Topic: Legal

Legal side of Reputation Management

1H 2017 Quick Links Part 7 (Fake News, RTBF, Censorship, Extremist Content) 0

Fake News

* NY Times: From Headline to Photograph, a Fake News Masterpiece

* Data and Society: Media Manipulation and Disinformation Online

* Wired: Inside the Macedonian Fake-News Complex

* Mark Verstraete, Derek E. Bambauer, & Jane R. Bambauer, Identifying and Countering Fake News:

fake news taxonomy

* Combating Fake News: An Agenda for Research and Action, May 2017

* David O. Klein and Joshua R. Wueller, Fake News: A Legal Perspective, Journal of Internet law, April 2017

* Columbia Journalism Review: Study: Breitbart-led right-wing media ecosystem altered broader media agenda

* ABA Journal: Lies and Libel: Fake news lacks straightforward cure

* Time: Inside Russia’s Social Media War on America

* WSJ: Document: Russia Uses Rigged Polls, Fake News to Sway Foreign Elections

* NY Times: With Big Red Stamp, Russia Singles Out What It Calls ‘Fake’ News

* NY Times: The Battle Over Your Political Bubble

* Ars Technica: State Department concocting “fake” intellectual property “Twitter feud”

* Washington Post: Spreading fake news becomes standard practice for governments across the world

Right to be Forgotten

* Techcrunch: Google wins ‘right to be forgotten’ battle in Japan

Michael  Geist: Did a Canadian court just establish a new right to be forgotten online?

* From India: The Supreme Court of India today directed Google, Yahoo and Microsoft to set up an in-house mechanism to remove online search results, which has “potential to go counter”to Section 22 of the Pre-Conception & Pre-Natal Diagnostic Techniques Act, 1994 (Act).

* Techdirt: Feds Say Jewelry Company CEO Scrubbed Google Results With Fake Court Orders And Forged Judge’s Signatures

* In IMDb v. Becerra, the case over the statute requiring the removal of actors’ age information from IMDb’s database, the court blasted California’s discovery requests:

the government’s discovery requests are more than annoying. They’re disturbing….It’s one thing for a legislature to enact a speech restriction without an adequate justification. That sometimes happens. It’s another thing for the government’s lawyers to double down on their client’s constitutional error by imposing irrelevant, burdensome, even harassing discovery obligations on a party that seeks only to vindicate its First Amendment rights in court. That should never happen.

* The Street: A “right to be forgotten”/expungement in FINRA’s BrokerCheck reputational database makes it of dubious credibility


* Milton Mueller: Internet Fragmentation Exists, But Not In the Way That You Think

* NY Times: Clearing Out the App Stores: Government Censorship Made Easier

* Vice: Apple’s Long History of Rejecting ‘Objectionable Content’ From the App Store

* Nathan Cortez, Regulation by Database, University of Colorado Law Review, Vol. 89, 2017. A critical examination of how it’s different when governments publish information about their constituents, especially for purposes of shaming them.

* Reuters: North Korea uses sophisticated tools to spy on citizens digitally – report

* Inside Higher Ed: “The University of California, Berkeley, will cut off public access to tens of thousands of video lectures and podcasts in response to a U.S. Justice Department order that it make the educational content accessible to people with disabilities.”  If you’re looking for paper topic, the ADA as a censorship tool might be an intriguing one.

* Bored Panda: Artist Uses 100,000 Banned Books To Build A Full-Size Parthenon At Historic Nazi Book Burning Site


* TorrentFreak: China Bans Unauthorized VPN Services in Internet Crackdown

* Reuters: China to further tighten its internet controls

* Reuters: ‘Sing the motherland’: China tightens rules for online content. Followup: China’s bloggers, filmmakers feel chill of internet crackdown

* Reuters: China’s cyber watchdog orders top tech platforms to increase self-censorship

* Reuters: Apple is removing VPN services from China App Store: providers

* NY Times: Silicon Valley Giants Confront New Walls in China

* NY Times: Apple Removes Apps From China Store That Help Internet Users Evade Censorship

* NY Times: Joining Apple, Amazon’s China Cloud Service Bows to Censors

* Reuters: China not to license Pokemon Go, similar games as it weighs security risks


* Reporters Without Borders: Russian parliament certifies free Internet’s death

* Reuters: Putin bans VPNs to stop Russians accessing prohibited websites

Extremist Content

* Wired: The One Hire Facebook Really Needs to Make to Curb Violence

* Facebook: Hard Questions: How We Counter Terrorism

* The Switch: Facebook adds 3,000 employees to screen for violence as it nears 2 billion users

* Wired: AI Isn’t Smart Enough (Yet) to Spot Horrific Facebook Videos

* Daphne Keller: Making Google the Censor:

making private companies curtail user expression in important public forums — which is what platforms like Twitter and Facebook have become — is dangerous. The proposed laws would harm free expression and information access for journalists, political dissidents and ordinary users. Policy makers should be candid about these consequences and not pretend that Silicon Valley has silver-bullet technology that can purge the internet of extremist content without taking down important legal speech with it….Governments that outsource speech control to private companies can effectively achieve censorship by proxy….Filters that can find child sexual abuse images work relatively well because those images are illegal in every instance. But violent and extremist material is different. Almost any such image or video is legal in some context. Filters can’t tell the difference between footage used for terrorist recruitment and the same footage used for journalism, political advocacy or human rights efforts.

