. The automatic word-filtering software did not remove offensive comments, although they did not contain “sophisticated metaphors or subtle threats”. The Court explicitly stated that asking a news portal to set up a team of dedicated moderators who will remove comments containing hate speech “can by no means be equated to ‘private censorship,’” and attached importance to the fact that “the ability of a potential victim of hate speech to continuously monitor the Internet is more limited than the ability of a large commercial Internet news portal to prevent or rapidly remove such comments.” As to the adequacy of the notice-and-takedown system, the Court stated that it can be used as a good balancing tool, but ultimately ruled that “the rights and interests of others and of society as a whole may entitle Contracting States to impose liability on Internet news portals, without contravening Article 10 of the Convention, if they fail to take measures to remove clearly unlawful comments without delay, even without notice from the alleged victim or from third parties.”
Lastly, the Court analyzed what kind of effect the ruling had on Delfi’s business model. It mainly focused on the fact that Delfi was only fined 320 EUR, and that it did not have to change its business model, as it remained one of the largest news portals in Estonia. It remains unclear why the Court did not acknowledge that hiring a professional team of moderators constitutes a change in the business model.
The Court therefore ruled that no violation of Delfi’s freedom of expression occurred.
Effect of the ruling
It remains unclear what kind of long-term effect this ruling might have.
The Court went to great lengths to emphasize that the ruling “does not concern other fora on the Internet where third-party comments can be disseminated, for example an Internet discussion forum or a bulletin board where users can freely set out their ideas on any topics without the discussion being channeled by any input from the forum’s manager; or a social media platform where the platform provider does not offer any content and where the content provider may be a private person running the website or a blog as a hobby”. Dissenting judges wrote as response: “[f]reedom of expression cannot be a matter of a hobby”.
Many prominent organizations have strongly criticized the ruling, claiming that it will chill the freedom of expression, principally because, as the dissenting judges have noted, “to avoid trouble, for active intermediaries the safe harbour will simply be to disable comments.” Moreover, as experiences from notice-and-takedown systems have shown in the past, moderators might be too cautious, and lacking legal training, might remove a significant amount of lawful content. Finally, users might feel the need to self-censor, as knowing that each comment they post will be read by a moderator can be a strong disincentive against posting.
The legal landscape seems rather confusing at this moment. On its surface, the only effect of the ruling is that imposing liability on active intermediaries is not a violation of the ECHR; but, being a violation of the ECD, the logical conclusion is that the EU Member States will not adopt such measures. However, unless such liability is imposed, “according to the logic of the present judgment, there is no proper protection for the rights of those who feel defamed by comments,” which creates a strong tension between the ECHR and ECD. For the time being, we can only wait – and hope that the system of shielding online intermediaries from liability, which is at the core of the modern Web, will not collapse under pressure.