* Google: Four steps we’re taking today to fight terrorism online

* The Guardian: Revealed: Facebook exposed identities of moderators to suspected terrorists

* Wired: Hacking Online Hate Means Talking to the Humans Behind It

* The Guardian: May and Macron plan joint crackdown on online terror

* Washington Post: YouTube is tricking people who search for ISIS videos

* YouTube: An update on our commitment to fight terror content online

* EFF: Industry Efforts to Censor Pro-Terrorism Online Content Pose Risks to Free Speech

* EFF: Payment Processors Are Profiling Heavy Metal Fans as Terrorists

Source: Eric Goldman Legal

Senate’s “Stop Enabling Sex Traffickers Act of 2017”–and Section 230’s Imminent Evisceration 0

drudge sirenA new anti-Section 230 bill, the Stop Enabling Sex Traffickers Act of 2017, will be introduced in the Senate imminently, perhaps tomorrow. It is being introduced by six senators (Senators Portman, Blumenthal, McCain, McCaskill, Cornyn, and Heitkamp), and I’ve been told there will be many other co-sponsors.

The Senate bill would create new exceptions to Section 230 for sex trafficking-related claims. The bill covers the same territory as Rep. Wagner’s “Allow States and Victims to Fight Online Sex Trafficking Act of 2017” bill. Though the Wagner bill contained additional problematic provisions, the Senate bill will have the same deleterious consequences for Section 230 and free speech online.

Given the significant co-sponsor support for these bills, it’s pretty clear Congress is on the cusp of gutting Section 230. This is the threat we’ve always knew was coming to Section 230, and in a jiffy, we’re already way behind in trying to save Section 230’s principal benefits.

What the Senate Bill Says

The Senate bill would make the following main substantive changes to Section 230:

1) Adding a new exclusion to Section 230’s immunity for “any State criminal prosecution or civil enforcement action targeting conduct that violates a Federal criminal law prohibiting (i) sex trafficking of children; or (ii) sex trafficking by force, threats of force, fraud, or coercion.”

2) Excluding 18 USC 1595 from Section 230’s immunity. 18 USC 1595 is a federal civil cause of action for sex trafficking victims against the bad guys and “whoever knowingly benefits, financially or by receiving anything of value from participation in a venture which that person knew or should have known has engaged in an act in violation of this chapter.”

3) Adding a new definition to 18 USC 1591 (the same provision that the SAVE Act amended–more on the SAVE Act in a moment), which criminalizes receiving money from sex trafficking, to expand the definition of “participation in a venture” to “include knowing conduct by an individual or entity, by any means, that assists, supports, or facilitates the violation.”

How The Senate Bill Differs From the Wagner Bill

From first circulated draft to introduced draft, the Wagner bill expanded to add a number of additional victim-advocate wish-list items. The Senate bill doesn’t include as many of those wish-list items:

* the Wagner bill would exclude state crimes related to “sexual exploitation of children” from Section 230. The Senate bill lacks this exclusion.
* both bills would exclude civil claims for sex trafficking from Section 230, but the Wagner bill would also exclude civil claims related to “sexual exploitation of children.”
* the Wagner bill would create a new 18 USC 1591 crime specifically targeting intermediaries.
* the Wagner bill also proposes to modify the definition of “participation in a venture,” but using more plaintiff-favorable language than the Senate bill.
* the Wagner bill has more onerous changes to Section 230’s policy statements.

So on balance, the Senate bill omits some of the worst policy ideas from the Wagner bill, but both bills are extremely troublesome. If the Wagner bill is a 10 on the worrisome-meter, the Senate bill is a 9.5.

Problems With the Bill

In my blog post on the Wagner bill, I asked numerous questions about the bill. I’ll recapitulate those questions and ask some new ones here:

* what online services will be regulated other than Backpage? The press release accompanying the Senate bill draft references Backpage a half-dozen times. Is this law only about making sure a single company, Backpage, is dead dead dead? Or will the bill reach other online services? If so, who? The most likely answer is that this law potentially implicates every online service that deals with user-generated content, which would make this an unusually wide-ranging bill.

* what about the SAVE Act, the law (sponsored by Rep. Wagner) that Congress passed in 2015 to kill Backpage? The bill’s press release doesn’t mention the SAVE Act once, even though it was designed to accomplish the same policy goals. Why not? Did Congress misjudge the policy efficacy of that law? Or perhaps it’s too early to judge the SAVE Act’s efficacy? A federal grand jury in Phoenix is considering indicting Backpage or its executives, and the odds are that the SAVE Act would be key ground for such an indictment. So perhaps Congress has already enacted all of the legislation it needs to kill Backpage…? If so, a new and major exclusion to Section 230 would not add any new policy benefits but would come with substantial policy costs.

* does the elimination of a centralized online prostitution ad venue actually improve the situation for victims of sex trafficking? This is the fundamental policy objective of the bill, but I have yet to see any good evidence demonstrating this outcome. Maybe it’s so intuitive (shut down Backpage, victims are better) that members of Congress don’t expect to see any proof, but this is hardly intuitive to me. We’ve seen over and over again that anti-prostitution regulations redirect the demand for prostitution elsewhere. If this bill accomplishes its goal, where will that demand get redirected, and how will that affect victims? We’ve also seen many successful victim protection efforts by law enforcement using the public ads as leads/evidence. What will happen to those enforcement efforts, and what does that mean for the overall protection of victims?

* what existing laws will be newly excluded from Section 230, and how will plaintiffs use those laws? I am extremely confident that none of the bill co-sponsors have comprehensively inventoried the existing state laws that will have tenable causes of action against online intermediaries once Section 230’s immunity is lifted. It could be zero laws (unlikely); it could be hundreds or thousands of new laws. Shouldn’t we model these effects before unleashing those laws?

While the bill co-sponsors may not have much clarity about the impending litigation flood, I imagine that victim advocacy groups and especially class-action lawyers have been thinking heavily about those questions.

We’ve already seen a spate of litigation against social media services over the “material support for terrorists” laws, even though Section 230 clearly preempts those. Without Section 230, the potential financial damages of those cases pose existential threats to every online service because they become financial guarantors of all harms caused by terrorists who used their services. In effect, this law would create the same deep pockets dynamic, except for sex trafficking instead of terrorism, and the potential financial exposure may pose an existential threat to many–if not all–online services.

* irrespective of existing laws, my #1 concern about the bill is what new laws will be enacted to take advantage of the newly plowable regulatory ground. Imagine the following laws:

– a new state law says: “if you don’t authenticate all of your users, you are liable for any non-authenticated content promoting sex trafficking of children”
– a new state law says: “if you don’t pre-screen your content, you are liable for all published posts promoting sex trafficking of children”

If these laws would now escape Section 230, then states may be able to effectively force sites to authenticate *all* users and prescreen *all* content to avoid the new laws’ criminal or civil sanctions. In effect, every state law regulator can try “if you don’t do X, you’ll be liable for Y” laws where Y focuses on sex trafficking of children and X involves efforts or policies by online services whose effects aren’t limited to sex trafficking but would change the Internet as we know it. (I recognize the First Amendment might limit those laws even without Section 230 protection, but even easy First Amendment cases are so much more litigation-intensive than straightforward Section 230 cases).

* what steps would online services proactively take to reduce their risk of liability, and how much would those steps cost them? As usual, Section 230’s real benefit comes from leveling the playing field for new industry entrants; so as the costs of entry go up, we see less entry. What does that mean for the emergence of new socially beneficial services to supplement or replace the existing incumbents? We also would expect more “collateral damage” from any proactive steps taken by service providers, i.e., removal or moderation of completely legal content in the effort to mitigate risk of liability.

I explain more about the serious risks of creating state law criminal exceptions to Section 230 in this 2013 primer.

Odds of Passage

On Friday, the Wagner bill added another SIXTY co-sponsors, raising its total to 98 (roughly 2/3 Republican, 1/3 Democrats). The Senate bill has 6 named co-sponsors, and I’ve been told it may have 25 total co-sponsors. Parallel anti-Section 230 bills with 25% of each house’s membership in support are very, very bad news.

Furthermore, this Congress has already proven that it will advance terrible policy ideas (e.g., 49 Senators voting for a healthcare reform bill that not a single one of them wanted to become law). Section 230 is one of Congress’ best tech policy success stories of the past quarter-century, which apparently makes it a primo reform target for this off-kilter Congress.

The goal of protecting sex trafficking victims is, of course, a compelling one, but policy changes of this magnitude come at a serious cost (putting aside the efficacy question). I wonder if all of the bills’ co-sponsors fully understand those costs. If one of your legislators is a co-sponsor of the bills, you might consider letting them know how you feel.

Source: Eric Goldman Legal

1H 2017 Quick Links, Part 6 (Defamation, Section 230, Consumer Reviews) 0


* Gillon v. Bernstein, Civ. No. 2:12-4891 (WJM) (D.N.J. Nov. 3, 2016). No liability for negative Ripoff Report.

* Jackson v. Mayweather, 2017 WL 1131869 (Cal. Ct. App. March 27, 2017). CA’s anti-SLAPP law protects the following Facebook/Instagram post: “the real reason me and Shantel Christine Jackson @MissJackson broke up was because she got an abortion, and I’m totally against killing babies. She killed our twin babies. #ShantelJackson #Floyd Mayweather #TheMoneyTeam #TMT”

* Gottwald v. Geragos, INDEX NO. 162075/201 (N.Y. Sup. Ct. April 18, 2017). Allowing a twibel lawsuit to proceed.

* Snow Systems, Inc. v. Tanner, 2017 WL 716047 (Ill. Ct. App. Feb. 22, 2017):

“an internet posting on a publicly accessible website like the Ripoff Report is, in essence, a type of “mass-media publication,” and as such, is not subject to the discovery rule.”


“the Craigslist Posting contained a single word, “Beware” and a hyperlink that directed the reader to the Cook County Clerk of the Circuit Court’s electronic docket, which contained a list of lawsuits in which Snow Systems has been a party since its incorporation. Plaintiffs do not suggest that the information displayed on the electronic docket is in any way inaccurate. Moreover, even when the docket list is construed in conjunction with the word “Beware,” the Craigslist Posting is not inherently negative or defamatory. The word “Beware” is a very broad, non-specific admonishment, and as such, does not satisfy the heightened specificity and particularity requirement applicable to defamation per se claims. Moreover, although the term “Beware” when read in context with the electronic docket list can be interpreted in a way that disparages Snow Systems as a business, it can also, as Tanner correctly observes, be subject to a number of innocent innocuous interpretations, “including a warning to other business regarding the litigiousness of society, a warning about conducting business in Cook County, and a warning about [the dangers] of operating a snow business in Chicagoland.” Given that the admonishment may be reasonably innocently interpreted, it is not actionable per se.”

* Ars Technica: History by lawsuit: After Gawker’s demise, the “inventor of e-mail” targets Techdirt

Section 230

* Doe v. Oesterblad, 2015 WL 12940181 (D. Ariz. June 9, 2015). This showed up in Westlaw inexplicably late.

The TAC alleges that each of these Plaintiffs have been required to register as sex offenders. The TAC also alleges that Defendant had the “names, photos, and other personal information” of these Plaintiffs pulled from preexisting non-profit and government websites and subsequently displayed this information on at least one of his websites ([also] stating that the contents of websites at issue are “virtually the same” and each was “based upon a database of about 775,000 profiles of individuals previously required to register as sex offenders throughout the United States”). Although Plaintiffs argue in their Response that Defendant “assembled, arranged for presentation, [and] organized [Plaintiffs’ information] into a form designed to send a particular message,” the TAC contains no similar allegations. Even so, the Court cannot conclude that these actions go substantially beyond the traditional editorial role of a publisher. The Court also cannot conclude that an interactive computer service provider only qualifies for protection under Section 230 of the CDA if it maintains a “neutral bulletin board” on which internet users post comments. Although online messaging boards may be the prototypical service qualifying for statutory immunity under the CDA, the Ninth Circuit and other appellate courts have not limited CDA immunity to those websites that merely passively display content created by third parties. Because the Court concludes that the TAC fails to allege that Defendant was responsible for the development or creation of information related to Plaintiffs John Does #1-6, the CDA bars their claims.

The Court will not dismiss the claims of Plaintiffs John Doe #8, Jane Does #9-10, John Doe #11, and David Ellis under Section 230 of the CDA. The TAC alleges that these Plaintiffs have never registered as sex offenders or been convicted of a sex-related offense.The TAC specifically alleges that Defendant obtained the personal information of Plaintiffs required to register as sex offenders from existing databases, but “[p]ersonal information regarding the remaining Plaintiffs was subsequently added by Defendant.” (noting that Defendant “is the only person who adds content or removes content” from the websites listed in the TAC). On at least one or more of his websites, Defendant falsely identified these Plaintiffs as individuals required to register as sex offenders. Based on these allegations, the Court could reasonably conclude that Defendant created a portion of his websites’ content by adding the personal information of those Plaintiffs not listed on preexisting sex offender registries and misidentifying them as individuals who have been convicted of a sex-related offense. Because the TAC sufficiently alleges that Defendant was responsible for his websites’ content concerning Plaintiffs John Doe #8, Jane Does #9-10, John Doe #11, and David Ellis, Defendant is not entitled to CDA immunity for their claims.

* Nail v. Schrauben, 2016 WL 8737183 (W.D. Mich. Jan. 22, 2016): “Plaintiff does not allege that Defendants made the determination that Plaintiff was a sex offender or reported, posted, or otherwise published such in the first instance. Plaintiff likewise does not allege that Defendants modified, edited, or otherwise altered the information in question. Instead, Plaintiff merely alleges that Defendants operated a website which posted, or directed individuals to, information created by a third party. Plaintiff alleges that this conduct constitutes various torts under state law. As discussed above, however, Defendants enjoy immunity from such claims. ”

* Federal court denies habeus corpus petition of Kenneth Gourlay, who provided web hosting (and other) services to an underage child who published child pornography. Gourlay was convicted without  the jury hearing about Section 230. Gourlay v. Barrett, 2017 WL 1329434 (E.D. Mich. April 11, 2017). Prior blog post.

* Ramirez v. Boost Mobile/Sprint, 2017 WL 2535871 (N.D. Cal. June 12, 2017): “Plaintiff has identified the CDA as a potential claim, however, the CDA is used as a tool for immunity by interactive computer services. Furthermore, Plaintiff has failed to identify what information was provided. Therefore, Plaintiff has failed to plead a CDA claim.”

* The Economist: Internet firms’ legal immunity is under threat

* Washington Post: Backpage has always claimed it doesn’t control sex-related ads. New documents show otherwise. A counter-perspective from Elizabeth Nolan Brown at Reason.

* Jeff Kosseff, “Twenty Years of Intermediary Immunity: The US Experience,” 14:1 SCRIPTed 5 (2017). Read the whole thing, but a highlight:

Although the US approach to intermediary immunity is not without its flaws and inequities, it demonstrates that even under a system of robust intermediary immunity, online platforms will develop reasonable safeguards for users.


It is difficult to conceive of how online user reviews – at least in their current form – could continue to exist in the United States without Section 230. User reviews often are blunt, harsh, and, in some cases, subject to factual dispute. The businesses that are the subjects of these reviews may file defamation lawsuits, seeking to be compensated for what they believe are false claims in the user reviews. The people who posted the allegedly defamatory content may have used an anonymity service such as Tor, allowing them to mask their true identities, therefore making it difficult for the subject to name them in a lawsuit. Moreover, even if the posters are identifiable, they may not have sufficient assets to make a defamation lawsuit worthwhile for the plaintiff. Accordingly, the sites hosting the user comments may be an easier and more attractive defendant for a defamation lawsuit.

Section 230 generally has prevented such lawsuits, allowing sites such as Yelp and other consumer review services to act as neutral intermediaries without facing the burden of pre-screening every user comment for accuracy.

* Sekiya v. Facebook, 2017 WL 3084476 (D. N.M. July 17, 2017):

Plaintiff asserts invasion of privacy claims against Defendants Mark Zuckerberg and Facebook. Plaintiff alleges Defendants Zuckerberg and Facebook have not shut down Plaintiff’s Facebook account which Plaintiff has not been able to access for over three years. Plaintiff also alleges that Defendants James Comey, FISA and NSA have not stopped Zuckerberg and Facebook from showing videos on Plaintiff’s Facebook account.
*2 Plaintiff has previously filed two complaints asserting privacy claims against Zuckerbeg and Facebook based on the same facts. See Sekiya v. Facebook, No. 16cv1368 KG/SCY (D.N.M.); Sekiya v. Zuckerberg, No. 17cv283 JCH/KK (D.N.M.). Judge Gonzales dismissed the complaint against Facebook and the unknown owner of Facebook stating:

Defendants are immune to Plaintiff’s cause of action. The Communications Decency Act, 47 U.S.C. § 230, “creates a federal immunity to any cause of action that would make service providers liable for information originating with a third-party user of the service.” Zeran v. America Online, Inc., 129 F.3d 327, 330 (D.C. Cir. 1997); 47 U.S.C. § 230(c)(1) (“No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider”). “Facebook qualifies as an interactive computer service.” Klayman v. Zuckerberg, 753 F.3d 1354, 1357 (D.C. Cir. 2014).

Doc. 5 at 3, in Sekiya v. Facebook, No. 16cv1368 KG/SCY. Judge Herrera dismissed the complaint against Zuckerberg for the same reasons. The Court will dismiss the claims against Defendants Zuckerberg and Facebook in this case without prejudice because they are immune to Plaintiff’s claims pursuant to the Communication Decency Act as discussed by Judge Gonzales and Judge Herrera. The Court notifies Plaintiff that it may impose filing restrictions on Plaintiff if he continues to file privacy claims against Mark Zuckerberg and Facebook based on the facts alleged in this case and his cases before Judge Gonzales and Judge Herrera.

Consumer Reviews

* OnMedica: Increasing number of doctors upset by ‘Trip Advisor’ style online ratings. Related essay.

* Ars Technica: “IoT garage door opener maker bricks customer’s product after bad review.” Does this expose a gap in the Consumer Review Fairness Act? I’m going to add this incident to my primer.

* Consumerist: Jewelry Store Worker Ordered To Pay $34.5K For Posting Fake Yelp Review Of Rival

* Consumerist: Study Finds Paying For Online Reviews Leads To Fewer Reviews. An interesting coda to why Epinions failed where Yelp succeeded?


* AdWeek: Even Bad Online Reviews Can Be Very Good for Business, New Study Finds

Source: Eric Goldman Legal

Courts Keep Shredding Online Contract Formation Processes–McGhee v. NAB; Applebaum v. Lyft 0

A couple more online contract formation cases to enliven your Saturday:

McGhee v. NAB

This case involves mobile credit card processing services. The plaintiffs are merchants who think they were overcharged for card readers. The vendor invoked the arbitration clause in its terms. The vendor alleges that “when McGhee signed up for NAB’s credit card processing services, McGhee was required to accept the ‘Terms and Conditions’ by clicking on a button next to the words ‘I have read and agree to the Terms and Conditions.’” Sounds like an easy case, right?

First, the court calls this implementation a “modified clickwrap” and says a “modified clickwrap agreement is similar to a browsewrap in that the user is not required to scroll through a list of terms and conditions before reaching the “I Agree” button, but the user is otherwise ‘required to affirmatively acknowledge the agreement before proceeding with use of the website.’” Why is it a “modified” click”wrap” and not just a clickwrap? (Better yet, let’s chuck all the BS “wrap” nomenclature and just call it a “clickthrough agreement”). It doesn’t make sense to distinguish between a “modified clickwrap” and a “clickwrap” unless courts thinks the results might differ based on this characterization, which would be really stupid. This reinforces the point we’ve expressed a few times that contract formation calls-to-action should require an extra click, like “By checking this box, I agree to the T&Cs” with a separate checkbox from the account creation button. If courts are going to do stupid things like distinguish between “modified clickwraps” and “clickwraps,” go ahead and take away any grounds for courts to make that stupid distinction.

Second, the vendor’s contract was articulated in at least two documents, the “terms and conditions” and a separate “user agreement.” The account formation page (and associated call to action) linked to the terms, which then further linked to the user agreement that contained the desired arbitration clause. This two-step process isn’t inherently defective (the court says “Even this two-step process could conceivably fall within the parameters of a modified clickwrap agreement”), but it can lead to trouble. Here, the vendor muffs it because the terms contained its own forum selection clause specifying litigation in Georgia courts, while the user agreement contained a conflicting mandatory arbitration clause. Oops. Plus, the terms had a merger clause integrating all other agreements, so the court has no problem giving priority to the directly linked terms instead of the merged-out user agreement two links away.

The vendor tries some fancy footwork to get out of this jam, saying that the terms don’t apply because the merchants never filled out the requisite account application. The court turns this argument around, asking why the terms were referenced in the contract formation call-to-action if the vendor was going to argue they didn’t apply. With a hint of deadpan, the court says “If NAB intended to have the “Pay Anywhere User Agreement” control McGhee’s claims, perhaps NAB should have linked the application to that agreement.”

The vendor did not intrinsically do something wrong by having two documents with terms and linking one into the other. This happens all of the time. However, if an online vendor has multiple separate documents all purporting to be part of the termset governing users, extra care has to be taken to ensure all of the documents work together seamlessly. When those separate documents have separate owners, the risk of a collision between the documents goes up substantially. Those documents need a permanent single owner, or perhaps the separation of documents entails more risk than it’s worth.

While the merchants won this round, the court has a parting gift for the vendor. In a footnote, the court notes: “While McGhee may have won the day, the Court notes that the Terms and Conditions McGhee himself points to as controlling include a forum selection clause and class action waiver.” Be careful what you wish for!

Case citation: McGhee v. North American Bancard, 2017 WL 3118799 (S.D. Cal. July 21, 2017)

Applebaum v. Lyft:

The Uber and Lyft contract formation rulings are coming out faster than we can blog ’em, so we have not attempted to blog them comprehensively or even representatively. Still, this one caught my eye, so I’m blogging it even if it may overlap with multitudinous other rulings we haven’t been able to digest.

The case involves passengers suing for allegedly being overcharged for tolls. To install the Lyft app, at the time the named plaintiff registered, passengers saw this screen:

lyft formation

“The plaintiff could not click the pink “Next” bar (which was necessary to create a registered profile) until he entered his phone number into the “Phone” field and clicked the box (the “Box”) adjacent to the phrase ‘I agree to Lyft’s Terms of Services.’”

During the litigation, Lyft modified its terms of service, including its arbitration clause. Apparently Lyft implemented the new terms by requiring clickthroughs from all members:

The screen contained the entire September 30, 2016 Terms of Service and was scrollable. An existing Lyft customer could not book a ride after September 30, 2016 unless the customer clicked the pink “I accept” bar.

The court says the initial contract formation process failed even though it was a “clickwrap”:

a reasonably prudent consumer would not have been on inquiry notice of the terms of the February 8, 2016 Terms of Service. Lyft’s registration process, as it existed when the plaintiff accessed it in April 2016, did not alert reasonable consumers to the gravity of the “clicks,” namely, that clicking the Box and then the pink “Next” bar at the bottom of the screen constituted acceptance of a contract, including an arbitration agreement, governing the parties’ obligations going forward. Instead, the design and content of the registration process — especially compared to the respective processes that existed for the July 28, 2014 Terms of Services and the September 30, 2016 Terms of Service (the latter of which is discussed below) — discouraged recognition of the existence of lengthier contractual terms that should be reviewed.

Initially, the text is difficult to read: “I agree to Lyft’s Terms of Service” is in the smallest font on the screen, dwarfed by the jumbo-sized pink “Next” bar at the bottom of the screen and the bold header “Add Phone Number” at the top. The “Terms of Service” are colored in light blue superimposed on a bright white background, making those “Terms of Service” — which Lyft argues are the operative words that would alert a reasonable consumer to inquire about a contract — even more difficult to read.

A reasonable consumer would not have understood that the light blue “Terms of Service” hyperlinked to a contract for review. Lyft argues that coloring words signals “hyperlink” to the reasonable consumer, but the tech company assumes too much. Coloring can be for aesthetic purposes. Courts have required more than mere coloring to indicate the existence of a hyperlink to a contract…. Beyond the coloring, there were no familiar indicia to inform consumers that there was in fact a hyperlink that should be clicked and that a contract should be reviewed, such as words to that effect, underlining, bolding, capitalization, italicization, or large font.

Seriously, did the court just say that online users don’t know that light blue text on a screen is probably a hyperlink? I’d love to put that question to a focus group or a survey. We’ve had 20 years of experience with 10 blue links from Google search results! Anyway, the court continues:

Rather than providing notice to consumers that they were agreeing to the terms of a contract, the screen with the Box to be checked for “Lyft’s Terms of Service” was misleading. The screen was titled in bold terms “Add Phone Number.” The entire screen was structured as part of a process to verify a phone number, not to enter a detailed contractual agreement. A reasonable consumer would have thought that the consumer had agreed to be contacted by Lyft, specifically, to receive a text message from the company. A consumer would have arrived at this screen after entering personal information on several similar screens; there was nothing to denote that this screen was special or distinguishable from the rest. The “Add phone number” screen called upon the consumer to input the consumer’s telephone number, with the proviso: “We’ll send a text [message] to verify your phone.” Only then would the consumer click the Box next to “I agree to Lyft’s Terms of Services,” which was immediately below Lyft’s statement that it was going to contact the consumer. The reasonable inference for the reasonable consumer was that the Terms of Service related only to the text verification because the consumer had just agreed to receive a text message. This conclusion would have been reinforced by the design of the screen, including the inconspicuousness of the hyperlink and the absence of cautionary language to indicate that there were contractual terms for review, let alone important contract terms. A reasonable consumer may have understood that the consumer had agreed to something, but not to the lengthy February 8, 2016 Terms of Service.

I’m sure most of you have spit your coffee on the screen by now. “A reasonable consumer may have understood that the consumer had agreed to something, but not to the lengthy February 8, 2016 Terms of Service.” WTF? The court then discounts the significance of the mandatory “next” bar:

a consumer could not click “Next” until the consumer had entered his or her phone information for text verification. As Lyft conceded at oral argument, this was true for every screen in the registration process: each screen contained a “Next” bar (or similar prompt), and a consumer could not proceed to the next screen until completing the required fields on each screen. Thus, this limitation would not have alerted a consumer to the significance of the Box, especially given that the word “Next” implied that there were additional steps in the registration process.

The court also says consumers don’t understand what the words “terms of service” mean in the abstract.

Yet, Lyft saves the day. The September 30, 2016 amendments were presented as a “scrollwrap” (UGH), so…

The method for presenting the September 30, 2016 Terms of Service to existing customers who had already registered conspicuously cured the defects in the notice for the February 8, 2016 Terms of Service. Rather than a screen headed “Add Phone Number,” the September 30, 2016 Terms of Service were presented on a screen titled “Terms of Service,” where, given the content and design of the screen, there could be no doubt as to what that phrase signified. The screen explicitly stated: “Before you can proceed you must read & accept the latest Terms of Service.” The Terms of Service were set out on the screen to be scrolled through. No subtle hyperlink was needed. The Terms of Service begin with the warning: “These Terms of Service constitute a legally binding agreement … between you and Lyft, Inc.” Before proceeding, the user was required to click on a conspicuous bar that said: “I accept.”

So Lyft gets its arbitration despite the failed initial clickwrap, and current and future users will get more prolix contract formation implementations. Yay…?

I’m pretty sure that 5 years ago, 95%+ of us would have said that the Lyft screenshot depicted above was an airtight contract formation. After all, it required an extra click! Yet, the formation bar has moved substantially in the past few years. When the engineers and product teams keep pushing for “lightweight” contract formation processes because everyone else is doing “lightweight” implementations, show them how Lyft barely dodged the bullet in this case.

Case citation: Applebaum v. Lyft, Inc., 2017 WL 2774153 (SDNY June 26, 2017)

Source: Eric Goldman Legal

1H 2017 Quick Links, Part 5 (Advertising, Contracts) 0


* David A. Hyman et al, Going Native: Can Consumers Recognize Native Advertising? Does It Matter?, 19 Yale J.L. & Tech. 77 (2017): “We tested sixteen examples of native advertising. For fifteen of the sixteen examples, fewer than 50% of respondents knew that native ads were paid content. Averaged across all sixteen examples, only 38% of respondents knew that native ads were paid content.”

* NY Times: While retailers and brands may feel like they have no choice but to spend on search advertising, many do so willingly because it works. Kristi Argyilan, a senior vice president of marketing at Target, said search advertising was accounting for a larger portion of her digital spending because the ads were “becoming so useful” and more relevant.

* Reuters: Burger King debuts Whopper ad that triggers Google Home devices

* Adweek: 1 in 3 Internet Users Has Made a Purchase Based on Sponsored Content. Reminder: Most people don’t care if content is sponsored as long as it’s useful and high quality

* Marketing Land: Study: Gen Z more discriminating, more advertising-resistant than Gen X or Y

* Adweek: “In an era of increased media fragmentation, brands are finding that the best way to market their products is to create entertainment that consumers actively seek out,” said Maude Standish, vp of programming strategy of Fullscreen. “This is particularly true when marketing to Gen Z, who grew up with the internet and are not only demanding that all brands entertain them, but also that entertainment shifts to behave like a friend. It’s also why influencers are so effective in selling to this generation.”

* FTC: Economic Analysis of Hotel Resort Fees. The fact that the FTC is doing nothing about resort fees is ridiculous.

* Washington Post: ‘It wasn’t even a question’: The simple calculation for pulling advertising off Breitbart

* Daily Mail Online: Are these the biggest food lies ever? Hilarious photos reveal the misleading packaging that’s led to some VERY disappointed (and hungry) customers

* NY Times: A Bank Had Ads on 400,000 Sites. Then Just 5,000. Same Results.

* WSJ: Online Publishers Try Reducing Ads to Boost Revenue

* FTC Staff Reminds Influencers and Brands to Clearly Disclose RelationshipThe Recorder: Who Got Those Social ‘Influencer’ Letters From the FTC? Read the Full List. Prior blog post.

* Venable: Operation Full Disclosure Part 2: FTC Compliance Sweep—Influencing the Use of Influencers. How companies responded.

* Wired: What Really Happens Inside a PR Crisis War Room

* Jalopnik: Here’s How Much Businesses Pay To Get On Those Big Blue Exit Signs


* Artifex Software v. Hancom, 3:16-cv-06982-JSC (N.D. Cal. April 25, 2017): “Defendant contends that Plaintiff’s reliance on the unsigned GNU GPL fails to plausibly demonstrate mutual assent, that is, the existence of a contract. Not so. The GNU GPL, which is attached to the complaint, provides that the Ghostscript user agrees to its terms if the user does not obtain a commercial license. Plaintiff alleges that Defendant used Ghostscript, did not obtain a commercial license, and represented publicly that its use of Ghostscript was licensed under the GNL GPU. These allegations sufficiently plead the existence of a contract.”

* Oren Bar-Gill et al, “Drawing False Inferences from Mandated Disclosures.”  Abstract:

Disclosure mandates are pervasive. Though designed to inform consumers, such mandates may lead consumers to draw false inferences – for example, that a product is harmful when it is not. When deciding to require disclosure of an ingredient in or characteristic of a product, regulators may be motivated by evidence that the ingredient or characteristic is harmful to consumers. But they may also be motivated by a belief that consumers have a right to know what they are buying or by interest-group pressure. Consumers who misperceive the regulator’s true motive, or mix of motives, will draw false inferences from the mandated disclosure. If consumers think that the disclosure is motivated by evidence of harm, when in fact it is motivated by a belief in a right-to-know or by interest-group pressure, then they will be inefficiently deterred from purchasing the product. We analyze this general concern about disclosure mandates. We also offer survey evidence demonstrating that the risk of false inferences is serious and real. Our framework has implications for the ongoing debate over the labeling of food with genetically modified organisms (GMOs); it suggests that the relevant labels might prove misleading to some or many consumers, producing a potentially serious welfare loss. Under prevailing executive orders, regulators must consider that loss and if feasible, quantify it.

* Jane Bambauer et al, A Bad Education, Univ. of Ill. L. Rev.:

Mandated-disclosure laws achieve their regulatory goals by educating the public about latent attributes of a product or service. At their best, they improve the accuracy of consumers’ cost-benefit analyses compared to a world without disclosure and inspire firms to reduce unnecessary risks. When mandated disclosures, however, do not improve cost-benefit assessments–when they are useless or, worse still, when they reduce the quality of those assessments– then they constitute a bad education.

American privacy law, which is principally a mandated-disclosure regime, imposes a bad education on consumers. This Article proposes a theory for differentiating valuable disclosures from wasteful and harmful ones. Valuable disclosures provide notice about material attributes without inducing an overreaction. After validating the theory in an experimental setting using disclosures about health risks, moral risks, and pseudoscience, we apply the model to four distinct forms of privacy-invasive practices. We find that the disclosures required by regulators are usually wasteful and may cause consumers to overreact. This is the first study to compare disclosures about privacy practices to disclosures about other types of attributes. It raises, for the first time, a troubling insight: if consumer law were guided by the same justifications as our privacy law, it would have to mandate disclosures about GMOs, animal testing, and an unlimited range of other attributes that produce visceral responses.

* Mike Hintze, In Defense of the Long Privacy Statement, Md. L. Rev.

* CouponCabin LLC v., Inc., 2017 WL 83337 (N.D. Ind. Jan. 10, 2017): In a scraping case, denying a defendant’s judgment on the pleadings even though it remains unclear how the defendant got any notice from the plaintiff. Prior blog post.

* Bob Sullivan: CFPB finally issues ban on class-action waivers; but with bureau’s future uncertain, will it stick?

Source: Eric Goldman Legal